Siemens SINUMERIK 840D sl Function Manual
Hide thumbs Also See for SINUMERIK 840D sl:
Table of Contents

Advertisement

Quick Links

SINUMERIK
SINUMERIK 840D sl / 828D
Special functions
Function Manual
Valid for
Controllers
SINUMERIK 840D sl / 840DE sl
SINUMERIK 828D
Software
CNC software
01/2015
6FC5397-2BP40-5BA2
Version
4.7 SP1
Preface
Fundamental safety
instructions
F2: Multi-axis transformations
G1: Gantry axes
K6: Contour tunnel monitoring
K7: Kinematic chain - only
840D sl
K8: Geometric machine
modeling - only 840D sl
K9: Collision avoidance - only
840D sl
M3: Coupled axes
R3: Extended stop and retract
S9: Setpoint exchange -
840D sl only
T3: Tangential control - 840D
sl only
T4: Automatic retuning with
AST - only 840D sl
TE01: Installation and
activation of loadable
compile cycles
TE02: Simulation of Compile
Cycles (only HMI Advanced)
TE1: Clearance control -
840D sl only
Continued on next page
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Advertisement

Table of Contents
loading

Summary of Contents for Siemens SINUMERIK 840D sl

  • Page 1 T4: Automatic retuning with AST - only 840D sl Valid for TE01: Installation and Controllers activation of loadable SINUMERIK 840D sl / 840DE sl compile cycles SINUMERIK 828D TE02: Simulation of Compile Software Version Cycles (only HMI Advanced) CNC software 4.7 SP1...
  • Page 2 Siemens AG Order number: 6FC5397-2BP40-5BA2 Copyright © Siemens AG 1995 - 2015. Division Digital Factory Ⓟ 02/2015 Subject to change All rights reserved Postfach 48 48 90026 NÜRNBERG GERMANY...
  • Page 3 TE3: Speed/torque coupling, master-slave TE4: Handling transformation package - 840D sl only TE6: MCS coupling - 840D sl only SINUMERIK 840D sl / 828D Special functions TE7: Continue machining at the contour (retrace support) - 840D sl only TE8: Cycle-independent...
  • Page 4 Note the following: WARNING Siemens products may only be used for the applications described in the catalog and in the relevant technical documentation. If products and components from other manufacturers are used, these must be recommended or approved by Siemens. Proper transport, storage, installation, assembly, commissioning, operation and maintenance are required to ensure that the products operate safely and without any problems.
  • Page 5: Preface

    Training For information about the range of training courses, refer under: ● www.siemens.com/sitrain SITRAIN - Siemens training for products, systems and solutions in automation technology ● www.siemens.com/sinutrain SinuTrain - training software for SINUMERIK FAQs You can find Frequently Asked Questions in the Service&Support pages under Product Support.
  • Page 6 Preface SINUMERIK You can find information on SINUMERIK under the following link: www.siemens.com/sinumerik Target group This publication is intended for: ● Project engineers ● Technologists (from machine manufacturers) ● System startup engineers (Systems/Machines) ● Programmers Benefits The function manual describes the functions so that the target group knows them and can select them.
  • Page 7 The description of functions include as <signal address> of an NC/PLC interface signal, only the address valid for SINUMERIK 840D sl. The signal address for SINUMERIK 828D should be taken from the data lists "Signals to/from ..." at the end of the particular description of functions.
  • Page 8 Preface Quantity structure Explanations concerning the NC/PLC interface are based on the absolute maximum number of sequential components: ● Mode groups (DB11) ● Channels (DB21, etc.) ● Axes/spindles (DB31, etc.) Data types The control provides the following data types that can be used for programming in part programs: Type Meaning...
  • Page 9: Table Of Contents

    Table of contents Preface.................................5 Fundamental safety instructions.........................29 General safety instructions.....................29 Industrial security........................29 F2: Multi-axis transformations........................31 Brief description........................31 2.1.1 5-axis Transformation......................31 2.1.2 3-axis and 4-axis transformation....................33 2.1.3 Orientation transformation with a swiveling linear axis............34 2.1.4 Universal milling head......................36 2.1.5 Orientation axes........................37 2.1.6 Cartesian manual travel......................38 2.1.7...
  • Page 10 Table of contents Orientation..........................85 2.9.1 Basic orientation........................85 2.9.2 Orientation movements with axis limits..................87 2.9.3 Orientation compression......................88 2.9.4 Smoothing of orientation characteristic..................92 2.9.5 Orientation relative to the path....................94 2.9.6 Programming of orientation polynominals................98 2.9.7 System variable for tool orientation..................101 2.10 Orientation axes........................103 2.10.1 JOG mode..........................105 2.10.2...
  • Page 11 Table of contents 3.2.3 Extended monitoring of the synchronism difference............148 3.2.4 Referencing and synchronization of gantry axes..............148 3.2.5 Control dynamics.........................149 3.2.6 Opening the gantry grouping....................149 Referencing and synchronization of gantry axes..............150 3.3.1 Introduction..........................150 3.3.2 Automatic synchronization....................156 3.3.3 Points to note........................157 Start-up of gantry axes......................160 Parameter assignment: Response to faults.................164 PLC interface signals for gantry axes..................165...
  • Page 12 Table of contents 5.2.1.2 Structure of the system variables..................188 5.2.2 Machine data........................190 5.2.2.1 Maximum number of elements.....................190 5.2.2.2 Root element........................190 5.2.3 System variables........................190 5.2.3.1 Overview..........................190 5.2.3.2 $NK_NAME..........................191 5.2.3.3 $NK_NEXT...........................192 5.2.3.4 $NK_PARALLEL........................193 5.2.3.5 $NK_TYPE...........................194 5.2.3.6 $NK_OFF_DIR, $NK_AXIS, $NK_A_OFF (parameterization for $NK_TYPE = AXIS_LIN)..195 5.2.3.7 $NK_OFF_DIR, $NK_AXIS, $NK_A_OFF (parameterization for $NK_TYPE = AXIS_ROT)..........................198...
  • Page 13 Table of contents 6.2.3.10 $NP_INDEX.........................231 6.2.4 System variables: Protection area elements for machine protection areas......232 6.2.4.1 Overview..........................232 6.2.4.2 $NP_NAME..........................233 6.2.4.3 $NP_NEXT...........................234 6.2.4.4 $NP_NEXTP........................235 6.2.4.5 $NP_COLOR........................237 6.2.4.6 $NP_D_LEVEL........................238 6.2.4.7 $NP_USAGE........................239 6.2.4.8 $NP_TYPE...........................240 6.2.4.9 $NP_FILENAME........................243 6.2.4.10 $NP_PARA...........................245 6.2.4.11 $NP_OFF..........................246 6.2.4.12 $NP_DIR..........................247 6.2.4.13 $NP_ANG..........................249 6.2.5 System variables: Protection area elements for automatic tool protection areas....250...
  • Page 14 Table of contents 7.3.2 Request recalculation of the machine model of the collision avoidance (PROTA)....277 7.3.3 Setting the protection area state (PROTS)................278 7.3.4 Determining the clearance of two protection areas (PROTD)..........278 Example..........................280 7.4.1 Specifications........................280 7.4.2 Part program of the machine model..................284 Data lists..........................292 7.5.1 Machine data........................292...
  • Page 15 Table of contents 8.3.1 Product brief.........................329 8.3.1.1 Function..........................329 8.3.1.2 Preconditions........................329 8.3.2 General functionality......................329 8.3.3 Programming........................333 8.3.4 Behavior in AUTOMATIC, MDA and JOG modes..............337 8.3.5 Effectiveness of PLC interface signals.................339 8.3.6 Special characteristics of the axis master value coupling function........339 8.3.7 Supplementary conditions....................340 Electronic gear (EG)......................340 8.4.1...
  • Page 16 Table of contents 8.5.5.9 Behavior of the following axis at switch-off (CPFMOF)............395 8.5.5.10 Position of the following axis when switching off (CPFPOS+CPOF)........396 8.5.5.11 Condition at RESET (CPMRESET)..................396 8.5.5.12 Condition at parts program start (CPMSTART)..............398 8.5.5.13 Status during part program start in search run via program test (CPMPRT).......399 8.5.5.14 Offset / scaling (CPLINTR, CPLINSC, CPLOUTTR, CPLOUTSC)........400 8.5.5.15...
  • Page 17 Table of contents 8.8.2.1 Axis/spindle-specific setting data..................451 8.8.3 System variables........................451 8.8.4 Signals..........................454 8.8.4.1 Signals to axis/spindle......................454 8.8.4.2 Signals from axis/spindle.....................455 R3: Extended stop and retract........................457 Brief description........................457 Control-managed ESR - 840D sl only..................457 9.2.1 Extended stop and retract (ESR)..................457 9.2.2 Drive-independent reactions....................459 9.2.3 Power failure detection and bridging..................459...
  • Page 18 Table of contents 9.6.1.2 Axis/spindlespecific machine data..................494 9.6.2 System variables........................494 9.6.3 Signals..........................495 9.6.3.1 Signals to channel........................495 9.6.3.2 Signals to axis/spindle......................495 9.6.3.3 Signals from axis/spindle.....................495 S9: Setpoint exchange - 840D sl only......................497 10.1 Brief description........................497 10.2 Startup..........................497 10.3 Interface signals........................500 10.4 Interrupts..........................503 10.5 Position control loop......................503...
  • Page 19 Table of contents 12.3.5 CYCLE755 - Backup/restore data..................523 12.3.6 CYCLE756 - Activate optimization results................524 12.3.7 CYCLE757 - Save optimization data..................525 12.3.8 CYCLE758 - Change a parameter value................527 12.3.9 CYCLE759 - Read parameter value..................527 12.3.10 List of the parameters for the automatic servo optimization..........529 12.3.11 Manufacturer-defined identifier for cycle calls and parameters..........532 12.4...
  • Page 20 Table of contents 15.2.3 Control loop structure......................568 15.2.4 Compensation vector......................569 15.3 Technological features of clearance control.................572 15.4 Sensor collision monitoring....................573 15.5 Startup..........................574 15.5.1 Activating the technological function..................574 15.5.2 Configuring the memory.......................574 15.5.3 Parameter settings for input signals (840D sl)..............574 15.5.4 Parameters of the programmable compensation vector............575 15.5.5 Parameter settings for clearance control................577...
  • Page 21 16.9 Examples..........................630 16.9.1 Master-slave coupling between AX1=Master and AX2=Slave..........630 16.9.2 Close coupling via the PLC....................630 16.9.3 Closing/separating the coupling via the part program for the SINUMERIK 840D sl....631 16.9.4 Release the mechanical brake.....................632 16.10 Data lists..........................633 16.10.1 Machine data........................633 16.10.1.1 Axis/spindlespecific machine data..................633...
  • Page 22 Table of contents 17.14 Data lists..........................686 17.14.1 Machine data........................686 17.14.1.1 General machine data......................686 17.14.1.2 Channelspecific machine data.....................686 17.14.1.3 Channel-specific machine data for compile cycles..............686 17.14.2 Signals..........................688 17.14.2.1 Signals from channel......................688 TE6: MCS coupling - 840D sl only......................689 18.1 Brief description........................689 18.2 Description of MCS coupling functions................690 18.2.1...
  • Page 23 Table of contents 19.5.2 Main program (CC_RESU.MPF)..................713 19.5.3 INI program (CC_RESU_INI.SPF)..................714 19.5.4 END program (CC_RESU_END.SPF).................716 19.5.5 Retrace support ASUB (CC_RESU_BS_ASUP.SPF)............716 19.5.6 RESU ASUB (CC_RESU_ASUP.SPF)................717 19.6 Retrace support........................718 19.6.1 General..........................718 19.6.2 Block search with calculation on contour................718 19.6.3 Reposition..........................719 19.6.4 Temporal conditions concerning NC start................720 19.6.5 Block search from last main block..................721 19.7...
  • Page 24 Table of contents 20.3.3 Parameterizing the digital on-board outputs................736 20.3.4 Parameterizing the switching signal..................737 20.3.5 Parameterization of the geometry axes................737 20.4 Programming........................738 20.4.1 Activating the block-related switching signal output (CC_FASTON)........738 20.4.2 Activating the path length-related switching signal output (CC_FASTON_CONT)....739 20.4.3 Deactivation (CC_FASTOFF)....................740 20.5 Function-specific alarm texts....................740...
  • Page 25 Table of contents 21.6.2.2 Axis/Spindle-specific machine data..................758 21.6.3 User data..........................758 21.6.3.1 Global user data (GUD).......................758 V2: Preprocessing............................759 22.1 Brief description........................759 22.2 Program handling.........................761 22.3 Program call.........................764 22.4 Constraints...........................766 22.5 Examples..........................767 22.5.1 Preprocessing individual files....................767 22.5.2 Preprocessing in the dynamic NC memory................769 22.6 Data lists..........................769 22.6.1...
  • Page 26 Table of contents 24.3.1 General activation........................800 24.3.2 Data groups..........................800 24.4 Examples..........................801 24.4.1 Traversal per part program....................801 24.5 Data lists..........................802 24.5.1 Machine data........................802 24.5.1.1 NC-specific machine data....................802 24.5.1.2 Axis/spindlespecific machine data..................802 Z3: NC/PLC interface signals........................803 25.1 F2: 3 to 5-axis transformation....................803 25.1.1 Signals from channel (DB21, ...)..................803 25.2...
  • Page 27 Table of contents Appendix..............................825 List of abbreviations......................825 Overview..........................834 Glossary..............................835 Index.................................857 Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 28 Table of contents Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 29: Fundamental Safety Instructions

    Siemens recommends strongly that you regularly check for product updates. For the secure operation of Siemens products and solutions, it is necessary to take suitable preventive action (e.g. cell protection concept) and integrate each component into a holistic, state-of-the-art industrial security concept.
  • Page 30 ● Keep the software up to date. You will find relevant information and newsletters at this address (http:// support.automation.siemens.com). ● Incorporate the automation and drive components into a holistic, state-of-the-art industrial security concept for the installation or machine. You will find further information at this address (http://www.siemens.com/...
  • Page 31: F2: Multi-Axis Transformations

    F2: Multi-axis transformations Brief description Note The transformations described below require that individual names are assigned to machine axes, channels and geometry axes when the transformation is active. Compare macchine data: MD10000 $MN_AXCONF_MACHAX_NAME_TAB (machine axis name) MD20080 $MC_AXCONF_CHANAX_NAME_TAB (name of the channel axis in the channel) MD20060 $MC_AXCONF_GEOAX_NAME_TAB (name of the geometry axis in the channel) Besides this no unambiguous assignments are present.
  • Page 32 F2: Multi-axis transformations 2.1 Brief description Tool orientation Tool orientation can be specified in two ways: ● Machine-related orientation The machine-related orientation is dependent on the machine kinematics. ● Workpiece-related orientation The workpiece-related orientation is not dependent on the machine kinematics. It is programmed by means of: –...
  • Page 33: 3-Axis And 4-Axis Transformation

    F2: Multi-axis transformations 2.1 Brief description 2.1.2 3-axis and 4-axis transformation Function The 3- and 4-Axis transformations are distinguished by the following characteristics: Transformation Features 3-axis Transformation 2 linear axes 1 rotary axis 4-Axis transformation 3 linear axes 1 rotary axis Both types of transformation belong to the orientation transformations.
  • Page 34: Orientation Transformation With A Swiveling Linear Axis

    F2: Multi-axis transformations 2.1 Brief description Figure 2-2 Schematic diagram of a 4-axis transformation with moveable workpiece 2.1.3 Orientation transformation with a swiveling linear axis. Function The orientation transformation with swiveling linear axis is similar to the 5-axis transformation of Machine Type 3, though the 3rd linear axis is not always perpendicular to the plane defined by the other two linear axes.
  • Page 35 F2: Multi-axis transformations 2.1 Brief description Figure 2-3 Schematic diagram of a machine with swiveling linear axis Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 36: Universal Milling Head

    F2: Multi-axis transformations 2.1 Brief description 2.1.4 Universal milling head Function A machine tool with a universal milling head has got at least 5 axes: ● 3 linear axes – for linear movement [X, Y, Z] – move the machining point to any random position in the working area ●...
  • Page 37: Orientation Axes

    F2: Multi-axis transformations 2.1 Brief description 2.1.5 Orientation axes Model for describing change in orientation There is no such simple correlation between axis motion and change in orientation in case of robots, hexapodes or nutator kinamatics, as in the case of conventional 5-axes machines. For this reason, the change in orientation is defined by a model that is created independently of the actual machine.
  • Page 38: Cartesian Manual Travel

    F2: Multi-axis transformations 2.1 Brief description 2.1.6 Cartesian manual travel Function The "Cartesian Manual Operation" function can be used to set one of the following coordinate systems as reference system for JOG motion to be selected separately for translation and orientation as: ●...
  • Page 39: Activation Via Parts/Program/Softkey

    F2: Multi-axis transformations 2.2 5-axis transformation If the tool orientation changes, the tool length offsets that apply are rotated so that the pivot point for the orientation movement always refers to the corrected tool tip. 2.1.10 Activation via parts/program/softkey The machine data relevant to the kinematic transformation has thus far been activated mostly through POWER ON.
  • Page 40: Machine Types For 5-Axis Transformation

    F2: Multi-axis transformations 2.2 5-axis transformation Fields of application The "5-axis transformation" machining package is provided for machine tools, which have two additional rotary axes (rotation about the linear axes) in addition to three linear axes X, Y and Z: This package thus allows an axially symmetrical tool (milling cutter, laser beam) to be oriented in any desired relation to the workpiece in every point of the machining space.
  • Page 41: Configuration Of A Machine For 5-Axis Transformation

    F2: Multi-axis transformations 2.2 5-axis transformation 6. The following applies to machine type 3: – 1. Rotary axis (4th machine axis of transformation) turns the tool. – 2. Rotary axis (5th machine axis of transformation) turns the tool. 7. Initial tool position: –...
  • Page 42 F2: Multi-axis transformations 2.2 5-axis transformation Machine type The machine types have been designated above as types 1 to 3 and are stored in the following machine data as a two-digit number: MD24100 $MC_TRAFO_TYPE_1 (definition of channel transformation 1) MD24480 $MC_TRAFO_TYPE_10 (definition of channel transformation 10) The following table contains a list of machine types, which are suitable for 5-axis transformation.
  • Page 43 F2: Multi-axis transformations 2.2 5-axis transformation MD24500 $MC_TRAFO5_PART_OFFSET_1 (workpiece-oriented offset) ● for machine type 1 (two-axis swivel head) Vector from machine reference point to table zero point (zero vector) ● for machine type 2 (two-axis rotary table) Vector from last table swivel joint to zero point of table Figure 2-6 Machine data MD24500 $MC_TRAFO5_PART_OFFSET_1 for machine type 2 ●...
  • Page 44 F2: Multi-axis transformations 2.2 5-axis transformation Position vector in MCS $MC_TRAFO5_PART_OFFSET_n[0 ..2] Vector of programmed position in BCS Tool correction vector $MC_TRAFO5_BASE_TOOL_n[0 .. 2] $MC_TRAFO5_JOINT_OFFSET_n[0 .. 2] Figure 2-7 Schematic diagram of CA kinematics, moved tool Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 45 F2: Multi-axis transformations 2.2 5-axis transformation Figure 2-8 Schematic diagram of CB kinematics, moved workpiece Figure 2-9 Schematic diagram of AC kinematics, moved tool, moved workpiece Assignment of direction of rotation The sign interpretation setting for a rotary axis is stored in the sign machine data for 5-axis transformation.
  • Page 46: Tool Orientation

    F2: Multi-axis transformations 2.2 5-axis transformation MD24520 $MC_TRAFO5_ROT_SIGN_IS_PLUS_1[n] (sign of rotary axis 1/2/3 for 5-axis transformation 1) MD24620 $MC_TRAFO5_ROT_SIGN_IS_PLUS_2[n] (sign of rotary axis 1/2/3 for 5-axis transformation 2) Transformation types Ten transformation types per channel can be configured in the following machine data: MD24100 $MC_TRAFO_TYPE_1 ...MD24480 $MC_TRAFO_TYPE_10 (definition of transformation 1 in channel …...
  • Page 47 F2: Multi-axis transformations 2.2 5-axis transformation Programming The orientation of the tool can be programmed in a block directly by specifying the rotary axes or indirectly by specifying the Euler angle, RPY angle and direction vector. The following options are available: ●...
  • Page 48 F2: Multi-axis transformations 2.2 5-axis transformation The orientation is selected via NC language commands ORIWKS and ORIMKS. Figure 2-11 Change in cutter orientation while machining inclined edges Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 49 F2: Multi-axis transformations 2.2 5-axis transformation Figure 2-12 Change in orientation while machining inclined edges ORIMKS constitutes the basic setting The basic setting can be changed via the following machine data: MD20150 MC_GCODE_RESET_VALUES (RESET position of G groups) MD20150 $MC_GCODE_RESET_VALUES [24] = 1 ⇒ ORIWKS is basic setting MD20150 $MC_GCODE_RESET_VALUES [24] = 2 ⇒...
  • Page 50: Singular Positions And Handling

    F2: Multi-axis transformations 2.2 5-axis transformation Alarm 17630 or 17620 is output for G74 and G75 if a transformation is active and the axes to be traversed are involved in the transformation. This applies irrespective of orientation programming. If the start and end vectors are inverse parallel when ORIWKS is active, then no unique plane is defined for the orientation programming, resulting in the output of alarm 14120.
  • Page 51 F2: Multi-axis transformations 2.2 5-axis transformation Alarm 10910 "Irregular velocity run in a path axis" is then triggered. The programmed velocity is then reduced to a value, which does not exceed the maximum axis velocity. Behavior at pole Unwanted behavior of fast compensating movements can be controlled by making an appropriate selection of the following machine data (see following Figure): ●...
  • Page 52 F2: Multi-axis transformations 2.2 5-axis transformation of the pole, a deviation is made from the specified path because the interpolation runs exactly through the pole point. ● MD24530 $MC_TRAFO5_NON_POLE_LIMIT_1 ● MD24630 $MC_TRAFO5_NON_POLE_LIMIT_2 As a result, the position at the end point of the fourth axis (pole axis) deviates from the programmed value.
  • Page 53: 3-Axis And 4-Axis Transformations

    F2: Multi-axis transformations 2.3 3-axis and 4-axis transformations Does not define the treatment of changes in orientation during large circle interpolation unless the starting orientation is equal to the pole orientation or approximates to it and the end orientation of the block is outside the tolerance circle defined in the following machine data. ●...
  • Page 54: Transformation With Swiveled Linear Axis

    F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis Zero position Tool orientation at zero position is the position of the tool with G17 as the active working plane and position of the rotary axis at 0 degrees. Axis assignments The three translatory axes included in the transformation are assigned to any channel axes via machine data $MC_TRAFO_GEOAX_ASSIGN_TAB_n[0..2] and $MC_TRAFO_AXES_IN_n[0..2].
  • Page 55 F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis ● The first rotary axis (A) is moved by two Cartesian linear axes. It rotates the third linear axis (Z) that moves the tool. ● The tool is aligned parallel to the third linear axis (Z). ●...
  • Page 56 F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis MD24100, ... MD25190 $MC_TRAFO_TYP_n = <type>, with n = 1, 2, 3, ... Kinematics <type> Bits 6 - 0 1. Rotary axis 2. rotary axis swiveled linear axis 10,00 000 10,00 001 10,00 010 10,00 011 10,00 100...
  • Page 57 F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis Determining the machine data values As an aid for determining the values for the above-mentioned machine data, the following two sketches clarify the relationships between the vectors. Note Requirement The machine has been traversed so that the toolholding flange aligns with the table zero (*). Is this is technically not possible, vector to must be corrected by the deviations.
  • Page 58 F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis Note A physically identical point on the 1st rotary axis (e.g. point of intersection between the tool axis and the 1st rotary axis) must be assumed for both views. Figure 2-16 Machine in the zero position Figure 2-17 Front view: Vectors for machine in the zero position...
  • Page 59 F2: Multi-axis transformations 2.4 Transformation with swiveled linear axis Figure 2-18 Top view: Vectors for machine in the zero position Determination of the machine data values Perform the following operation: 1. Determine, as shown in the lower part for vector jo in the "Vectors for machine in zero position"...
  • Page 60: Cardan Milling Head

    F2: Multi-axis transformations 2.5 Cardan milling head Cardan milling head 2.5.1 Fundamentals of cardan milling head Note The following description of the cardan milling head transformation has been formulated on the assumption that the reader has already read and understood the general 5-axis transformation described in Section "5-axis transformation (Page 39)".
  • Page 61 F2: Multi-axis transformations 2.5 Cardan milling head Tool orientation Tool orientation at zero position can be specified as follows: ● parallel to the first rotary axis or ● perpendicular to it, and in the plane of the specified axis sequence Types of kinematics The axis sequence of the rotary axes and the orientation direction of the tool at zero position are set for the different types of kinematics using the following machine data:...
  • Page 62: Parameterization

    F2: Multi-axis transformations 2.5 Cardan milling head Axis A' is positioned in the plane spanned by the rectangular axes of the designated axis sequence. If, for example, the axis sequence is CA', then axis A' is positioned in plane Z-X. The angle φ...
  • Page 63: Traverse Of The Cardan Milling Head In Jog Mode

    F2: Multi-axis transformations 2.5 Cardan milling head Axis se‐ Moving component: Bits 6 - 5 quence: Tool Workpiece Tool/workpiece Bits 0 - 2 Zero position Zero position Zero position BC' / B'C CA' / C'A CB' / C'B 1) Orientation of the tool in the zero position: Bits 3 - 4 x: Transformation type can be set -: Transformation type cannot be set Active machining plane...
  • Page 64: Programming Of The 3- To 5-Axis Transformation

    F2: Multi-axis transformations 2.6 Programming of the 3- to 5-axis transformation Programming of the 3- to 5-axis transformation Switch on The 3- to 5-axis transformations, including the transformations with swiveled linear axis and cardan milling head, are enabled with the TRAORI(<transformation-no.>) command. The enable of the transformation sets the NC/PLC interface signal: DB21, ...
  • Page 65: Generic 5-Axis Transformation And Variants

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Generic 5-axis transformation and variants 2.7.1 Functionality Scope of functions The scope of functions of generic 5-axis transformation covers implemented 5-axis transformations (see Section "5-axis transformation (Page 39)") for perpendicular rotary axes as well as transformations for the cardan milling head (one rotary axis parallel to a linear axis, the second rotary axis at any angle to it, see Section "Cardan milling head (Page 60)").
  • Page 66: Description Of Machine Kinematics

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants 2.7.2 Description of machine kinematics Machine types Like the existing 5-axis transformations, there are three different variants of generic 5-axis transformation: 1. Machine type: Rotatable tool Both rotary axes change the orientation of the workpiece. The orientation of the workpiece is fixed.
  • Page 67: Generic Orientation Transformation Variants

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants MD24572 $MC_TRAFO5_AXIS2_1[0] = 0.0 (direction 2nd rotary axis) MD24572 $MC_TRAFO5_AXIS2_1[1] = 1.0 MD24572 $MC_TRAFO5_AXIS2_1[2] = 0,0 2.7.3 Generic orientation transformation variants Extension Generic orientation transformation for 5-axis transformation has been extended with the following variants for 3-and 4-axis transformation: Variant 1 4-axis transformations...
  • Page 68 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Effects on orientations Generic 3-axis or 4-axis transformation has the following effect on the various orientations: The resulting tool orientation is defined according to the hierarchy specified for generic 5-axis transformation. Priority: ●...
  • Page 69: Parameterization Of Orientable Toolholder Data

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants 2.7.4 Parameterization of orientable toolholder data Application Machine types for which the table or tool can be rotated, can either be operated as true 5-axis machines or as conventional machines with orientable toolholders. In both cases, machine kinematics is determined by the same data, which, due to different parameters, previously had to be entered twice - for toolholder via system variables and for transformations via machine data.
  • Page 70 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Note The transformation only takes place if the orientable toolholder concerned is available and the value of $TC_CARR23 contains a valid entry for type M, P or T kinematics in lower or upper case.
  • Page 71 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Assignment for all types of transformation together identical MD24510 $MC_TRAFO5_ROT_AX_OFFSET_1[1] $TC_CARR25 (+ $TC_TCARR65) MD24520 $MC_TRAFO5_ROT_SIGN_IS_PLUS_1[0] (sign of rotary TRUE* axis 1/2/3 for 5-axis transformation 1) MD24520 $MC_TRAFO5_ROT_SIGN_IS_PLUS_1[1] TRUE* *) Machine data MD24520/MD24620 $MC_TRAFO5_ROT_SIGN_IS_PLUS_1/2 are redundant.
  • Page 72 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Transformation type "P" (in accordance with MD24100 $MC_TRAFO_TYPE_1 = 40) MD24500 $MC_TRAFO5_PART_OFFSET_1[0] $TC_CARR18 (+$TC_TCARR58) MD24500 $MC_TRAFO5_PART_OFFSET_1[1] $TC_CARR19 (+$TC_TCARR59) MD24500 $MC_TRAFO5_PART_OFFSET_1[2] $TC_CARR20 (+$TC_TCARR60) Assignments for transformation type 56 Toolholder data assignments dependent on transformation type 56 Transformation type "M"...
  • Page 73: Extension Of The Generic Transformation To Six Axes - 840D Sl Only

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants 2.7.5 Extension of the generic transformation to six axes - 840D sl only Application With the maximum 3 linear axes and 2 rotary axes, the motion and direction of the tool in space can be completely described with the generic 5-axis transformation.
  • Page 74 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Configuration For configuration of a 6-axis transformation the extensions of the following machine data are required: ● The channel axis index of the 3rd rotary axis must be entered in the following machine data: MD24110 $MC_TRAFO_AXES_IN_1[5] (axis assignment for transformation) ●...
  • Page 75 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants normal vector should be perpendicular to the orientation and is only possible when both programmed vectors are not parallel or anti-parallel. Otherwise, alarm 4342 is output. The direction of the first axis, the X axis, is then uniquely defined. Default setting of the orientation normal vector The default setting of the orientation normal vector in the transformation can also be defined as for the default setting of the orientation in one of three ways:...
  • Page 76: Extension Of The Generic Transformation To Seven Axes - 840D Sl Only

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Note The orientation vector of a tool can also be defined via system the variables $TC_DPV or $TC_DPV3 - $TC_DPV5 in tool data - see Function Manual Basic Machine, Tool Corrections (W1), Section: Sum and setup offsets.
  • Page 77 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants reference to the workpiece is also rotated by the 7th axis. This way it is possible to program the orientation in relation to the workpiece. The transformation uses the 7th axis as the observed input variable. To configure the 7th axis, the channel machine data of the 5-/6-axis transformation is extended by one field containing the 3 components of the direction vector of the 7th axis and an axis offset.
  • Page 78 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants The extensions of the following machine data are required to configure a generic 7-axis transformation: Machine data extension $MC_TRAFO_AXES_IN_1[9] The channel axis index of the 4th rotary axis is recor‐ ded here. $MC_TRAFO_AXES_IN_1[10] and $MC_TRA‐...
  • Page 79 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Programming 1. Programming the Cartesian position The position of the 7th axis must be programmed in the workpiece coordination system in addition to the Cartesian position. The Cartesian position is thus programmed in relation to the constant workpiece.
  • Page 80: Cartesian Manual Travel With Generic Transformation

    F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants Frames The basic coordinate system sits on the 7th axis. It is also rotated when the 7th axis rotates. This way the workpiece coordinate system (WCS) does not remain stationary when the workpiece is rotated over the 7th axis.
  • Page 81 F2: Multi-axis transformations 2.7 Generic 5-axis transformation and variants SD42660 $SC_ORI_JOG_MODE (definition of virtual kinematics for JOG) As opposed to the generic 5-/6-axis transformation, only kinematics can be set in which the rotary axes are perpendicular to one another. The traversing of the geometry and orientation axes is performed via the VDI interface signals of the geometry or orientation axes.
  • Page 82: Restrictions For Kinematics And Interpolation

    F2: Multi-axis transformations 2.8 Restrictions for kinematics and interpolation the first axis rotates around the x direction, the second axis rotates around the y direction, the third axis (if present) rotates around the new z direction. SD42660 $SC_ORI_JOG_MODE = 3: When jogging, RPY angles are traversed with rotation sequence ZYX, i.e.: the first axis rotates around the z direction, the second axis rotates around the y direction,...
  • Page 83: Singularities Of Orientation

    F2: Multi-axis transformations 2.8 Restrictions for kinematics and interpolation 3-and 4-axis kinematics For 3- and 4-axis kinematics, only one degree of freedom is available for orientation. The respective transformation determines the relevant orientation angle. In this case, it only makes sense to traverse the orientation axis using ORIAXES.
  • Page 84 F2: Multi-axis transformations 2.8 Restrictions for kinematics and interpolation Example for machine type 1 Rotatable tool Both rotary axes change the orientation of the workpiece. The orientation of the workpiece is fixed. 2-axis swivel head with rotary axis RA 1 (4th transformation axis) and rotary axis RA 2 (5th transformation axis) Figure 2-22 Generic 5-axis transformation;...
  • Page 85: Orientation

    F2: Multi-axis transformations 2.9 Orientation End point within the circle If the end point is within the circle, the first axis comes to a standstill and the second axis moves until the difference between target and actual orientation is minimal. However, since the first rotary axis does not move, the orientation will generally deviate from the programmed value (see previous figure).
  • Page 86 F2: Multi-axis transformations 2.9 Orientation Definition The basic orientation can be defined in three different ways: ● Definition by calling the transformation ● Definition by the orientation of the active tool ● Definition using a machine data Definition by calling the transformation When the transformation is called, the direction vector of the basic orientation can be specified in the call, e.g.
  • Page 87: Orientation Movements With Axis Limits

    F2: Multi-axis transformations 2.9 Orientation If the tool is de-selected, thereby canceling the definition of tool orientation, the basic orientation programmed in machine data becomes operative. Definition using a machine data If the basic orientation is not defined by either of the two variants described above, it is specified with reference to the following machine data: $MC_TRAFO5_BASE_ORIENT_n (basic tool orientation) This machine data must not be set to a zero vector or else an alarm will be generated during...
  • Page 88: Orientation Compression

    F2: Multi-axis transformations 2.9 Orientation Determining permissible axis limits The control system attempts to define another permissible solution if the axis limits are violated, by approaching the desired axis position along the shortest path. The second solution is then verified, and if this solution also violates the axis limits, the axis positions for both solutions are modified by multiples of 360 until a valid position is found.
  • Page 89 F2: Multi-axis transformations 2.9 Orientation Conditions The orientation movement is compressed in the following cases: ● Active orientation transformation (TRAORI) ● Active large radius circular interpolation (i.e. tool orientation is changed in the plane which is determined by start and end orientation). Large circle interpolation is performed under the following conditions: –...
  • Page 90 F2: Multi-axis transformations 2.9 Orientation Compression mode The manner in which the tolerances are to be considered is set via the unit position in the machine data: MD20482 $MC_COMPRESSOR_MODE (mode of compression) Value Meaning The tolerances specified with MD33100 $MA_COMPRESS_POS_TOL are observed for all the axes (geo and orientation axes).
  • Page 91 F2: Multi-axis transformations 2.9 Orientation The hundreds position of MD20482 is used to select which blocks outside the linear blocks (G1) should be compressed. Value Meaning Circular blocks and G0 blocks are not compressed. This is compatible with earlier SW versions.
  • Page 92: Smoothing Of Orientation Characteristic

    F2: Multi-axis transformations 2.9 Orientation N... X=<...> Y=<...> Z=<...> A3=<...> B3=<...> C3=<...> THETA=<...> F=<...> N... X=<...> Y=<...> Z=<...> A2=<...> B2=<...> C2=<...> THETA=<...> F=<...> Programming tool orientation using rotary axis positions Tool orientation can be also specified using rotary axis positions, e.g. with the following structure: N...
  • Page 93 F2: Multi-axis transformations 2.9 Orientation Function The "Smoothing the orientation characteristic (ORISON)" function can be used to smooth oscillations affecting orientation over several blocks. The aim is to achieve a smooth characteristic for both the orientation and the contour. Prerequisites The "Smoothing the orientation characteristic (ORISON)"...
  • Page 94: Orientation Relative To The Path

    F2: Multi-axis transformations 2.9 Orientation Example Program code Comments TRAORI() ; Activation of orientation transformation. ORISON ; Activation of orientation smoothing. $SC_ORISON_TOL=1.0 ; Orientation tolerance smoothing = 1.0 degrees. X10 A3=1 B3=0 C3=1 X10 A3=–1 B3=0 C3=1 X10 A3=1 B3=0 C3=1 X10 A3=–1 B3=0 C3=1 X10 A3=1 B3=0 C3=1 X10 A3=–1 B3=0 C3=1...
  • Page 95 F2: Multi-axis transformations 2.9 Orientation orientation not only at the block end, but also throughout the entire trajectory. The desired orientation is achieved: ● By settable orientation methods with ORIPATH, specifying how interpolation is to be performed relative to the path. ●...
  • Page 96 F2: Multi-axis transformations 2.9 Orientation Deviation from the desired orientation During the interpolation of the block, the orientation may deviate more or less from the desired relative orientation. The orientation achieved in the previous block is transferred to the programmed end orientation using large circular interpolation. The resulting deviation from the desired relative orientation has two main causes: 1.
  • Page 97 F2: Multi-axis transformations 2.9 Orientation 1: Current tool direction (z coordinate) and orientation change (x coordinate) 2: Active plane (z coordinate is normal vector to the active plane) and orientation change (x coordinate) Smoothing of the orientation jump ORIPATHS Smoothing of the orientation jump is done within the setting data SD42670 $SC_ORIPATH_SMOOTH_DIST (path distance to smoothing orientation) of the specified path.
  • Page 98: Programming Of Orientation Polynominals

    F2: Multi-axis transformations 2.9 Orientation Path relative interpolation of the rotation ORIROTC With 6-axis transformations, in addition to the complete interpolation of the tool orientation relative to the path and the rotation of the tool, there is also the option that only the rotation of the tool relative to the path tangent is interpolated.
  • Page 99 F2: Multi-axis transformations 2.9 Orientation Type 2 polynomials Orientation polynomials of type 2 are polynomials for coordinates PO[XH]: x coordinate of the reference point on the tool PO[YH]: y coordinate of the reference point on the tool PO[ZH]: z coordinate of the reference point on the tool Polynomials for angle of rotation and rotation vectors For 6-axis transformations, the rotation of the tool around itself can be programmed for tool orientation.
  • Page 100 F2: Multi-axis transformations 2.9 Orientation Rotations of rotation vectors with ORIROTC The rotation vector is interpolated relative to the path tangent with an offset that can be programmed using the THETA angle. A polynomial up to the 5th degree can also be programmed with PO[THT]=(c2,c3,c4,c5) for the offset angle.
  • Page 101: System Variable For Tool Orientation

    F2: Multi-axis transformations 2.9 Orientation 2.9.7 System variable for tool orientation The tool orientation can be read in various coordinate systems (BCS, WCS, SZS) via system variables as well as also via OPI variables. Tool orientation in BCS System variable Meaning $AC_TOOLO_ACT[<i>] ;...
  • Page 102 F2: Multi-axis transformations 2.9 Orientation System variable Meaning $AC_TOOLR_DIFF Residual angle in degrees, i.e. this is the angle be‐ tween vectors $AC_TOOLR_END[<i>] and $AC_TOOLR_ACT[<i>] $VC_TOOLR[<i>] ; <i> = 1, 2, 3 i-th component of the vector of the actual value of the orientation rotation $VC_TOOLR_DIFF Angle in degrees between the setpoint and actual...
  • Page 103: Orientation Axes

    F2: Multi-axis transformations 2.10 Orientation axes System variable $AC_TOOL_R_ACT[<i>,<j>] ; <i> =1, 2, 3 i-th component of the actual rotation vector in the coordinate system <j> ; <j> = 0, 1, 2 $AC_TOOL_R_END[<i>,<j>] ; <i> =1, 2, 3 i-th component of the rotation vector at the end of the actual block in the coordinate system <j>...
  • Page 104 F2: Multi-axis transformations 2.10 Orientation axes Direction of the tool vector The direction of the tool vector in the initial machine setting is defined in the following machine data: MD24580 $MC_TRAFO5_TOOL_VECTOR_1 (orientation vector direction) or MD24680 $MC_TRAFO5_TOOL_VECTOR_2 (orientation vector direction) Assignment to channel axes Machine data MD24585 $MC_TRAFO5_ORIAX_ASSIGN_TAB_1[0..2] (ORI/channel assignment Transformation 1) are used to assign up to a total of 3 virtual orientation axes to...
  • Page 105: Jog Mode

    F2: Multi-axis transformations 2.10 Orientation axes 2.10.1 JOG mode It is not possible to traverse orientation axes in JOG mode until the following conditions are fulfilled: ● The orientation axis must be defined as such, that is, a value must be set in the following machine data: MD24585 $MC_TRAFO5_ORIAX_ASSIGN_TAB (ORI/channel axis assignment Transformation 1)
  • Page 106: Programming For Orientation Transformation

    F2: Multi-axis transformations 2.10 Orientation axes 2.10.2 Programming for orientation transformation The values can only be programmed in conjunction with an orientation transformation. Programming of orientation Orientation axes are programmed by means of axis names A2, B2 and C2. Euler and RPY values are distinguished on the basis of G-group 50: ●...
  • Page 107 F2: Multi-axis transformations 2.10 Orientation axes References: Programming Manual Fundamentals Note The four variants of orientation programming are mutually exclusive. If mixed values are programmed, alarm 14130 or alarm 14131 is generated. Exception: For 6-axis kinematics with a 3rd degree of freedom for orientation, C2 may also be programmed for variants 3 and 4.
  • Page 108: Programmable Offset For Orientation Axes

    F2: Multi-axis transformations 2.10 Orientation axes 2.10.3 Programmable offset for orientation axes How the programmable offset works The additional programmable offset for orientation axes acts in addition to the existing offset and is specified when transformation is activated. Once transformation has been activated, it is no longer possible to change this additive offset and no zero offset will be applied to the orientation axes in the event of an orientation transformation.
  • Page 109: Orientation Transformation And Orientable Tool Holders

    F2: Multi-axis transformations 2.10 Orientation axes Orientable toolholder with additive offset On an orientable toolholder, the offset for both rotary axes can be programmed with the system variables $TC_CARR24 and $TC_CARR25. This rotary axis offset can be transferred automatically from the zero offset effective at the time the orientable toolholder was activated. Automatic transfer of offset from zero offset is made possible via the following machine data: MD21186 $MC_TOCARR_ROT_OFFSET_FROM_FR = TRUE (offset of TOCARR rotary axes from NPV)
  • Page 110: Orientation Vectors

    F2: Multi-axis transformations 2.11 Orientation vectors Parameterization The modulo display of orientation axes is activated as follows: MD21132 $MC_ORI_DISP_IS_MODULO[0...2] = TRUE The modulo range is defined with the help of the following machine data: ● MD21134 $MC_ORI_MODULO_RANGE[0...2] (Size of the modulo range for the display of the orientation axes) ●...
  • Page 111 F2: Multi-axis transformations 2.11 Orientation vectors MD10674 If machine data MD10674 $MN_PO_WITHOUT_POLY = FALSE (polynomial programming without G-function POLY programmable) it can be specified, whether the following programming is possible: ● PO[...] or PO(...) is possible only if POLY is active, or ●...
  • Page 112 F2: Multi-axis transformations 2.11 Orientation vectors If ORIAXES is active, the interpolation of the rotary axis can also take place using polynomials like polynomial interpolation of axes with POLY . On the other hand, if ORIVECT is active, "normal" large circle interpolation is carried out through linear interpolation of the angle of the orientation vector in the plane that is defined by the start and end vector.
  • Page 113 F2: Multi-axis transformations 2.11 Orientation vectors Figure 2-23 Rotation of the orientation vector in the plane between start and end vector PHI and PSI angle Programming of polynomials for the two angles PO[PHI] and PO[PSI] is always possible. Whether the programmed polynomials are actually interpolated for PHI and PSI depends on: ●...
  • Page 114: Rotations Of Orientation Vector

    F2: Multi-axis transformations 2.11 Orientation vectors Special situations If no polynomial for angle PSI is programmed, the orientation vector is always interpolated in the plane defined by the start and end vector. The PHI angle in this plane is interpolated according to the programmed polynomial for PHI. As a result the orientation vector moves through a "normal"...
  • Page 115 F2: Multi-axis transformations 2.11 Orientation vectors 3. Programming in RPY angles via A2, B2, C2. 4. Programming the direction vector via A3, B3, C3 (the length of the vector is irrelevant). Switching between Euler and RPY angle programming can be selected via the following machine data or via the G-codes ORIEULER and ORIRPY : MD21100 $MC_ORIENTATION_IS_EULER (angle definition for orientation programming) Programming of orientation direction and rotation...
  • Page 116 F2: Multi-axis transformations 2.11 Orientation vectors In such cases, the end value of the angleand the constant and linear coefficient of the polynomial cannot be programmed directly. The linear coefficient is defined by means of the end angle in degrees. The end angle is derived from programming of the rotation vector.
  • Page 117 F2: Multi-axis transformations 2.11 Orientation vectors Activation of rotation A rotation of the orientation vector is programmed with the identifier THETA. The following options are available for programming: Programming of an angle of rotation at the end of the block. THETA=<value>...
  • Page 118: Extended Interpolation Of Orientation Axes

    F2: Multi-axis transformations 2.11 Orientation vectors A programmed orientation rotation is only interpolated if the machine kinematics allow rotation of the tool orientation (e.g. 6-axis machines). 2.11.3 Extended interpolation of orientation axes Functionality To execute a change in orientation along the peripheral surface of a cone located in space, it is necessary to perform an extended interpolation of the orientation vector.
  • Page 119 F2: Multi-axis transformations 2.11 Orientation vectors ● The opening angle of the cone is programmed degrees with the identifier (nutation angle). The value range of this angle is limited to the interval between 0 degrees and 180 degrees. The values 0 degrees and 180 degrees must not be programmed. If an angle is programmed outside the valid interval, an alarm is generated.
  • Page 120 F2: Multi-axis transformations 2.11 Orientation vectors Settings for intermediate orientation orientation interpolation on a cone with intermediate orienta‐ ORICONIO tion: Interpolation on a conical peripheral surface with inter‐ mediate orientation setting If this G-code is active, it is necessary to specify an intermediate orientation with A7, B7, C7 which is specified as a (normalized) vector.
  • Page 121 F2: Multi-axis transformations 2.11 Orientation vectors PO[XH] = (xe, x2, x3, x4, x5): (xe, ye, ze) the end point of the curve, and PO[YH] = (ye, y2, y3, y4, y5): xi, yi, zi the coefficients of the polynomials PO[ZH] = (ze, z2, z3, z4, z5): of the 5th degree maximum. This type of interpolation can be used to program points (G1) or polynomials (POLY) for the two curves in space.
  • Page 122: Online Tool Length Offset

    F2: Multi-axis transformations 2.12 Online tool length offset Examples Various changes in orientation are programmed in the following program example: Program code Comment N10 G1 X0 Y0 F5000 N20 TRAORI ; Orientation transformation active. N30 ORIVECT ; Interpolate tool orientation as a vector N40 ORIPLANE ;...
  • Page 123 F2: Multi-axis transformations 2.12 Online tool length offset Application The online tool length compensation function can be used for: ● Orientation transformations (TRAORI) ● Orientable tool carriers (TCARR) Note The online tool length offset is an option. This function is only practical in conjunction with an active orientation transformation or an active orientable toolholder.
  • Page 124 F2: Multi-axis transformations 2.12 Online tool length offset The following machine data and setting data are available for configuring online tool length compensation: Machine data / setting data Meaning for online tool length offset MD21190 $MC_TOFF_MODE The contents of $AA_TOFF[ ] are traversed as an absolute value or integrated MD21194 $MC_TOFF_VELO (speed online tool Speed of online tool length offset...
  • Page 125 F2: Multi-axis transformations 2.12 Online tool length offset $AA_IW[ ] and $AA_IB[ ] are changed. These variables now contain the deselected share of tool length compensation. Once "Online tool length offset" has been deselected for a tool direction, the value of system variable $AA_TOFF[ ] or $AA_TOFF_VAL[ ] is zero for this tool direction.
  • Page 126: Examples

    F2: Multi-axis transformations 2.13 Examples Reference: Parameter Manual, System Variables Boundary conditions The online tool length offset function is an option and is available during "generic 5-axis transformation" by default and for "orientable toolholders". If the tool is not perpendicular to the workpiece surface during machining or the contour contains curvatures whose radius is smaller than the compensation dimension, deviations compared to the actual offset surface are produced.
  • Page 127 F2: Multi-axis transformations 2.13 Examples $MC_TRAFO_GEOAX_ASSIGN_TAB_1[0]=1 $MC_TRAFO_GEOAX_ASSIGN_TAB_1[1]=2 $MC_TRAFO_GEOAX_ASSIGN_TAB_1[2]=3 $MC_TRAFO5_PART_OFFSET_1[0] = 0 $MC_TRAFO5_PART_OFFSET_1[1] = 0 $MC_TRAFO5_PART_OFFSET_1[2] = 0 $MC_TRAFO5_ROT_AX_OFFSET_1[0] = 0 $MC_TRAFO5_ROT_AX_OFFSET_1[1] = 0 $MC_TRAFO5_ROT_SIGN_IS_PLUS_1[0] = TRUE $MC_TRAFO5_ROT_SIGN_IS_PLUS_1[1] = TRUE $MC_TRAFO5_NON_POLE_LIMIT_1 = 2.0 $MC_TRAFO5_POLE_LIMIT_1 = 2.0 $MC_TRAFO5_BASE_TOOL_1[0] = 0.0 $MC_TRAFO5_BASE_TOOL_1[1] = 0.0 $MC_TRAFO5_BASE_TOOL_1[2] = 5,0 $MC_TRAFO5_JOINT_OFFSET_1[0] = 0.0 $MC_TRAFO5_JOINT_OFFSET_1[1] = 0.0...
  • Page 128 F2: Multi-axis transformations 2.13 Examples Approach initial position: N100 G1 x1 y0 z0 a0 b0 F20000 G90 G64 T1 D1 G17 ADIS=.5 ADISPOS=3 Orientation vector programming: N110 TRAORI(1) N120 ORIWKS N130 G1 G90 N140 a3 = 0 b3 = 0 c3 = 1 x0 N150 a3 = 0 b3 =-1 c3 = 0 N160 a3 = 1 b3 = 0 c3 = 0 N170 a3 = 1 b3 = 0 c3 = 1...
  • Page 129: Example Of A 3-Axis And 4-Axis Transformation

    F2: Multi-axis transformations 2.13 Examples 2.13.2 Example of a 3-axis and 4-axis transformation 2.13.2.1 Example of a 3-axis transformation Example: For the schematically represented machine (see "Figure 2-1 Schematic diagram of 3-axis transformation (Page 33)"), the 3-axis transformation can be projected as follows: Program code Comment $MC_TRAFO_TYPE_n = 18...
  • Page 130 F2: Multi-axis transformations 2.13 Examples Machine data ; machine kinematics CA' with tool orientation in zero position in the z direction $MC_TRAFO_TYPE_1 = 148 $MC_TRAFO_GEOAX_ASSIGN_TAB_1[0]=1 $MC_TRAFO_GEOAX_ASSIGN_TAB_1[1]=2 $MC_TRAFO_GEOAX_ASSIGN_TAB_1[2]=3 ; angle of second rotary axis $MC_TRAFO5_NUTATOR_AX_ANGLE_1 = 45 Program Program code Comment ;...
  • Page 131: Example For Orientation Axes

    F2: Multi-axis transformations 2.13 Examples 2.13.4 Example for orientation axes Example 1: 3 orientation axes for the 1st orientation transformation for kinematics with 6 transformed axes. Rotation must be done in the following sequence: ● firstly about the Z axis. ●...
  • Page 132 F2: Multi-axis transformations 2.13 Examples Example 2: 3 orientation axes for the 2nd orientation transformation for kinematics with 5 transformed axes. Rotation must be done in the following sequence: ● firstly about the X axis. ● then about the Y axis and ●...
  • Page 133: Examples For Orientation Vectors

    F2: Multi-axis transformations 2.13 Examples References: Programming Manual, Production Planning 2.13.5 Examples for orientation vectors 2.13.5.1 Example for polynomial interpretation of orientation vectors Orientation vector in Z-X plane The orientation vector is programmed directly in the examples below. The resulting movements of the rotary axes depend on the particular kinematics of the machine.
  • Page 134: Example Of Rotations Of Orientation Vector

    F2: Multi-axis transformations 2.13 Examples 2.13.5.2 Example of rotations of orientation vector Rotations with angle of rotation THETA In the following example, the angle of rotation is interpolated in linear fashion from starting value 0 degrees to end value 90 degrees. The angle of rotation changes according to a parabola or a rotation can be executed without a change in orientation.
  • Page 135 F2: Multi-axis transformations 2.13 Examples ; Rotatable tool $MC_TRAFO5_AXIS1_1[0] = 0.0 $MC_TRAFO5_AXIS1_1[1] = 0.0 $MC_TRAFO5_AXIS1_1[2] = 1,0 ; 1. Rotary axis is parallel to Z. $MC_TRAFO5_AXIS2_1[0] = 0.0 $MC_TRAFO5_AXIS2_1[1] = 1,0 $MC_TRAFO5_AXIS2_1[2] = 0.0 ; 2. Rotary axis is parallel to Y. $MC_TRAFO5_BASE_ORIENT_1[0] = 1.0 $MC_TRAFO5_BASE_ORIENT_1[1] = 0,0 $MC_TRAFO5_BASE_ORIENT_1[2] = 1.0...
  • Page 136: Example Of A Generic 6-Axis Transformation

    F2: Multi-axis transformations 2.13 Examples Program code Comment ; basic orientation → B0 C0 N220 TOFRAME ; Z axis points in the direction ; of the orientation N230 G91 Z7 ; 7 mm in new Z direction ; Traverse → X2 Y3 Z6 N240 C3=1 ;...
  • Page 137: Example Of A Generic 7-Axis Transformation

    F2: Multi-axis transformations 2.13 Examples Program code Comment N220 C3=1 ; Orientation in the Z direction → tool ; rotated by 26.565 degrees N230 THETA=IC(90) ; Orientation normal vector ; Incremental by 90 degrees ; rotate Vector points in negative ;...
  • Page 138: Example For The Modification Of Rotary Axis Motion

    F2: Multi-axis transformations 2.13 Examples 2.13.6.3 Example for the modification of rotary axis motion The machine is a 5-axis machine of machine type 1 (two-axis swivel head with CA kinematics) on which both rotary axes rotate the tool (transformation type 24). The first rotary axis is a modulo axis parallel to Z (C axis);...
  • Page 139: Data Lists

    F2: Multi-axis transformations 2.14 Data lists Programming Comment ; The movement describes a circle generated from polygons. The orientation moves on a taper around the Z axis with an opening an- gle of 45 degrees. N100 X0 Y0 A3=0 B3=-1 C3=1 N110 FOR COUNTER=0 TO NUMBER N120 ANGLE=360*COUNTER/NUMBER N130 X=RADIUS*cos(angle) Y=RADIUS*sin(angle)
  • Page 140 F2: Multi-axis transformations 2.14 Data lists Number Identifier: $MC_ Description 21094 ORIPATH_MODE Setting for path relative orientation 21100 ORIENTATION_IS_EULER Angle definition for orientation programming 21102 ORI_DEF_WITH_G_CODE Definition of orientation angles A2, B2, C2 21104 ORI_IPO_WITH_G_CODE Definition of interpolation type for orientation 21106 CART_JOG_SYSTEM Coordinate system for Cartesian JOG...
  • Page 141 F2: Multi-axis transformations 2.14 Data lists Number Identifier: $MC_ Description 24432 TRAFO_AXES_IN_5[n] Axis assignment for transformation 5 [axis index] 24434 TRAFO_GEOAX_ASSIGN_TAB_5[n] Assignment geometry axis to channel axis for transfor‐ mation 5 [geometry no.] 24440 TRAFO_TYPE_6 Definition of transformation 6 in channel 24442 TRAFO_AXES_IN_6[n] Axis assignment for transformation 6 [axis index]...
  • Page 142 F2: Multi-axis transformations 2.14 Data lists Number Identifier: $MC_ Description 24572 TRAFO5_AXIS2_1[n] Vector for the second rotary axis and the first transfor‐ mation [n = 0.. 2] 24673 TRAFO5_AXIS3_1[n] Direction of third rotary axis for general 6-axis transfor‐ mation (Transformer type 24, 40, 56, 57) 24574 TRAFO5_BASE_ORIENT_1[n] Basic orientation for the first transformation [n = 0..
  • Page 143: Setting Data

    F2: Multi-axis transformations 2.14 Data lists Number Identifier: $MC_ Description 24682 TRAFO5_TCARR_NO_2 TCARR number for the second 5-axis transformation 2 24685 TRAFO5_ORIAX_ASSIGN_TAB_2[n] Assignment of orientation axes to channel axes for ori‐ entation transformation 2 [n = 0.. 2] 24694 TRAFO7_EXT_ROT_AX_OFFSET_2 Angle offset of the 2nd external rotary axis 24695 TRAFO7_EXT_AXIS1_2...
  • Page 144: Signals

    F2: Multi-axis transformations 2.14 Data lists 2.14.3 Signals 2.14.3.1 Signals from channel Signal name SINUMERIK 840D sl SINUMERIK 828D Activate PTP traversal DB21, ... DBX29.4 Transformation active DB21, ... DBX33.6 Number of active G function of DB21, ... DBB232 G function group 25 PTP traversal active DB21, ...
  • Page 145: G1: Gantry Axes

    G1: Gantry axes Brief description For gantry machines, each of various machine elements, such as the gantry and the transverse beams, are moved by several axes that operate in parallel. The axes that together move a machine part, are designated as gantry axes or gantry grouping. Because of the mechanical structure, the gantry axes are rigidly connected with each other and so must always be traversed synchronously by the control.
  • Page 146: Gantry Axes" Function

    G1: Gantry axes 3.2 "Gantry axes" function "Gantry axes" function 3.2.1 Definition of a gantry grouping Definition The axes of a gantry grouping are specified via the following axial machine data: MD37100 $MA_GANTRY_AXIS_TYPE[AX1] = xy Tens decimal place: Type of gantry axis (guide or synchronous axis) Ones decimal place: ID of the gantry grouping A maximum of eight gantry groupings (gantry grouping ID: 1 - 8) can be defined.
  • Page 147: Monitoring The Synchronism Difference

    G1: Gantry axes 3.2 "Gantry axes" function 3.2.2 Monitoring the synchronism difference Limit values for monitoring 2 limit values can be specified for the synchronism difference. Gantry warning limit The gantry warning limit is set using the following machine data: MD37110 $MA_GANTRY_POS_TOL_WARNING (gantry warning limit) The "Alarm limit exceeded"...
  • Page 148: Extended Monitoring Of The Synchronism Difference

    G1: Gantry axes 3.2 "Gantry axes" function 3.2.3 Extended monitoring of the synchronism difference Activation of the extended monitoring An extended monitoring of the synchronism difference can be activated using the following machine data: MD37150 $MA_GANTRY_FUNCTION_MASK, Bit 0 = 1 For the extended monitoring, a synchronism difference between the leading and synchronous axis, obtained when tracking or when the gantry grouping is opened, is taken into account.
  • Page 149: Control Dynamics

    G1: Gantry axes 3.2 "Gantry axes" function 3.2.5 Control dynamics Use case From the user perspective, a gantry grouping is exclusively traversed via the leading axis. The NC generates the setpoints of the synchronous axes directly from the setpoints of the leading axis in time synchronism and outputs these to them.
  • Page 150: Referencing And Synchronization Of Gantry Axes

    G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Referencing and synchronization of gantry axes 3.3.1 Introduction Misalignment after starting Immediately after the machine is switched on, the leading and synchronous axes may not be ideally positioned in relation to one another (e.g. misalignment of a gantry). Generally speaking, this misalignment is relatively small so that the gantry axes can still be referenced.
  • Page 151 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes The appropriate synchronous axes traverse in synchronism with the leading axis. Interface signal "Referenced/synchronized" of the leading axis is output to indicate that the reference point has been reached. Section 2: Referencing of the synchronous axes As soon as the leading axis has approached its reference point, the synchronous axis is automatically referenced (as for reference point approach).
  • Page 152 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes The next step in the operating sequence depends on the difference calculated between the actual values of the leading and synchronous axes: ● difference is smaller than the gantry warning limit: MD37110 $MA_GANTRY_POS_TOL_WARNING (gantry warning limit) The gantry synchronization process is started automatically.
  • Page 153 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 154 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Figure 3-2 Flowchart for referencing and synchronization of gantry axes Synchronization process A synchronization process is always required in the following cases: ● after the reference point approach of all axes included in a grouping, ●...
  • Page 155 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Instead, the actual position of the leading axis is specified as the target position and is approached in the uncoupled state. Note For the leading axis, automatic synchronization can be locked using the following NC/PLC interface signal: DB31, ...
  • Page 156: Automatic Synchronization

    G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Selecting the reference point To ensure that the shortest possible paths are traversed when the gantry axes are referenced, the reference point values from leading and synchronous axes should be the same in the machine data: MD34100 $MA_REFP_SET_POS (reference point value/destination point for distancecoded system)
  • Page 157: Points To Note

    G1: Gantry axes 3.3 Referencing and synchronization of gantry axes difference between the positions of the leading and synchronized axes greater than the setting in the machine data: MD36030 $MA_STANDSTILL_POS_TOL (standstill tolerance) In this case, a new setpoint is specified for the synchronized axis (axes) without interpolation. The positional difference detected earlier is then corrected by the position controller.
  • Page 158 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes MD36500 $MA_ENC_CHANGE_TOL (Max. tolerance for position actual value switchover) The two position measuring systems must, however, have been referenced beforehand. The relevant measuring system must be selected before referencing is initiated. The operational sequence is then the same as that described above.
  • Page 159 G1: Gantry axes 3.3 Referencing and synchronization of gantry axes Activation of axis compensations Compensation functions can be activated for both the leading axis and the synchronized axes. Compensation values are applied separately for each individual gantry axis. These values must therefore be defined and entered for the leading axis and the synchronized axes during start-up.
  • Page 160: Start-Up Of Gantry Axes

    G1: Gantry axes 3.4 Start-up of gantry axes Start-up of gantry axes General Owing to the forced coupling which is normally present between guide and synchronous gantry axes, the gantry grouping must be started up as if it were an axis unit. For this reason, the axial machine data for the guide and synchronous axes must always be defined and entered jointly.
  • Page 161 G1: Gantry axes 3.4 Start-up of gantry axes Axial optimization The following control parameters must be set to the optimum axial value for both the guide axis and the synchronous axis: ● MD32200 $MA_POSCTRL_GAIN (servo gain factor) ● MD32620 $MA_FFW_MODE (feedforward control parameter) ●...
  • Page 162 G1: Gantry axes 3.4 Start-up of gantry axes Guide axis MD32810 $MA_EQUIV_SPEEDCTRL_TIME [n] = 5 ms (equivalent time constant speed control loop for feedforward control) Synchronous axis MD32810 $MA_EQUIV_SPEEDCTRL_TIME [n] = 3 ms ● Time constant of dynamic response adaptation for synchronous axis: MD32910 $MA_DYN_MATCH_TIME [n] = 5 ms - 3 ms = 2 ms (time constant of dynamic response adaptation) The dynamic response adaptation must be activated axially with the machine data:...
  • Page 163 G1: Gantry axes 3.4 Start-up of gantry axes Synchronizing gantry axes The gantry synchronization is activated via the NC/PLC interface signal (see Section "Referencing and synchronization of gantry axes (Page 148)"): DB31, ... DBX29.4 = 1 (start synchronization of gantry) The completion of the synchronization is displayed via the NC/PLC interface signal: DB31, ...
  • Page 164: Parameter Assignment: Response To Faults

    G1: Gantry axes 3.5 Parameter assignment: Response to faults If the gantry grouping is canceled, the following points must be noted: ● Always activate the traversing range limits and set them to the lowest possible values (position tolerance). ● Synchronize the gantry grouping first if possible and then execute a POWER-ON-RESET without referencing the axes again.
  • Page 165: Plc Interface Signals For Gantry Axes

    G1: Gantry axes 3.6 PLC interface signals for gantry axes MD30455 $MA_MISC_FUNCTION_MASK, bit 9 = <value> <value> Meaning When a fault occurs that triggers the pulse suppression (e.g. measuring-circuit fault), the pulses in all other axes of the gantry grouping will also be suppressed. Result: Coast down of all axes of the gantry grouping.
  • Page 166 G1: Gantry axes 3.6 PLC interface signals for gantry axes For example, all axes in the gantry groupings will be simultaneously shut down when the following interface signal is set to "0" from the leading axis: DB31, ... DBX2.1 (servo enable) The following table shows the effect of individual interface signals (from PLC to axis) on gantry axes: NC/PLC interface signal...
  • Page 167: Miscellaneous Points Regarding Gantry Axes

    G1: Gantry axes 3.7 Miscellaneous points regarding gantry axes Miscellaneous points regarding gantry axes Manual travel It is not possible to traverse a synchronized axis directly by hand in JOG mode. Traverse commands entered via the traversing keys of the synchronized axis are ignored internally in the control.
  • Page 168 G1: Gantry axes 3.7 Miscellaneous points regarding gantry axes Axis replacement All axes in the gantry grouping are released automatically in response to a RELEASE command (leading axis). A replacement of the leading axis of a closed gantry grouping is only possible, if all axes of the grouping are known in the channel in which they are to be transferred, otherwise alarm 10658 is signaled.
  • Page 169: Examples

    G1: Gantry axes 3.8 Examples Differences in comparison with the "Coupled motion" function The main differences between the "gantry axes" and "coupled motion" functions are listed below: ● The axis coupling between the gantry axes must always be active. Separation of the axis coupling via part program is therefore not possible for gantry axes.
  • Page 170 G1: Gantry axes 3.8 Examples Machine data The following machine data describes the original values at the beginning of the procedure. Individual settings must be corrected or added later according to the information below. Gantry machine data Axis 1 MD37100 $MA_GANTRY_AXIS_TYPE = 1 (gantry axis definition) MD37110 $MA_GANTRY_POS_TOL_WARNING =0 (gantry warning limit) MD37120 $MA_GANTRY_POS_TOL_ERROR = e.g.
  • Page 171: Setting Of Nck Plc Interface

    G1: Gantry axes 3.8 Examples 3.8.2 Setting of NCK PLC interface Introduction An automatic synchronization process during axis referencing must be disabled initially so as to prevent any damage to grouping axes that are misaligned. Disabling of automatic synchronization The PLC user program sets the following for the axis data block of axis 1: DB31, ...
  • Page 172 G1: Gantry axes 3.8 Examples DB31, ... DBB101.7 = 1 (gantry axis) In addition, the following steps must be taken: ● RESET ● Read off values in machine coordinate system: e.g. X = 0.941 Y = 0.000 XF = 0.000 ●...
  • Page 173: Setting Warning And Trip Limits

    G1: Gantry axes 3.8 Examples If Case A applies, the synchronization process can be started immediately (see "start synchronization process" step). If Case B applies, the offset "diff" must be calculated and taken into account: ● Measuring of diff ● By using two appropriate, right-angled reference points R' and R" in the machine bed (right in picture), the difference in position in JOG can be traversed.
  • Page 174: Data Lists

    G1: Gantry axes 3.9 Data lists MD37110 $MA_GANTRY_POS_TOL_WARNING (gantry warning limit) MD37120 $MA_GANTRY_POS_TOL_ERROR (gantry trip limit) MD37130 $MA_GANTRY_POS_TOL_REF (gantry trip limit for referencing) These should have the following scales of magnitude at the end of the customizing process: Note The same procedure must be followed when starting up a gantry grouping in which the coupled axes are driven by linear motors and associated measuring systems.
  • Page 175: Signals

    Gantry trip limit for referencing 37140 GANTRY_BREAK_UP Invalidate gantry axis grouping 3.9.2 Signals 3.9.2.1 Signals from mode group Signal name SINUMERIK 840D sl SINUMERIK 828D Active machine function REF DB11 DBX5.2 DB3100.DBX1.2 3.9.2.2 Signals from channel Signal name SINUMERIK 840D sl...
  • Page 176: Signals To Axis/Spindle

    G1: Gantry axes 3.9 Data lists 3.9.2.3 Signals to axis/spindle Signal name SINUMERIK 840D sl SINUMERIK 828D Start gantry synchronization DB31, ... DBX29.4 DB380x.DBX5005.4 No automatic synchronization DB31, ... DBX29.5 DB380x.DBX5005.5 3.9.2.4 Signals from axis/spindle Signal name SINUMERIK 840D sl...
  • Page 177: K6: Contour Tunnel Monitoring

    K6: Contour tunnel monitoring Brief description 4.1.1 Contour tunnel monitoring - 840D sl only Function The absolute movement of the tool tip in space is monitored. The function operates channel specific. Model A round tunnel with a definable diameter is defined around the programmed path of a machining operation.
  • Page 178: Programmable Contour Accuracy

    K6: Contour tunnel monitoring 4.1 Brief description Example The following figure is a diagram of the monitoring area shown by way of a simple example. Figure 4-1 Position of the contour tunnel around the programmed path As long as the calculated actual position of the tool tip remains inside the sketched tunnel, motion continues in the normal way.
  • Page 179: Contour Tunnel Monitoring - 840D Sl Only

    K6: Contour tunnel monitoring 4.2 Contour tunnel monitoring - 840D sl only Contour tunnel monitoring - 840D sl only Aim of the monitoring function The aim of the monitoring function is to stop the movement of the axes if axis deviation causes the distance between the tool tip (actual value) and the programmed path (setpoint) to exceed a defined value (tunnel radius).
  • Page 180: Programmable Contour Accuracy

    K6: Contour tunnel monitoring 4.3 Programmable contour accuracy Shutting down Monitoring can be stopped by enabling the machine data setting: MD21050 = 0.0. Analysis output The values of deviation of the actual value of the tool tip from the programmed path can – for analysis purposes –...
  • Page 181 K6: Contour tunnel monitoring 4.3 Programmable contour accuracy MD20470 $MC_CPREC_WITH_FFW (programmable contour accuracy) Value Meaning The "Programmable contour accuracy" function has no effect when feedforward control is also active. The "Programmable contour accuracy" function also acts for feedforward control. With active feedforward control, the reduction of the path velocity is calculated on the basis of the effective K factor with feedforward control.
  • Page 182 K6: Contour tunnel monitoring 4.3 Programmable contour accuracy Parameterization Contour accuracy The maximum contour error for the path of the geometry axes on curved contours is determined ● For MD20470 $MC_CPREC_WITH_FFW = 2 with the setting data: SD42450 $SC_CONTPREC (contour accuracy) ●...
  • Page 183: Constraints

    K6: Contour tunnel monitoring 4.4 Constraints The two CPRECON and CPRECOF modal G functions form the G function group 39 (programmable contour accuracy). Behavior for part program start and after reset / part program end For part program start and after reset / part program end, the configured control initial setting acts for the G function group 39: MD20110 $MC_RESET_MODE_MASK (definition of initial control settings after RESET / TP End)
  • Page 184: Data Lists

    K6: Contour tunnel monitoring 4.5 Data lists Data lists 4.5.1 Machine data 4.5.1.1 Channelspecific machine data Number Identifier: $MC_ Description 20110 RESET_MODE_MASK Determination of basic control settings after Reset / TP 20112 START_MODE_MASK Definition of the initial control settings after part program start 20470 CPREC_WITH_FFW...
  • Page 185: K7: Kinematic Chain - Only 840D Sl

    K7: Kinematic chain - only 840D sl Function description 5.1.1 Characteristics This section describes how the kinematic structure of a machine can be mapped using a kinematic chain and parameterized in the control using system variables for NC functions such as "collision avoidance"...
  • Page 186 K7: Kinematic chain - only 840D sl 5.1 Function description Kinematic chain A kinematic structure of a machine is described using a kinematic chain with the following properties: ● A kinematic chain consists of an arbitrary number of interconnected elements. ●...
  • Page 187 K7: Kinematic chain - only 840D sl 5.1 Function description A position or orientation change in an element, e.g. through a position change of the associated machine axis, affects all the following elements of the chain or parallel subchains. Parallel subchains If a parallel subchain branches off from an element e , then for the kinematics, the branch is always before the element.
  • Page 188: Commissioning

    K7: Kinematic chain - only 840D sl 5.2 Commissioning Origin and orientation of the world coordinate system can be freely selected. The following arrangement is recommended: ● Origin of the world coordinate system at the machine zero point ● Orientation of the world coordinate system so that the coordinate axes are arranged in the positive traversing direction of the machine linear main axes Direction vectors Direction vectors are always specified absolutely within the kinematic chain, i.e.
  • Page 189 K7: Kinematic chain - only 840D sl 5.2 Commissioning General The system variables to describe the elements of kinematic chains have the following properties: ● The prefix for all system variables of the kinematic chain is $NK_, (N for NC, K for kinematic). ●...
  • Page 190: Machine Data

    K7: Kinematic chain - only 840D sl 5.2 Commissioning 5.2.2 Machine data 5.2.2.1 Maximum number of elements The following machine data is used to specify how many elements are provided by the control for kinematic chains: MD18880 $MN_MM_MAXNUM_KIN_CHAIN_ELEM = <number> 5.2.2.2 Root element The following machine data is used to specify the root element, i.e.
  • Page 191: Nk_Name

    K7: Kinematic chain - only 840D sl 5.2 Commissioning System variable Meaning $NK_AXIS Machine axis or object name $NK_A_OFF Work offset for linear axis or rotary axis The system variables are described in detail in the following sections. Note Establish a defined initial state It is recommended that a defined initial state be generated before parameterizing the kinematic chain.
  • Page 192: Nk_Next

    K7: Kinematic chain - only 840D sl 5.2 Commissioning Example The 9th kinematic element is assigned the name "B axis": Program code Comment N100 $NK_NAME[8] = "B axis" ; 9th kinematic element, ; name = "B axis" 5.2.3.3 $NK_NEXT Function If the element is part of a kinematic chain, the name of the following element should be entered in the system variable.
  • Page 193: Nk_Parallel

    K7: Kinematic chain - only 840D sl 5.2 Commissioning 5.2.3.4 $NK_PARALLEL Function The name of the element that branches off before the current element should be entered in the system variable. The branching element is parallel to the current element. Changes to the current element, e.g.
  • Page 194: Nk_Type

    K7: Kinematic chain - only 840D sl 5.2 Commissioning 5.2.3.5 $NK_TYPE Function The element type should be entered in the system variable: Type Description AXIS_LIN The element describes a linear machine axis (linear axis) with the direction vector $NK_OFF_DIR (Page 195) and the work offset $NK_A_OFF (Page 195). AXIS_ROT The element describes a rotary machine axis (rotary axis) with the direction vector $NK_OFF_DIR (Page 198) and the work offset $NK_A_OFF (Page 198) or a spindle.
  • Page 195: Nk_Off_Dir, $Nk_Axis, $Nk_A_Off (Parameterization For $Nk_Type = Axis_Lin)

    K7: Kinematic chain - only 840D sl 5.2 Commissioning 5.2.3.6 $NK_OFF_DIR, $NK_AXIS, $NK_A_OFF (parameterization for $NK_TYPE = AXIS_LIN) $NK_OFF_DIR Function The direction vector along which the linear axis $NK_AXIS assigned to the element moves, should be entered in the system variable. The output coordinate system therefore results from the input coordinate system, offset by the current position value of the linear axis and the work offset specified in $NK_A_OFF.
  • Page 196 K7: Kinematic chain - only 840D sl 5.2 Commissioning Figure 5-4 Direction vector, general Program code Comment ; 9th kinematic element N100 $NK_OFF_DIR[8,0] = COS(90)*COS(10) ; 0 = X component N110 $NK_OFF_DIR[8,1] = SIN(90)*COS(10) ; 1 = Y component N120 $NK_OFF_DIR[8,2] = SIN(10) ;...
  • Page 197 K7: Kinematic chain - only 840D sl 5.2 Commissioning Example The 9th kinematic element is assigned the machine axis with the name V1 as linear axis. Program code Comment N100 $NK_AXIS[8] = "V1" ; 9. kin. element ; axis = machine axis V1 $NK_A_OFF Function An additional work offset can be entered in the system variable for the assigned machine axis...
  • Page 198: Nk_Off_Dir, $Nk_Axis, $Nk_A_Off (Parameterization For $Nk_Type = Axis_Rot)

    K7: Kinematic chain - only 840D sl 5.2 Commissioning 5.2.3.7 $NK_OFF_DIR, $NK_AXIS, $NK_A_OFF (parameterization for $NK_TYPE = AXIS_ROT) $NK_OFF_DIR Function The direction vector around which the rotary axis $NK_AXIS assigned to the element rotates, should be entered in the system variable. The output coordinate system is therefore calculated from the input coordinate system, rotated by the current position value of the rotary axis and the offset specified in $NK_A_OFF around the direction vector $NK_OFF_DIR.
  • Page 199 K7: Kinematic chain - only 840D sl 5.2 Commissioning Example The rotary axis of the 9th element rotates around the direction vector. The direction vector is the unit vector (1; 0; 0), rotated through α=90° in the X/Y plane and β=10° in the Y/Z plane, in relation to the world coordinate system.
  • Page 200 K7: Kinematic chain - only 840D sl 5.2 Commissioning Meaning Machine axis name $NK_AXIS: Data type: STRING Range of values: Machine axis name Default value: "" "" (empty string) System variable or element index <n>: Data type: Range of values: 0, 1, 2, ... ($MN_MM_MAXNUM_KIN_CHAIN_ELEM - 1) Machine axis name <value>: Data type:...
  • Page 201: Nk_Off_Dir, $Nk_Axis, $Nk_A_Off (Parameterization For $Nk_Type = Rot_Const)

    K7: Kinematic chain - only 840D sl 5.2 Commissioning Example The rotary axis zero point of the 9th kinematic elements is moved through 30.0° compared to the modelled kinematics. Program code Comment N100 $NK_A_OFF[8] = 30.0 ; 9. kin. element ;...
  • Page 202 K7: Kinematic chain - only 840D sl 5.2 Commissioning Example The output coordinate system of the 9th Elements therefore arises from the input coordinate system, rotated through the angle specified in $NK_A_OFF around the direction vector. The direction vector is the unit vector (1; 0; 0), rotated through α=90° in the X/Y plane and β=10° in the Y/Z plane, in relation to the world coordinate system.
  • Page 203: Nk_Off_Dir, $Nk_Axis, $Nk_A_Off (Parameterization For $Nk_Type = Offset)

    K7: Kinematic chain - only 840D sl 5.2 Commissioning Meaning Angle of rotation $NK_A_OFF: Data type: REAL Range of values: - max. REAL value ≤ x ≤ + max. REAL value Default value: System variable or element index <n>: Data type: Range of values: 0, 1, 2, ...
  • Page 204 K7: Kinematic chain - only 840D sl 5.2 Commissioning Coordinate value <value>: Data type: REAL Range of values: - max. REAL value ≤ x ≤ + max. REAL value Example The output coordinate system of the 9th element arises from the input coordinate system, offset by the offset vector with the following coordinates related to the world coordinate system: ●...
  • Page 205: Programming

    K7: Kinematic chain - only 840D sl 5.3 Programming Programming 5.3.1 Deletion of components (DELOBJ) Function The DELOBJ() function "deletes" components by resetting the assigned system variables to their default values: ● Elements from kinematic chains ● Protection areas, protection area elements and collision pairs ●...
  • Page 206 K7: Kinematic chain - only 840D sl 5.3 Programming Component type to be deleted <CompType>: Data type: STRING Value: "KIN_CHAIN_ELEM" Meaning: All kinematic elements (system variable $NK_... ) Value: "PROT_AREA" Meaning: Protection areas (Page 221) Value: "PROT_AREA_ELEM" Meaning: Protection area elements of machine protection areas (Page 232) and/ or automatic tool protection areas (Page 250) Value: "PROT_AREA_COLL_PAIRS"...
  • Page 207: Index Determination By Means Of Names (Nametoint)

    K7: Kinematic chain - only 840D sl 5.3 Programming Alarm suppression (optional) <NoAlarm>: Data type: BOOL Default value: FALSE Value Meaning FALSE In the event of an error (<RetVal> < 0), the program processing is stopped and an alarm displayed. TRUE In the event of an error, the program processing is not stopped and no alarm displayed.
  • Page 208: Data Lists

    K7: Kinematic chain - only 840D sl 5.4 Data lists Alarm suppression (optional) <NoAlarm>: Data type: BOOL Default value: FALSE Value Meaning FALSE In the event of an error (<RetVal> < 0), the program processing is stopped and an alarm displayed. TRUE In the event of an error, the program processing is not stopped and no alarm displayed.
  • Page 209 K7: Kinematic chain - only 840D sl 5.4 Data lists Identifier Description $NK_OFF_DIR Depends on $NK_TYPE: Offset or direction vector $NK_AXIS Name of the assigned machine axis or object name $NK_A_OFF Work offset in the element for linear or rotary axes Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 210 K7: Kinematic chain - only 840D sl 5.4 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 211: K8: Geometric Machine Modeling - Only 840D Sl

    K8: Geometric machine modeling - only 840D sl Function description 6.1.1 Features This section describes how the geometry of machine parts can be mapped based on protection areas and parameterized in the control using system variables for NC functions such as "collision avoidance".
  • Page 212 K8: Geometric machine modeling - only 840D sl 6.1 Function description ● Initialization status of the protection area ● Address of the geometric data of the machine element to the protected (only relevant for automatic protection areas) Each parameter is mapped by a system variable. The individual parameters and/or system variables are described in detail in Chapter "System variables: Protection areas (Page 221)".
  • Page 213 K8: Geometric machine modeling - only 840D sl 6.1 Function description Each parameter is mapped by a system variable. The individual parameters or system variables are described in detail in: ● Chapter "System variables: Protection area elements for machine protection areas (Page 232)"...
  • Page 214: Automatic Tool Protection Areas

    K8: Geometric machine modeling - only 840D sl 6.1 Function description 6.1.2 Automatic tool protection areas Unlike machine protection areas whose geometry is defined once during the machine modeling and then no longer changes, the geometry of a tool protection area can change with every tool change.
  • Page 215 K8: Geometric machine modeling - only 840D sl 6.1 Function description Tool modeling The model of a tool is created by the control heuristically from the tool data. The tool data used for this (L1, L2, L3, R) is always the the overall dimensions resulting from the individual components, e.g.
  • Page 216 K8: Geometric machine modeling - only 840D sl 6.1 Function description Tool type-dependent model generation The following tool types are distinguished for the model generation: ● Milling tool and all other tools that are neither turning tools or grinding tools –...
  • Page 217: Commissioning

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning The coordinate system of the geometric data always has its origin at the point from which the tool length offsets point to the tip of the tool. System variable All parameters for an automatic tool protection area can be read via System variable (Page 232).
  • Page 218 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Data type STRING All system variables of the STRING data type have the following properties: ● Maximum string length: 31 characters ● No distinction is made between upper and lower case Example: "Axis1"...
  • Page 219: Color Chart

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.1.3 Color chart The following color chart provides an overview of the RGB color values and the associated colors. A RGB color value comprises three bytes. One byte per color: 3rd byte 2nd byte 1st byte...
  • Page 220: Machine Data

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.2 Machine data 6.2.2.1 Maximum number of protection areas The maximum number of all types of parameterizable protection areas (Page 224) is specified with the machine data. MD18890 $MN_MM_MAXNUM_3D_PROT_AREAS = <number> 6.2.2.2 Maximum number of protection area elements for machine protection areas The maximum number of parameterizable protection area elements for machine protection...
  • Page 221: Creation Mode For Automatic Tool Protection Areas

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning The control system automatically models the protection area bodies based on the geometric data of the tool active at the time of creation. The number of triangles correspondingly increases: ● as the geometric complexity of the tool increases. ●...
  • Page 222: Np_Prot_Name

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning The system variables are described in detail in the following sections. Note Establish a defined initial state It is recommended that a defined initial state be generated before parameterizing the protection areas.
  • Page 223: Np_Chain_Elem

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.3.3 $NP_CHAIN_ELEM Function Enter the name of the kinematic element (Page 191) to which the protection area will be connected in the system variable. Note Reference coordinate system The geometric data of the protection area, starting from the first protection area element ($NP_1ST_PROT (Page 225)), refer to the local coordinate system of the kinematic element, with which the protection area is connected.
  • Page 224: Np_Prot_Type

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.3.4 $NP_PROT_TYPE Function The protection area type should be entered in the system variable: Type Description MACHINE Machine protection area The protection area body is defined using one or several protection area elements. $NP_1ST_PROT (Page 225) refers to the first protection area element.
  • Page 225: Np_1St_Prot

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.3.5 $NP_1ST_PROT Function Enter the name of the first protection area element (Page 233) in the system variable. Syntax $NP_1ST_PROT[<m>] = "<name>" Meaning Name of the first protection area element of the protection area $NP_1ST_PROT: Data type: STRING...
  • Page 226: Np_Prot_Color

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Behavior for value == name of a protection area element, type "FRAME" When activating the associated tool, the control creates a protection area element for the tool with a unique, internal name and protection area body generated from the geometric data of the tool.
  • Page 227: Np_Prot_D_Level

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Syntax $NP_PROT_COLOR[<m>] = <value> Meaning Alpha/transparency and color value of the protection area $NP_PROT_COLOR: Data type: DWORD Range of values: 00000000 - FFFFFFFF Default value: 0000000 (black, not visible) System variable or protection area index <m>: Data type: Range of values: 0, 1, 2, ...
  • Page 228: Np_Bit_No

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Syntax $NP_PROT_D_LEVEL[<m>] = <value> Meaning Detail level for the protection area $NP_PROT_D_LEVEL: Data type: Range of val‐ 0 ≤ D ≤ 3 ues: Default value: System variable or protection area index <m>: Data type: Range of val‐...
  • Page 229 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Syntax $NP_BIT_NO[<m>] = <number> Meaning Bit number of the interface signal to activate/deactivate the protection area $NP_BIT_NO: Data type: Range of values: -1, 0,1 ,2, ... ($MN_MM_MAXNUM_3D_INTERFACE_IN - 1) Default value: -1 (no interface signal selected) System variable or protection area index <m>:...
  • Page 230: Np_Init_Stat

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning DB10, DB10, DB10, DB10, DB10, DB10, DB10, DB10, → (PLC→NC) (NC→PLC) → (PLC→NC (NC→PLC → (PLC→NC (NC→PLC → (PLC→NC (NC→PLC DBX238.3 DBX230.3 DBX239.3 DBX231.3 DBX240.3 DBX232.3 DBX241.3 DBX233.3 DBX238.4 DBX230.4 DBX239.4 DBX231.4 DBX240.4 DBX232.4 DBX241.4 DBX233.4...
  • Page 231: Np_Index

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Example The initialization status of the 6th protection area is set to "P" (preactivated or PLC-controlled): Program code Comment N100 $NP_INIT_STAT[5] = "P" ; 6. Protection area, ; initialization status = "P" The current status depends on the state of the interface signal parameterized in $NP_BIT_NO (Page 228).
  • Page 232: System Variables: Protection Area Elements For Machine Protection Areas

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Type: Automatic tool protection area ($NP_PROT_TYPE == "TOOL") <i> <value> When tool management is active Without tool management Circular magazine: Tool location number Spindle number No circular magazine: Spindle number Magazine number TOA area 1) TOA area "1"...
  • Page 233: Np_Name

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Name Meaning $NP_COLOR Color and transparency of the protection area element. $NP_D_LEVEL Detail level for the protection area element $NP_USAGE Use of the protection area element $NP_TYPE Type of the protection area element $NP_FILENAME File name of the STL file that contains the geometric data of the pro‐...
  • Page 234: Np_Next

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Meaning Name of the protection area element $NP_NAME: Data type: STRING Default value: "" (empty string) System variable or protection area element index <n>: Data type: Range of values: 0, 1, 2, ... ($MN_MM_MAXNUM_3D_PROT_AREA_ELEM - Name of the protection area element <name>: Data type:...
  • Page 235: Np_Nextp

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Syntax $NK_NEXT[<n>] = "<name>" Meaning Name of the following protection area element $NP_NEXT: Data type: STRING Range of values: All names contained in $NP_NAME (Page 233) Default value: "" (empty string) System variable or protection area element index <n>: Data type:...
  • Page 236 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Application example The separate subchains can be used, for example, to model different machine parts of a protection area for visualization or collision avoidance. Typically "C" (collision avoidance) is specified for the protection area element referred to by $NP_NEXT for use in $NP_USAGE (Page 239), and for the protection area element referred to in $NP_NEXTP, the value “V”...
  • Page 237: Np_Color

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.4.5 $NP_COLOR Function Enter the protection area element-specific value for alpha/transparency and color (ARGB) in the system variable. This value is used for the display of the protection area element on the user interface.
  • Page 238: Np_D_Level

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Example The 19th protection area is to be displayed half transparent and in a green-blue color on the user interface: ● AA = 7F = 127 ≙ 50% transparency ● RR (red) = 00 ≙ no red component ●...
  • Page 239: Np_Usage

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning System variable or protection area element index <m>: Data type: Range of val‐ 0, 1, 2, ... ($MN_MM_MAXNUM_3D_PROT_AREA_ELEM ues: - 1) Detail level <value>: Data type: Example The 19th protection area is to always to be displayed ⇒ detail level 0: Program code Comment N100 $NP_PROT_D_LEVEL[18] = 0...
  • Page 240: Np_Type

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Meaning Use of the protection area element $NP_USAGE: Data type: CHAR Range of values: "V", "v", "C", "c", "A", "a" Value Meaning "V" or "v" Only visualization, no collision calculation "C"...
  • Page 241 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning coordinate system. At the same time as the definition of the body, the local coordinate system can be transformed via the following system variables: ● Offset: $NP_OFF (Page 246) ● Direction vector of the rotation: $NP_DIR (Page 247) ●...
  • Page 242 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Radius of the sphere: Parameter specification in $NP_PARA (Page 245): Radius Type: "CYLINDER" A protection area element of the "CYLINDER" type defines a cylinder in the local coordinate system of the protection area element. The mid-point of the cylinder is at the origin of the local coordinate system.
  • Page 243: Np_Filename

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Figure 6-3 Example bodies in STL format Syntax $NP_TYPE[<n>] = "<type>" Meaning Type of the protection area element $NP_TYPE: Data type: STRING Range of values: "FRAME", "BOX", "SPHERE", "CYLINDER", "FILE" Default value: ""...
  • Page 244 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning File format The description of the body must be available in STL format (StandardTesselationFormat) and in ASCII coding. Search path The specified file is sought in the predefined directories on the CompactFlash card in this order: 1.
  • Page 245: Np_Para

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Note Maximum length of the file name The maximum file name length, including point and the file extension is 49 characters. For more than 49 characters, an alarm is displayed when generating an archive. 6.2.4.10 $NP_PARA Function...
  • Page 246: Np_Off

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Example The 19th protection area element is a box with the dimensions: ● Length: 50.0 ● Width: 100.0 ● Height: 75.5 Program code Comment ; 19. Protection area element, N100 $NP_TYPE[18] = "BOX" ;...
  • Page 247: Np_Dir

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Example The local coordinate system of the 19th protection area element is moved by the following vector compared to the coordinate system of the previous protection area element: ● X direction: 25.0 ●...
  • Page 248 K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Meaning Direction vector $NP_DIR: Data type: REAL Range of values: - max. REAL value ≤ x ≤ ± max. REAL value Default value: (0.0, 0.0, 0.0) System variable or protection area element index <n>: Data type: Range of values: 0, 1, 2, ...
  • Page 249: Np_Ang

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.4.13 $NP_ANG Function The angle through which the local coordinate system of the protection area element is rotated around the direction vector ($NP_DIR (Page 247)) compared to the coordinate system of the previous protection area element should be entered in the system variable.
  • Page 250: System Variables: Protection Area Elements For Automatic Tool Protection Areas

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning Program code Comment ; 19. Protection area element, direction vector and angle of rotation N100 $NP_DIR[18.0] = COS(90)*COS(10) ; 0 = X component N110 $NP_DIR[18.1] = SIN(90)*COS(10) ; 1 = Y component N120 $NP_DIR[18.2] = SIN(10) ;...
  • Page 251: Boundary Conditions

    K8: Geometric machine modeling - only 840D sl 6.2 Commissioning 6.2.6 Boundary conditions Protection area bodies for spindles For spindles that are not in position-controlled operation, the associated protection area bodies are only modeled statically. As a result the following boundary conditions must be met when modeling protection area bodies connected with a spindle as a kinematic element: ●...
  • Page 252: Data Lists

    K8: Geometric machine modeling - only 840D sl 6.3 Data lists Data lists 6.3.1 Machine data 6.3.1.1 NC-specific machine data Number Identifier: $MN_ Description MD18890 $MN_MM_MAXNUM_3D_PROT_AREAS Maximum number of protection areas MD18892 $MN_MM_MAXNUM_3D_PROT_AREA_ELEM Maximum number of protection area elements MD18893 $MN_MM_MAXNUM_3D_T_PROT_ELEM Maximum number of tool protection area elements MD18897...
  • Page 253 K8: Geometric machine modeling - only 840D sl 6.3 Data lists Identifier Description $NP_OFF Offset vector $NP_DIR Direction vector $NP_ANG Angle of rotation Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 254 K8: Geometric machine modeling - only 840D sl 6.3 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 255: K9: Collision Avoidance - Only 840D Sl

    K9: Collision avoidance - only 840D sl Function description 7.1.1 Characteristics The "Collision avoidance" function is used to prevent collisions of machine parts and tool cutting edges while the machine axes are being traversed. To do this, the function cyclically calculates the clearance to the protection areas enveloping the bodies to be protected.
  • Page 256 K9: Collision avoidance - only 840D sl 7.1 Function description 7. Activation of the protection areas to be monitored. See Chapter "Setting the protection area state (PROTS) (Page 278)". 8. Optional: Use of the extended functions and system variables Limits of the collision avoidance The function cannot guarantee complete protection against a collision when traversing machine parts, tools of workpieces.
  • Page 257: Reaction Of The Control To A Risk Of Collision

    K9: Collision avoidance - only 840D sl 7.1 Function description 7.1.2 Reaction of the control to a risk of collision The collision avoidance takes the following parameterizable limit values into account for the collision detection: ● Collision tolerance ● Safety clearance ①...
  • Page 258 K9: Collision avoidance - only 840D sl 7.1 Function description The collision tolerance is set the same for all collision pairs via MD10619 $MN_COLLISION_TOLERANCE (Page 267). Note Difference between collision tolerance and safety clearance The collision tolerance can be violated and is permissible. The safety clearance is always maintained.
  • Page 259 K9: Collision avoidance - only 840D sl 7.1 Function description Traversing motion is always canceled when the collision clearance is reached. Continuation of the traversing motion always requires a new travel request (e.g by pressing a traversing key), irrespective of the traversing direction. Responses in the operating mode: MDI If two protection areas approach one another when traversing in the MDI mode, the traversing velocity is continuously braked down to standstill when the collision tolerance is reached.
  • Page 260: State Diagram: Protection Area

    K9: Collision avoidance - only 840D sl 7.1 Function description 7.1.3 State diagram: Protection area Protection area Operating mode ① UpdateAllCaSysVar(SB) function All system variables of the collision avoidance are read-in in NCK-internal variables: int... = $N... Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 261: Tools

    K9: Collision avoidance - only 840D sl 7.1 Function description ② UpdateAllCaSysVarExeptInitStat(SB) function As with the UpdateAllCaSysVar(SB) function, but the $NP_INIT_STAT system variable is not read in. Internally in the NCK, the last value of the intInitStat initialization status is therefore retained.
  • Page 262 ● TMMVTL: PI service "Prepare magazine location for loading, unload tool" Reference Function Manual Basic Functions; Chapter: "P3: Basic PLC program for SINUMERIK 840D sl" > "Component descriptions" > "PI services" > "PI service: TMMVTL" ● MVTOOL: Command to move a tool Reference Function Manual Tool Management;...
  • Page 263: Boundary Conditions

    K9: Collision avoidance - only 840D sl 7.1 Function description Example: Revolver magazine on a lathe The revolver magazine on a lathe is fully modeled in the machine model of the collision avoidance: ● The geometry of the magazine and the tools located inside it ●...
  • Page 264 K9: Collision avoidance - only 840D sl 7.1 Function description Compensations The various compensation functions of the NC – for instance, temperature, spindle and pitch error and sag compensation – ensure that positions programmed in the workpiece coordinate system are actually assumed in the machine coordinate system. The collision avoidance takes into account the position corrections made by the compensation functions.
  • Page 265: Commissioning

    K9: Collision avoidance - only 840D sl 7.2 Commissioning AUTOMATIC modes: Incomplete protection area data for collision When a large number of protection areas have been configured, in exceptional cases, the following behavior can occur: ● Several protection areas have approached one another, down to the collision tolerance ●...
  • Page 266: Machine Data

    K9: Collision avoidance - only 840D sl 7.2 Commissioning Data type STRING All system variables of the STRING data type have the following properties: ● Maximum string length: 31 characters ● No distinction is made between upper and lower case Example: "Axis1"...
  • Page 267: Collision Tolerance

    K9: Collision avoidance - only 840D sl 7.2 Commissioning MD19830 $ON_COLLISION_MASK.Bit x = 1 Meaning Collision avoidance (machine, tool) 1 - 31 reserved 7.2.2.2 Collision tolerance The collision tolerance (accuracy of the collision check) for all protection areas of the NC monitored for collision is set with the machine data.
  • Page 268: Maximum Memory Space

    K9: Collision avoidance - only 840D sl 7.2 Commissioning See also Reaction of the control to a risk of collision (Page 257) 7.2.2.4 Maximum memory space The maximum value of the memory space in KB that can be allocated to the collision avoidance is set with the machine data.
  • Page 269: Protection Levels For Collision Avoidance On/Off

    K9: Collision avoidance - only 840D sl 7.2 Commissioning MD18898 $MN_MM_MAXNUM_3D_COLL_PAIRS = <value> <Val‐ Meaning ue> The following applies to the maximum number of possible collision pairs MCP: MCP = maximum value of the machine data x > 0 The following applies to the maximum number of possible collision pairs MCP: MCP = x, with 0 <...
  • Page 270: Np_Coll_Pair

    K9: Collision avoidance - only 840D sl 7.2 Commissioning The system variables are described in detail in the following sections. Note Establish a defined initial state It is recommended that a defined initial state be generated before parameterizing the collision avoidance.
  • Page 271: Np_Safety_Dist

    K9: Collision avoidance - only 840D sl 7.2 Commissioning Meaning Name of the first or second protection area of a collision pair $NP_COLL_PAIR: Data type: STRING Default value: "" "" (empty string) System variable or collision pair index <m>: Data type: Range of values: 0, 1, 2, ...
  • Page 272 K9: Collision avoidance - only 840D sl 7.2 Commissioning If the value 0.0 is entered in the system variable, the safety clearance set in the machine data applies. Syntax $NP_SAFETY_DIST[<m>] = <value> Meaning Safety clearance of the collision pair $NP_SAFETY_DIST: Data type: REAL Default value:...
  • Page 273: Extend System Variables

    K9: Collision avoidance - only 840D sl 7.2 Commissioning 7.2.4 Extend system variables 7.2.4.1 Overview Further information on the internal states and values of the collision avoidance can be read via the following system variables: ● State data (Page 273) ●...
  • Page 274: Memory Requirement

    K9: Collision avoidance - only 840D sl 7.2 Commissioning 7.2.4.3 Memory requirement The data for the memory requirement of the collision avoidance can be read via the following system variables (OPI variables). System variable OPI variable Meaning $AN_COLL_MEM_AVAILABLE anCollMemAvailable Size of the memory space in KB reserved for the collision avoidance.
  • Page 275: Programming

    K9: Collision avoidance - only 840D sl 7.3 Programming Table 7-2 Machine coordinate system (MCS) System variable OPI variable Meaning Total braking distance $AA_DTBREM[<a>] aaDtbrem Estimated, linearly approximated total braking dis‐ tance Proportional braking distances for superimposed motions $AA_DTBREM_CMD[<a>] aaDtbremCmd Command component $AA_DTBREM_CORR[<a>] aaDtbremCorr...
  • Page 276 K9: Collision avoidance - only 840D sl 7.3 Programming Meaning Check whether part of a collision pair COLLPAIR: Function return value <RetVal>: Data type: Value Meaning ≥ 0 The two protection areas form a collision pair. Return value == collision pair index m, see Section "$NP_COLL_PAIR (Page 270)" Either two strings have not been specified or at least one of the two is the zero string.
  • Page 277: Request Recalculation Of The Machine Model Of The Collision Avoidance (Prota)

    K9: Collision avoidance - only 840D sl 7.3 Programming 7.3.2 Request recalculation of the machine model of the collision avoidance (PROTA) Function If system variables of the kinematic chain $NK_..., the geometric machine modeling or the collision avoidance $NP_... are written in the part program, the PROTA procedure must subsequently be called so that the change becomes effective in the NC-internal machine model of the collision avoidance.
  • Page 278: Setting The Protection Area State (Prots)

    K9: Collision avoidance - only 840D sl 7.3 Programming See also Setting the protection area state (PROTS) (Page 278) 7.3.3 Setting the protection area state (PROTS) Function The PROTS() procedure sets the state of protection areas to the specified value. Syntax PROTS(<state>{, <name>}) Meaning...
  • Page 279 K9: Collision avoidance - only 840D sl 7.3 Programming Function properties: ● The clearance calculation is performed independent of the protection area state (activated, deactivated, preactivated). ● The clearance is calculated at the interpretation instant of the function with the axis positions valid at the end of the previous block.
  • Page 280: Example

    K9: Collision avoidance - only 840D sl 7.4 Example Measuring system (inch/metric) for clearance and clearance vector (optional) <System>: Data type: BOOL Default value: FALSE Value Meaning FALSE Measuring system corresponding to the currently active G func‐ tion from G group 13 (G70, G71, G700, G710). TRUE Measuring system corresponding to the set basic system: MD52806 $MN_ISO_SCALING_SYSTEM...
  • Page 281 K9: Collision avoidance - only 840D sl 7.4 Example Machine data: $MN_ Value MD18896 MM_MAXNUM_3D_COLLISION MD18897 MM_MAXNUM_3D_INTERFACE_IN MD18899 PROT_AREA_TOOL_MASK Principle design of the 3-axis milling machine The principle machine design is shown in the following diagram. The machine parts and/or protection areas are assigned to the following machine axes. Machine parts and/or protection areas Machine axis Table...
  • Page 282 K9: Collision avoidance - only 840D sl 7.4 Example Dimension drawing The dimensions of the protection area elements as well as their position (vectors to the center point of the protection area element) referred to the machine zero point are specified in the following dimension drawing.
  • Page 283 K9: Collision avoidance - only 840D sl 7.4 Example Collision pairs For the example, it is assumed that only the following collision pairs are to be taken into account: ● Tool adapter - table ● Tool - table Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 284: Part Program Of The Machine Model

    K9: Collision avoidance - only 840D sl 7.4 Example 7.4.2 Part program of the machine model Program code ;*********************************************************** ;************************* Example ************************ ; milling machine: 3 linear axes, 1 spindle table => X1, Y1 Z axis, tool adapter, tool => Z1 ;*********************************************************** ;...
  • Page 285 K9: Collision avoidance - only 840D sl 7.4 Example Program code ;=========================================================== ; initialization of the collision data ;=========================================================== MSG("protection areas") G4 F3 ; reset all parameters to their initial setting RETVAL = DELOBJ("KIN_CHAIN_ELEM") IF (RETVAL <> 0) MSG("error: DELOBJ KIN_CHAIN_ELEM") G4 F5 ENDIF RETVAL = DELOBJ("PROT_AREA_ALL")
  • Page 286 K9: Collision avoidance - only 840D sl 7.4 Example Program code ; ---------------------------------------------------------- ; kinematic element: X axis ; ---------------------------------------------------------- $NK_NAME[C_NKE] = "X axis" $NK_NEXT[C_NKE] = "Y axis" $NK_PARALLEL[C_NKE] = "Z axis" $NK_TYPE[C_NKE] = "AXIS_LIN" $NK_OFF_DIR[C_NKE, 0] = 1.0 $NK_OFF_DIR[C_NKE, 1] = 0.0 $NK_OFF_DIR[C_NKE, 2] = 0.0 $NK_AXIS[C_NKE] = "X1"...
  • Page 287 K9: Collision avoidance - only 840D sl 7.4 Example Program code ;=========================================================== ; protection areas with protection area elements ;=========================================================== ; protection area 1: Column ; ---------------------------------------------------------- $NP_PROT_NAME[C_NPC] = "column" $NP_PROT_TYPE[C_NPC] = "MACHINE" $NP_CHAIN_ELEM[C_NPC] = "ROOT" $NP_1ST_PROT[C_NPC] = "SBE column" $NP_PROT_COLOR[C_NPC] = 'HFFA0A0A4' ;...
  • Page 288 K9: Collision avoidance - only 840D sl 7.4 Example Program code ;++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ; protection area 2: Tool adapter ; ---------------------------------------------------------- $NP_PROT_NAME[C_NPC] = "tool adapter" $NP_PROT_TYPE[C_NPC] = "MACHINE" $NP_CHAIN_ELEM[C_NPC] = "Z axis" $NP_1ST_PROT[C_NPC] = "SBE tool adapter" $NP_PROT_COLOR[C_NPC] = 'HFF0000FF' ; AARRGGBB $NP_BIT_NO[C_NPC] = -1 $NP_INIT_STAT[C_NPC]...
  • Page 289 K9: Collision avoidance - only 840D sl 7.4 Example Program code ; ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ; protection area 3: Tool ; ---------------------------------------------------------- $NP_PROT_NAME[C_NPC] = "WKZ" $NP_PROT_TYPE[C_NPC] = "TOOL" $NP_CHAIN_ELEM[C_NPC] = "Z axis" $NP_1ST_PROT[C_NPC] = "" $NP_PROT_COLOR[C_NPC] = 'HFFFF0000' ; AARRGGBB $NP_BIT_NO[C_NPC] = -1 $NP_INIT_STAT[C_NPC] = "A"...
  • Page 290 K9: Collision avoidance - only 840D sl 7.4 Example Program code ; ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ; protection area 4: Z axis ; ---------------------------------------------------------- $NP_PROT_NAME[C_NPC] = "Z axis" $NP_PROT_TYPE[C_NPC] = "MACHINE" $NP_CHAIN_ELEM[C_NPC] = "Z axis" $NP_1ST_PROT[C_NPC] = "SBE Z axis" $NP_PROT_COLOR[C_NPC] = 'HFFA0A0A4' ;...
  • Page 291 K9: Collision avoidance - only 840D sl 7.4 Example Program code ; ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ; protection area 5: Table ; -------------------------------------------------------- $NP_PROT_NAME[C_NPC] = "table" $NP_PROT_TYPE[C_NPC] = "MACHINE" $NP_CHAIN_ELEM[C_NPC] = "Y axis" $NP_1ST_PROT[C_NPC] = "SBE table" $NP_PROT_COLOR[C_NPC] = 'HFF00FF00' ; AARRGGBB $NP_BIT_NO[C_NPC] = -1 $NP_INIT_STAT[C_NPC] = "A"...
  • Page 292: Data Lists

    K9: Collision avoidance - only 840D sl 7.5 Data lists Program code ;=========================================================== ; collision pairs ;=========================================================== $NP_COLL_PAIR[C_NPP, 0] = "Tool adapter" $NP_COLL_PAIR[C_NPP, 1] = "Table" C_NPP = C_NPP + 1 ; next collision pair $NP_COLL_PAIR[C_NPP, 0] = "T" $NP_COLL_PAIR[C_NPP, 1] = "Table" C_NPP = C_NPP + 1 ;...
  • Page 293: Signals

    Correction component (MCS) $AA_DTBREM_DEP Coupling component (MCS) 7.5.3 Signals 7.5.3.1 Signals to NC Signal name SINUMERIK 840D sl SINUMERIK 828D Collision avoidance: Deactivate protection area group DB10.DBX58.0 - 7 Collision avoidance: Activate protection area DB10.DBX234.0 - DBX241.7 7.5.3.2 Signals from NC...
  • Page 294: Signals From Channel

    K9: Collision avoidance - only 840D sl 7.5 Data lists 7.5.3.3 Signals from channel Signal name SINUMERIK 840D sl SINUMERIK 828D Collision avoidance: Stop DB21, ..DBX377.0 7.5.3.4 Signals from axis Signal name SINUMERIK 840D sl SINUMERIK 828D Collision avoidance: Velocity reduction DB31, ...
  • Page 295: M3: Coupled Axes

    M3: Coupled axes Coupled motion 8.1.1 Brief description 8.1.1.1 Function The "coupled motion" function enables the definition of simple axis links between a master axis and a slave axis, taking into consideration a coupling factor. Coupled motion has the following features: ●...
  • Page 296: General Functionality

    M3: Coupled axes 8.1 Coupled motion However, for basic operation of generic coupling, the following restrictions apply: ● The maximum number of coupled motion groupings is limited to 4. ● Only 1 leading axes may be assigned to each coupled motion axis. ●...
  • Page 297 M3: Coupled axes 8.1 Coupled motion Figure 8-1 Application example: Two-sided machining Multiple couplings Up to 2 leading axes can be assigned to one coupled motion axis. The traversing movement of the coupled motion axis then results from the sum of the traversing movements of the leading axes.
  • Page 298 M3: Coupled axes 8.1 Coupled motion Activating/deactivating Coupled motion can be activated/deactivated via the part programs and synchronous actions. In this context please ensure activating/deactivating is undertaken with the same programming: ● Activate: Part program → Deactivate: Part program ● Activate: Synchronous action → Deactivate: Synchronized action Synchronization on-the-fly If switch on is performed while the leading axis is in motion, the coupled motion axis is first accelerated to the velocity corresponding to the coupling.
  • Page 299 M3: Coupled axes 8.1 Coupled motion Distance-to-go: Coupled motion axis The distance-to-go of a coupled motion axis refers to the total residual distance to be traversed from dependent and independent traversing. Delete distance-to-go: Coupled motion axis Delete distance-to-go for a coupled motion axis only results in aborting of the independent traversing movement of the leading axis.
  • Page 300: Programming

    M3: Coupled axes 8.1 Coupled motion 8.1.3 Programming 8.1.3.1 Definition and switch on of a coupled axis grouping (TRAILON) Definition and switch on of a coupled axis grouping take place simultaneously with the TRAILON part program command. Programming Syntax: TRAILON(<coupled motion axis>, <leading axis>, [<coupling factor>]) Effectiveness: modal...
  • Page 301: Effectiveness Of Plc Interface Signals

    M3: Coupled axes 8.1 Coupled motion Parameters: Coupled motion Type: AXIS axis: Range of val‐ All defined axis and spindle names in the channel ues: Leading axis: Type: AXIS Range of val‐ All defined axis and spindle names in the channel ues: Example: Program code...
  • Page 302: Status Of Coupling

    M3: Coupled axes 8.1 Coupled motion Leading axis When a coupled axis grouping is active, the interface signals (IS) of the leading axis are applied to the appropriate coupled motion axis via the axis coupling, i.e. ● A position offset or feed control action of the leading axis is applied via the coupling factor to effect an appropriate position offset or feed control action in the coupled motion axis.
  • Page 303: Dynamics Limit

    M3: Coupled axes 8.1 Coupled motion Value Meaning Master value coupling Following axis of electronic gearbox Note Only one coupling mode may be active at any given time. 8.1.6 Dynamics limit The dynamics limit is dependent on the type of activation of the coupled axis grouping: ●...
  • Page 304: Supplementary Conditions

    M3: Coupled axes 8.1 Coupled motion 8.1.7 Supplementary conditions Control system dynamics It is recommended to align the position control parameters of the leading axis and the coupled motion axis within a coupled axis group. Note Alignment of the position control parameters of the leading axis and the coupled motion axis can be performed via a parameter set changeover.
  • Page 305: Curve Tables - 840D Sl Only

    M3: Coupled axes 8.2 Curve tables - 840D sl only Program code Comment G0 Z10 ; Infeed Z and W axes in opposite axial directions G0 Y20 ; Infeed of Y and V axes in same axis direction G1 Y22 V25 ;...
  • Page 306: Preconditions

    M3: Coupled axes 8.2 Curve tables - 840D sl only Curve tables can be saved in dynamic NC memory for faster access. Please note that tables need to be reloaded after run-up. Axis groupings with curve tables must be reactivated independently of the storage location of the curve table after POWER ON.
  • Page 307: Memory Organization

    M3: Coupled axes 8.2 Curve tables - 840D sl only Reference: Programming Manual, Production Planning Curve segments are used if: ● Polynomes or circles are programmed ● Spline is active ● Compressor is active ● Polynomials or circles are generated internally (chamfer/rounding, approximate positioning with G643, WRK etc.) Tool radius compensation Curve tables are available in which it is possible to specify the tool radius compensation in the...
  • Page 308 M3: Coupled axes 8.2 Curve tables - 840D sl only Memory optimization In a curve table with linear segments, the linear segments can be stored efficiently in the memory only if the two following machine data items are > 0: MD18403 $MC_MM_NUM_CURVE_SEG_LIN (number of linear curve segments in the static NC memory) MD18409 $MC_MM_NUM_CURVE_SEG_LIN_DRAM (number of linear curve segments in...
  • Page 309: Commissioning

    M3: Coupled axes 8.2 Curve tables - 840D sl only Overwriting curve tables Curve tables that are not active in a master value coupling and are locked with CTABLOCK() may be overwritten. Deleting curve tables Curve tables that are not active in a master value coupling and are locked with CTABLOCK() may be overwritten.
  • Page 310: Tool Radius Compensation

    M3: Coupled axes 8.2 Curve tables - 840D sl only 8.2.4.2 Tool radius compensation MD20900 Tool radius compensation can produce segments for which the following axis or leading axis have no movement. A missing movement of the following axis does not normally represent any problem.
  • Page 311: Programming

    M3: Coupled axes 8.2 Curve tables - 840D sl only 8.2.5 Programming Definition The following modal language commands work with curve tables: (The parameters are explained at the end of the list of functions.) ● Beginning of definition of a curve table: CTABDEF(following axis, leading axis, n, applim, memType) ●...
  • Page 312 M3: Coupled axes 8.2 Curve tables - 840D sl only Access to curve table segments ● Read start value (following axis value) of a table segment CTABSSV(leading value, n, degrees, [following axis, leading axis]) ● Read end value (following axis value) of a table segment CTABSEV(master value, n, degrees, [following axis, master axis]) Note If curve table functions such as CTAB(), CTABINV(), CTABSSV() etc., in synchronous...
  • Page 313 M3: Coupled axes 8.2 Curve tables - 840D sl only All curve tables, irrespective of memory type CTABUNLOCK() All curve tables in the specified memory type CTABUNLOCK(, , memType) Other commands for calculating and differentiating between curve tables for applications for diagnosing and optimizing the use of resources: ●...
  • Page 314 M3: Coupled axes 8.2 Curve tables - 840D sl only ● Maximum number of possible curve segments in memory memType. CTABMSEG(memType, segType) ● Number of polynomials already used in memory memType. CTABPOL(memType) ● Number of curve polynomials used by curve table number n. CTABPOLID(n) ●...
  • Page 315 M3: Coupled axes 8.2 Curve tables - 840D sl only ● applim: Behavior at the curve table edges. – 0 non-periodic (table is processed only once, even for rotary axes). – 1 periodic, modulo (the modulo value corresponds to the LA table values). –...
  • Page 316 M3: Coupled axes 8.2 Curve tables - 840D sl only ● Axis names from gantry axis groups cannot be used to define a table (only leading axis are possible). ● Depending on the following machine data, jumps in the following axis may be tolerated if a movement is missing in the leading axis.
  • Page 317: Access To Table Positions And Table Segments

    M3: Coupled axes 8.2 Curve tables - 840D sl only Example 2 Example of a curve table with active tool radius compensation: Prior to definition of a curve table with CTABDEF(), tool radius compensation must not be active; otherwise alarm 10942 is generated. This means that tool radius compensation must be activated within the definition of the curve table.
  • Page 318 M3: Coupled axes 8.2 Curve tables - 840D sl only 1:1 to the internal segments of the curve table. This is always the case if only G1 blocks or axis polynomials are used to define the curve tables and no other functions are active. Programmed sections may under certain circumstances not be transformed unchanged into internal curve segments if: 1.
  • Page 319 M3: Coupled axes 8.2 Curve tables - 840D sl only Identifying the segment associated with master value X Example of reading the segment starting and end values for determining the curve segment associated with master value X = 30 using CTABSSV and CTABSEV: Program code Comment N10 DEF REAL STARTPOS...
  • Page 320 M3: Coupled axes 8.2 Curve tables - 840D sl only R10 =CTABTEP(n, degrees, F axis), following value at the beginning of the curve table Value range of the following value The following example illustrates how the minimum and maximum values of the table are determined using CTABTMIN and CTABTMAX: Program code Comment...
  • Page 321: Activation/Deactivation

    M3: Coupled axes 8.2 Curve tables - 840D sl only Figure 8-3 Determining the minimum and maximum values of the table 8.2.7 Activation/deactivation Activation The coupling of real axes to a curve table is activated through this command: LEADON (<Following axis>, <Leading axis>, <n>) with <n>...
  • Page 322: Modulo-Leading Axis Special Case

    M3: Coupled axes 8.2 Curve tables - 840D sl only Deactivation is possible: ● In the part program ● in synchronized actions Note While programming LEADOF, the abbreviated form is also possible without specification of the leading axis. Example: N1010 LEADOF(A,X) ;...
  • Page 323: Effectiveness Of Plc Interface Signals

    M3: Coupled axes 8.2 Curve tables - 840D sl only Basic setting after run-up No curve tables are active after run-up. 8.2.10 Effectiveness of PLC interface signals Dependent following axis With respect to the motion of a following axis that is dependent on the leading axis, only the following axis interface signals that effect termination of the motion (e.g.
  • Page 324 M3: Coupled axes 8.2 Curve tables - 840D sl only a) Curve tables ● Determine total number of defined tables. The definition applies to all memory types (see also CTABNOMEM) CTABNO() ● Number of defined tables in SRAM or DRAM of NC memory. CTABNOMEM (memType) If memType is not specified, the memory type specified in the following machine data: MD20905 $MC_CTAB_DEFAULT_MEMORY_TYPE (default memory type for curve tables)
  • Page 325 M3: Coupled axes 8.2 Curve tables - 840D sl only To prevent this from happening, the curve tables concerned can be locked, using the CTABLOCK(...) language command. In this case, it should be noted that the curve tables concerned are then unlocked with CTABUNLOCK(). ●...
  • Page 326 M3: Coupled axes 8.2 Curve tables - 840D sl only ● Determine number of used curve segments of the type memType in the memory range CTABSEGID(n, segType) Result: >= 0: Number of curve segments -1: Curve table with number n does not exist -2: segType not equal "L"...
  • Page 327: Supplementary Conditions

    M3: Coupled axes 8.2 Curve tables - 840D sl only 8.2.12 Supplementary conditions Transformations Transformations are not permissible in curve tables. TRAANG is an exception. TRAANG If TRAANG is programmed, the rule of motion programmed in the basic co-ordinate system is transformed to the associated machine co-ordinate system.
  • Page 328 M3: Coupled axes 8.2 Curve tables - 840D sl only %_N_TAB_1_NOTPERI_MPF N21 PO[X]=(147.754,-0.116,0.103) PO[Y]=(3.359,-0.188,0.277) N22 PO[X]=(174.441,0.578,-0.206) PO[Y]=(0.123,1.925,0.188) N23 PO[X]=(185.598,-0.007,0.005) PO[Y]=(-0.123,0.430,-0.287) N24 PO[X]=(212.285,0.040,-0.206) PO[Y]=(-3.362,-2.491,0.190) N25 PO[X]=(227.395,-0.193,0.103) PO[Y]=(-6.818,-0.641,0.276) N26 PO[X]=(254.098,0.355,-0.053) PO[Y]=(-11.710,0.573,0.723) N26 PO[X]=(254.098,0.355,-0.053) PO[Y]=(-11.710,0.573,0.723) N27 PO[X]=(310.324,0.852,-0.937) PO[Y]=(-7.454,11.975,-1.720) N28 PO[X]=(328.299,-0.209,0.169) PO[Y]=(-3.197,0.726,-0.643) N29 PO[X]=(360.031,0.885,-0.413) PO[Y]=(0.000,-3.588,0.403) CTABEND N30 M30 Definition of a periodic curve table...
  • Page 329: Master Value Coupling - 840D Sl Only

    M3: Coupled axes 8.3 Master value coupling - 840D sl only N10 DEF REAL DEPPOS N230 M30 Master value coupling - 840D sl only 8.3.1 Product brief 8.3.1.1 Function The "axial master value coupling" function can be used to process short programs cyclically with close coupling of the axes to one another and a master value that is either generated internally or input from an external source.
  • Page 330 M3: Coupled axes 8.3 Master value coupling - 840D sl only Virtual leading axis / simulated master value If the leading axis is not interpolated by the same NCU, the interpolator that is implemented in the NCU for this particular leading axis can be used for master value simulation. The following machine data settings must be defined for this: MD30132 $MA_IS_VIRTUAL_AX[n] = 1 (axis is virtual axis) MD30130 $MA_CTRLOUT_TYPE[n] = 0 (simulation as output type of setpoint)
  • Page 331 M3: Coupled axes 8.3 Master value coupling - 840D sl only Figure 8-4 Master value coupling offset and scaling (multiplied) Figure 8-5 Master value coupling offset and scaling (with increment offset) Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 332 M3: Coupled axes 8.3 Master value coupling - 840D sl only Reaction to Stop All leading value coupled following axes react to channel stop and MODE GROUP stop. Master value coupled following axes react to a stop due to end of program (M30, M02) if they have not been activated by static synchronized actions (IDS=...).
  • Page 333: Programming

    M3: Coupled axes 8.3 Master value coupling - 840D sl only Example: Activation from synchronized action Program code Comment SPOS=0 B=IC(0) ; switch spindle to axis operation. RELEASE(Y) ; Enable for synchronized action. ID=1 WHEN ($AA_IM[X]<-50) DO LEADON(B,X,2) ; Y is coupled to X via curve table No. 8.3.3 Programming Definition and activation...
  • Page 334 M3: Coupled axes 8.3 Master value coupling - 840D sl only Boundary conditions: ● No reference point is required to activate the coupling. ● A defined following axis cannot be traversed in the JOG mode (not even if the "Synchronized run fine"...
  • Page 335 M3: Coupled axes 8.3 Master value coupling - 840D sl only Meaning: Following axis as geometry-, channel- or machine axis name (X, Y, Z,...) <FA> Leading axis as geometry-, channel- or machine axis name (X, Y, Z,...) <LA> Software axis is also possible: MD30130 $MA_CTRLOUT_TYPE=0 (setpoint output type) Example: Program code...
  • Page 336 M3: Coupled axes 8.3 Master value coupling - 840D sl only System variables of the master value The following master value system variables can only be read from part program and from synchronous actions: System variable Meaning $AA_LEAD_V[ax] Velocity of the leading axis $AA_LEAD_P[ax] Position of the leading axis $AA_LEAD_P_TURN...
  • Page 337: Behavior In Automatic, Mda And Jog Modes

    M3: Coupled axes 8.3 Master value coupling - 840D sl only Note If the following axis is not enabled for travel, it is stopped and is no longer synchronous. 8.3.4 Behavior in AUTOMATIC, MDA and JOG modes Efficiency A master value coupling is active depending on the settings in the part program and in the following machine data: MD20110 $MC_RESET_MODE_MASK (definition of initial control settings after RESET / TP End)
  • Page 338 M3: Coupled axes 8.3 Master value coupling - 840D sl only MD20112 $MC_START_MODE_MASK (bit 13) (definition of initial control system settings with NC-START) ● MD20110 $MC_RESET_MODE_MASK=2001H && MD20112 $MC_START_MODE_MASK=0H → Master value coupling remains valid after RESET and START ● MD20110 $MC_RESET_MODE_MASK=2001H &&...
  • Page 339: Effectiveness Of Plc Interface Signals

    M3: Coupled axes 8.3 Master value coupling - 840D sl only 8.3.5 Effectiveness of PLC interface signals Leading axis When a coupled axis group is active, the interface signals (IS) of the leading axis are applied to the appropriate following axis via axis coupling. i.e.: ●...
  • Page 340: Supplementary Conditions

    M3: Coupled axes 8.4 Electronic gear (EG) The curve tables are not lost when the control system is switched off. These functions have no effect on cyclic machines because they are performed without operator actions. Nor does it make sense to perform automatic (re-)positioning via the NC with external master values.
  • Page 341: Preconditions

    M3: Coupled axes 8.4 Electronic gear (EG) Curve tables Non-linear relationships between lead and following axes can also be implemented using curve tables. Cascading Electronic gearboxes can be cascaded, i.e. the following axis of an electronic gearbox can be the leading axis for a subsequent electronic gearbox. Synchronous position An additional function for synchronizing the following axis allows a synchronous position to be selected:...
  • Page 342 M3: Coupled axes 8.4 Electronic gear (EG) All paths are referred to the basic co-ordinate system BCS. When an EG axis group is activated, it is possible to synchronize the leading axes and following axis in relation to a defined starting position. From the part program a gearbox group can be: ●...
  • Page 343 M3: Coupled axes 8.4 Electronic gear (EG) Coupling factor The coupling factor must be programmed for each leading axis in the group. It is defined by numerator/denominator. Coupling factor values numerator and denominator are entered per leading axis with the following activation calls: EGON EGONSYN...
  • Page 344 M3: Coupled axes 8.4 Electronic gear (EG) EG cascading The following axis of an EG can be the leading axis of another EG. For a sample configuration file, see Chapter "Examples". Figure 8-7 Block diagram of an electronic gearbox Synchronous positions To start up the EG axis group, an approach to defined positions for the following axis can first be requested.
  • Page 345 M3: Coupled axes 8.4 Electronic gear (EG) Activation response An electronic gearbox can be activated in two different ways: 1. On the basis of the axis positions that have been reached up to now in the course of processing the command to activate the EG axis group is issued without specifying the synchronizing positions for each individual axis.
  • Page 346 M3: Coupled axes 8.4 Electronic gear (EG) Synchronization abort with EGONSYN and EGONSYNE 1. The EGONSYN/EGONSYNE command is aborted under the following conditions and changed to an EGON command: ● RESET ● Axis switches to tracking The defined synchronization positions are ignored. Synchronous traverse monitoring still takes synchronized positions into account.
  • Page 347 M3: Coupled axes 8.4 Electronic gear (EG) Difference < .. TOL_COARSE As long as the synchronism difference is smaller than the following machine data, IS "Coarse synchronism" DB 31, ... DBX 98.1 is at the interface and IS "Fine synchronism" DB31, ... DBX99.4 is deleted: MD37200 $MN_COUPLE_POS_TOL_COARSE Difference >...
  • Page 348 M3: Coupled axes 8.4 Electronic gear (EG) IS "overlaid movement" DB31, ... DBX98.4: axis is overlaid, IS "Enable following axis override" DB31, ... DBX26.4 In the case of the commands EGON() and EGONSYNE(), the "Enable following axis override" signal must be present for the gear to synchronize to the specified synchronization position for the following axis.
  • Page 349: Definition Of An Eg Axis Group

    M3: Coupled axes 8.4 Electronic gear (EG) 5. "COARSE": Block change if "Coarse synchronism" is present 6. "IPOSTOP": Block change if "Setpoint synchronism" is present Note When programmed in activation calls EGON, EGONSYN, EGONSYNE, each of the above strings can be abbreviated to the first two characters. If no block change has been defined for the EG axis group and none is currently specified, "FINE"...
  • Page 350: Activating An Eg Axis Group

    M3: Coupled axes 8.4 Electronic gear (EG) No existing axis coupling may already be defined for the following axis. (If necessary, an existing axis must be deleted with EGDEL.) EGDEF triggers preprocessing stop with alarm. For an example of how to use the EG gearbox for gear hobbing, please see Chapter "Examples", "Electronic Gearbox for Gear Hobbing".
  • Page 351 M3: Coupled axes 8.4 Electronic gear (EG) 1. EGONSYN EGONSYN(FA, block change mode, SynPosFA, LA , SynPosLA , Z_LA , N_LA With: FA: Following axis Block change mode: "NOC": Block change takes place immediately "FINE": Block change is performed in "Fine synchronism" "COARSE": Block change is performed in "Coarse synchronism"...
  • Page 352 M3: Coupled axes 8.4 Electronic gear (EG) 2. EGONSYNE EGONSYNE(FA, block change mode, SynPosFA, Approach mode, LA ,SynPosLA , Z_LA N_LA with: "FA": Following axis Block change mode: "NOC": Block change takes place immediately "FINE": Block change is performed in "Fine synchronism" "COARSE": Block change is performed in "Coarse synchronism"...
  • Page 353 M3: Coupled axes 8.4 Electronic gear (EG) Tooth gap: 360*2/10 = 72 (degrees) Approach response with FA at standstill In this case, the time-optimized and path-optimized traversing modes are identical. The table below shows the target positions and traversed paths with direction marker (in brackets) for the particular approach modes: Programmed Position of the fol‐...
  • Page 354 M3: Coupled axes 8.4 Electronic gear (EG) Figure 8-9 Reaching the next tooth gap, FA path-optimized (top) vs. time-optimized (bottom) Sample notations EGONSYNE(A, "FINE", 110, "NTGT", B, 0, 2, 10) couple A to B, synchronized position A = 110, B = 0, coupling factor 2/10, approach mode = NTGT EGONSYNE(A, "FINE", 110, "DCT", B, 0, 2, 10) couple A to B, synchronized position A = 110, B = 0, coupling factor 2/10, approach mode =...
  • Page 355: Deactivating An Eg Axis Group

    M3: Coupled axes 8.4 Electronic gear (EG) 8.4.4 Deactivating an EG axis group Variant 1 There are different ways to deactivate an active EG axis grouping. EGOFS(following axis) The electronic gear is deactivated. The following axis is braked to a standstill. This call triggers a preprocessing stop.
  • Page 356: Interaction Between Rotation Feedrate (G95) And Electronic Gearbox

    M3: Coupled axes 8.4 Electronic gear (EG) This call triggers a preprocessing stop. 8.4.6 Interaction between rotation feedrate (G95) and electronic gearbox The FPR( ) part program command can be used to specify the following axis of an electronic gear as the axis, which determines the rotational feedrate. The following behavior is applicable in this case: ●...
  • Page 357: System Variables For Electronic Gearbox

    M3: Coupled axes 8.4 Electronic gear (EG) References (K1) Mode group, channel, program operation, reset response , Section "Block search type 5 SERUPRO" > "Lock program section for continue machining at the contour" 8.4.8 System variables for electronic gearbox Application The following system variables can be used in the part program to scan the current states of an EG axis group and to initiate appropriate reactions if necessary: Table 8-1...
  • Page 358: Examples

    M3: Coupled axes 8.4 Electronic gear (EG) Name Type Access Preprocessing Meaning, value Cond. Index stop part Sync part pro‐ Sync pro‐ act. gram act. gram $P_EG_BC[a] STRING Block change criterion for Axis name EG activation calls: EGON, a: Following axis EGONSYN: "NOC": immediately "FINE": Synchronoous tra‐...
  • Page 359 M3: Coupled axes 8.4 Electronic gear (EG) ● The radial axis (X) for infeeding the cutter to depth of tooth. ● The cutter swivel axis (A) for setting the hobbing cutter in relation to the workpiece as a function of cutter lead angle and angle of inclination of tooth. Figure 8-10 Definition of axes on a gear hobbing machine (example) The functional interrelationships on the gear hobbing machine are as follows:...
  • Page 360 M3: Coupled axes 8.4 Electronic gear (EG) The setpoint of the following axis is calculated cyclically with the following logic equation: * (z ) + v * (u ) + v * (u with: = Rotational speed of workpiece axis (C) = Rotational speed of milling spindle (B) = Number of gears of the hobbing machine = Number of teeth of the workpiece...
  • Page 361 M3: Coupled axes 8.4 Electronic gear (EG) Extract from part program: Program code Comment EGDEF(C,B,1,Z,1,Y,1) ; Definition of EG axis grouping with setpoint coupling (1) from B, Z, Y to C (following axis). EGON(C,"FINE",B,z0,z2,Z,udz,z2,Y,udy,z2) ; Activate coupling. … Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 362: Extended Example With Non-Linear Components

    M3: Coupled axes 8.4 Electronic gear (EG) 8.4.9.2 Extended example with non-linear components Introduction The following example extends the example (see "Figure 8-10 Definition of axes on a gear hobbing machine (example) (Page 359)") with the following: ● Machine error compensations which are not linearly dependent on the Z axis, and ●...
  • Page 363 M3: Coupled axes 8.4 Electronic gear (EG) The following section of a part program is intended to illustrate the general concept; supplementary curve tables and gear wheel/machine parameters are still to be added. Components to be added are marked with <...> . Stated parameters may also have to be modified, e.g.
  • Page 364 M3: Coupled axes 8.4 Electronic gear (EG) Program code Comment Y, <SynPosC99_Y>, ; Switch-on of leading axis Y R1 * π, 1, & Z, <SynPosC99_Z>, ; Switch-on of leading axis Z 10, 1) ; "&" character means: command continued in next line, no LF nor comment permissible in program ;...
  • Page 365 M3: Coupled axes 8.4 Electronic gear (EG) The system variables listed below are only used for explanatory purposes! ; ************** Gear X (G1) $AA_EG_TYPE[X, Z] = 1 ; Setpoint value coupling $AA_EG_NUMERA[X, Z] = 1 ; curve table No. = 1 $AA_EG_DENOM[X, Z] = 0 ;...
  • Page 366 M3: Coupled axes 8.4 Electronic gear (EG) $AA_EG_AX[2, C99] = B ; name of leading axis B $AA_EG_SYN[C99, Y] = <SynPosC99_Y> ; Synchronized position of leading axis Y $AA_EG_SYN[C99, Z] = <SynPosC99_Z> ; Synchronized position of leading axis Z $AA_EG_SYN[C99, B] = <SynPosC99_B> ;...
  • Page 367 M3: Coupled axes 8.4 Electronic gear (EG) ; *************** Axis 3, "Z" $MC_AXCONF_GEOAX_NAME_TAB[2] = "Z" $MC_AXCONF_CHANAX_NAME_TAB[2] = "Z" $MC_AXCONF_MACHAX_USED[2]=3 $MN_AXCONF_MACHAX_NAME_TAB[2] = "Z1" $MA_SPIND_ASSIGN_TO_MACHAX[AX3] = 0 $MA_IS_ROT_AX[AX3] = FALSE ; *************** Axis 4, "A" $MC_AXCONF_CHANAX_NAME_TAB[3] = "A" $MC_AXCONF_MACHAX_USED[3]=4 $MN_AXCONF_MACHAX_NAME_TAB[3] = "A1" $MA_SPIND_ASSIGN_TO_MACHAX[AX4]=0 $MA_IS_ROT_AX[AX4] = TRUE $MA_ROT_IS_MODULO[AX4] = TRUE...
  • Page 368: Generic Coupling

    M3: Coupled axes 8.5 Generic coupling Generic coupling 8.5.1 Brief description 8.5.1.1 Function Function "Generic Coupling" is a general coupling function, combining all coupling characteristics of existing coupling types (coupled motion, master value coupling, electronic gearbox and synchronous spindle). The function allows flexible programming: ●...
  • Page 369 M3: Coupled axes 8.5 Generic coupling This structure is based on the following considerations: ● Functional scope and required application knowledge increase from the basic version to the optional CP_EXPERT version. ● The number of required couplings (following axes, following spindles) and their properties are decisive in the selection of versions.
  • Page 370 M3: Coupled axes 8.5 Generic coupling Type A Type B Type C Type D Type E Coupling between a rotary axis and a linear axis From part program and synchronous actions Superimposition / speed difference permitted Cascading permitted BCS / BCS / BCS / Coordinate reference (default): CPFRS="BCS")
  • Page 371 M3: Coupled axes 8.5 Generic coupling Type A Type B Type C Type D Type E Non-linear coupling law (CPLCTID) permitted CP - free generic coupling Maximum number of free generic couplings with the following properties: Default (corresponds to CPSETTYPE="CP" Maximum number of master values From part program and synchronous actions Superimposition / speed difference permitted...
  • Page 372: Fundamentals

    M3: Coupled axes 8.5 Generic coupling 8.5.2 Fundamentals 8.5.2.1 Coupling module With the aid of a coupling module, the motion of one axis, (→ following axis), can be interpolated depending on other (→ leading) axes. Coupling rule The relationships between leading axis/values and a following axis are defined by a coupling rule (coupling factor or curve table).
  • Page 373: Keywords And Coupling Characteristics

    M3: Coupled axes 8.5 Generic coupling The following axis position results from the overlay (summation) of the dependent motion components (FA and FA ), which result from the individual coupling relationships to the DEP1 DEP2 leading axes, and of the independent motion component (FA ) of the following axis: = FA + FA...
  • Page 374 M3: Coupled axes 8.5 Generic coupling In order to be uniquely assigned, keywords are furnished with the prefix "CP", for Coupling). Depending on meaning and application position, a third letter is used: Keyword prefix Meaning Example Describes the characteristics of the entire coupling. CPON CPF* Describes the characteristics of the following axis (Fol‐...
  • Page 375 M3: Coupled axes 8.5 Generic coupling Keyword Coupling characteristics / meaning Default setting (CPSET‐ TYPE="CP") Behavior of the following axis at complete switch-off STOP CPFMOF Switch-off position of the following axis when switch‐ Not set CPFPOS + CPOF ing off Coupling response to RESET NONE CPMRESET...
  • Page 376: System Variables

    M3: Coupled axes 8.5 Generic coupling 8.5.2.3 System variables The current state of a coupling characteristic set with a keyword, can be read and written to with the relevant system variable. Note When writing in the part program, PREPROCESSING STOP is generated. Notation The names of system variables are normally derived from the relevant keywords and a corresponding prefix.
  • Page 377: Delete Coupling Module (Cpdel)

    M3: Coupled axes 8.5 Generic coupling Programming Syntax: CPDEF= (<following axis/spindle>) Designation: Coupling Definition Functionality: Definition of a coupling module The coupling is not activated. Following axis/ Type: AXIS spindle: Range of values: All defined axis and spindle names in the channel Example: Programming Comment...
  • Page 378: Defining Leading Axes (Cpldef Or Cpdef+Cpla)

    M3: Coupled axes 8.5 Generic coupling Example: Programming Comment CPDEL=(X2) ; Deletion of the coupling module with following axis X2. Boundary conditions ● The switch command CPDEL results in a preprocessing stop with active coupling. Exception: CPSETTYPE="COUP" does not result in a preprocessing stop. ●...
  • Page 379: Delete Leading Axes (Cpldel Or Cpdel+Cpla)

    M3: Coupled axes 8.5 Generic coupling Functionality: Definition of leading axis/spindle for following axis/spindle FAx. Leading axis/spin‐ Type: AXIS dle: Range of values: All defined axis and spindle names in the channel Example: Programming Comment CPDEF=(X2) CPLA[X2]=(X1) ; Definition of leading axis X1 for following axis Boundary conditions ●...
  • Page 380 M3: Coupled axes 8.5 Generic coupling Range of values: All defined axis and spindle names in the channel Example: Programming Comment CPLDEL[X2]=(X1) ; Deletion of leading axis X1 of the coupling to following axis X2. Programming with CPLA and CPDEL Syntax: CPLA[FAx]= (<leading axis/spindle>) Designation:...
  • Page 381: Switching Coupling On/Off

    M3: Coupled axes 8.5 Generic coupling 8.5.4 Switching coupling on/off 8.5.4.1 Switching on a coupling module (CPON) A defined coupling module is switched on with the switch command CPON. Coupling characteristics like coupling reference can be programmed together with the switch on command (see Section "Programming coupling characteristics (Page 384)").
  • Page 382: Switching On Leading Axes Of A Coupling Module (Cplon)

    M3: Coupled axes 8.5 Generic coupling Programming Syntax: CPOF= (<Following axis / spindle>) Designation: Coupling Off Functionality: Deactivate the coupling of the following axis to all defined leading axes. Following axis/ Type: AXIS spindle: Range of values: All defined axis and spindle names in the channel Example: Programming Comment...
  • Page 383: Switching Off Leading Axes Of A Coupling Module (Cplof)

    M3: Coupled axes 8.5 Generic coupling Range of values: All defined axis and spindle names in the channel Example: Programming Comment CPLON[X2]=(X1) ; The coupling of leading axis X1 to following axis X2 is ac- tivated. Boundary conditions CPON can be programmed in synchronous actions. 8.5.4.4 Switching off leading axes of a coupling module (CPLOF) CPLOF deactivates the coupling of a leading axis to a following axis.
  • Page 384: Implicit Creation And Deletion Of Coupling Modules

    M3: Coupled axes 8.5 Generic coupling 8.5.4.5 Implicit creation and deletion of coupling modules Switch-on commands may also be used to create coupling modules (without prior definition with CPDEF). Example Programming Comment CPON=(X2) CPLA[X2]=(X1) ; Creates a coupling module for following axis X2 with leading axis X1 and activates the coupling module.
  • Page 385 M3: Coupled axes 8.5 Generic coupling Numerator of the coupling factor Syntax: CPLNUM[FAx,LAx]= <value> Designation: Coupling Lead Numerator Functionality: Defines the numerator of the coupling factor for the coupling rule of the following axis/spindle FAx to the leading axis/spindle LAx. Value: Type: REAL...
  • Page 386: Coupling Relationship (Cplsetval)

    M3: Coupled axes 8.5 Generic coupling Programming: Curve table When programming a table number, a previously activated non-linear coupling relationship (coupling factor) is deactivated. The leading axis specific coupling component for the leading value of the leading axis is calculated using the specified curve table. Syntax: CPLCTID[FAx,LAx]= <value>...
  • Page 387: Co-Ordinate Reference (Cpfrs)

    M3: Coupled axes 8.5 Generic coupling The following couplings can be programmed accordingly: ● Setpoint value coupling ● Speed coupling ● Actual value coupling Programming Syntax: CPLSETVAL[FAx,LAx]= <value> Identifiers: Coupling Lead Set Value Functionality: Defines tapping of the leading axis/spindle LAx and the reaction point on the following axis/spindle FAx.
  • Page 388: Block Change Behavior (Cpbc)

    M3: Coupled axes 8.5 Generic coupling Programming Syntax: CPFRS[FAx]= (<co-ordinate reference>) Identifiers: Coupling Following Relation System Functionality: Defines the co-ordinate reference system for the coupling module of the following axis/spindle FAx. Co-ordinate refer‐ Type: STRING ence: Range of values: ”BCS” Basis Co-ordinate System Basic Coordinate System ”MCS”...
  • Page 389 M3: Coupled axes 8.5 Generic coupling Programming with PCBC Syntax: CPBC[FAx]= "<block change criterion>" Identifiers: Coupling Block Change Criterium Functionality: Defines block change criterion with active coupling. Block change criteri‐ Type: STRING Range of values: "NOC" Block change is performed irrespective of the coupling status.
  • Page 390: Synchronized Position Of The Following Axis When Switching On (Cpfpos+Cpon)

    M3: Coupled axes 8.5 Generic coupling Type: STRING Range of values: "NOC" Block change is performed irrespective of the coupling status. "IPOSTOP" Block change is performed with setpoint synchronism. "COARSE" Block change is performed with actual value synchron‐ ism “coarse”. "FINE"...
  • Page 391: Synchronized Position Of The Leading Axis When Switching On (Cplpos)

    M3: Coupled axes 8.5 Generic coupling Example: Programming Comment CPON=X2 CPFPOS[X2]=100 ; Activation of coupling to following axis X2. 100 is taken as synchronized position of following axis X2. Supplementary conditions ● CPFPOS is only effective as synchronized position with the switch-on command CPON/ CPLON.
  • Page 392: Synchronization Mode (Cpfmson)

    M3: Coupled axes 8.5 Generic coupling Functionality: Defines the synchronized position of the following axis at switch on. Only AC is possible in the position specification. Value: Type: REAL Range of values: All position within the traverse range boundaries Example: Programming Comment CPLPOS[X2,X1]=200...
  • Page 393 M3: Coupled axes 8.5 Generic coupling Identifiers: Coupling Following Mode Strategy On Functionality: Determines the synchronization mode during coupling. Synchronization Type: STRING mode: Range of values: "CFAST" Closed Coupling Fast The coupling is closed time-op‐ timized. "CCOARSE" Closed If Gab Coarse The coupling is only closed when the following axis posi‐...
  • Page 394: Behavior Of The Following Axis At Switch-On (Cpfmon)

    M3: Coupled axes 8.5 Generic coupling "DCP" Direct Co-ordinate For rotary axes only! Path Optimized The rotary axis traverses to the programmed synchronized po‐ sition in path-optimized fash‐ ion. Synchronization is effec‐ ted immediately. Default value: "CFAST" Example: Programming Comment CPFMSON[X2]="CFAST"...
  • Page 395: Behavior Of The Following Axis At Switch-Off (Cpfmof)

    M3: Coupled axes 8.5 Generic coupling "ADD" Additional For spindles only! The motion components of the cou‐ pling operate in addition to the cur‐ rently overlaid motion, i.e. the current motion of the following axis/spindle is retained as overlaid motion. Default value: "STOP"...
  • Page 396: Position Of The Following Axis When Switching Off (Cpfpos+Cpof)

    M3: Coupled axes 8.5 Generic coupling Example: Programming Comment CPFMOF[S2]="CONT" ; The following spindle S2 continues to traverse at the speed that was applied at the instant of deactivation. 8.5.5.10 Position of the following axis when switching off (CPFPOS+CPOF) When switching off a coupling (CPOF) traversing to a certain position can be requested for the following axis.
  • Page 397 M3: Coupled axes 8.5 Generic coupling Programming Syntax: CPMRESET[FAx]= "<Reset behavior>" Identifiers: Coupling Mode RESET Functionality: Defines the behavior of a coupling at RESET. Reset response: Type: STRING Range of values: "NONE" The current state of the coupling is retained. "ON"...
  • Page 398: Condition At Parts Program Start (Cpmstart)

    M3: Coupled axes 8.5 Generic coupling Constraints ● The coupling characteristics set with CPMRESET is retained until the coupling module is deleted with (CPDEL). ● For the coupling type (CPSETTYPE="TRAIL", "LEAD", "EG" or "COUP") the response is defined by the following machine data during RESET: MD20110 $MC_RESET_MODE_MASK (definition of initial control system settings after RESET/TP-End) →...
  • Page 399: Status During Part Program Start In Search Run Via Program Test (Cpmprt)

    M3: Coupled axes 8.5 Generic coupling Constraints ● The coupling characteristics set with CPMSTART are retained until the coupling module is deleted with (CPDEL). ● For the set coupling type (CPSETTYPE="TRAIL", "LEAD", "EG" or "COUP"), the response is defined by the following machine data during part program start: MD20112 $MC_START_MODE_MASK (Definition of the control default settings in case of NC START) →...
  • Page 400: Offset / Scaling (Cplintr, Cplinsc, Cplouttr, Cploutsc)

    M3: Coupled axes 8.5 Generic coupling Example: Programming Comment CPMPRT[X2]="ON" ; At part program start during search run via program test, coupling to following axis X2 is switched on. Constraints ● The coupling characteristics set with CPMPRT is retained until the coupling module is deleted with (CPDEL).
  • Page 401 M3: Coupled axes 8.5 Generic coupling Programming Offset of the input value Syntax: CPLINTR[FAx,LAx]= <value> Designation: Coupling Lead In Translation Displacement Functionality: Defines the offset value for the input value of the LAx leading axis. Value: Type: REAL Default value: Example: Programming Comment...
  • Page 402 M3: Coupled axes 8.5 Generic coupling Default value: Example: Programming Comment CPLOUTTR[X2,X1]=100 ; The output value of the coupling of the following axis X2 with leading axis X1 is displaced by the value 100 in the positive direction. Scaling of the output value Syntax: CPLOUTSC[FAx,LAx]= <value>...
  • Page 403: Synchronism Monitoring Stage 1 (Cpsyncop, Cpsynfip, Cpsyncov, Cpsynfiv)

    M3: Coupled axes 8.5 Generic coupling 8.5.5.15 Synchronism monitoring stage 1 (CPSYNCOP, CPSYNFIP, CPSYNCOV, CPSYNFIV) Synchronism monitoring stage 1 In each interpolator clock cycle, the synchronous operation of the coupling group is monitored – both on the setpoint and actual value sides. The synchronous operation monitoring responds as soon as the synchronous operation difference (the difference between the setpoint or actual value of the following axis and the value calculated from the setpoints or actual values of the leading axes according to the coupling rule) reaches one of the following programmed...
  • Page 404 M3: Coupled axes 8.5 Generic coupling Status of the coupling during synchronous operation State Description Not synchronized Provided the synchronous operation difference is greater than the threshold value for position "coarse" synchronous operation or "coarse" speed synchronous operation, the coupled group is des‐ ignated as non-synchronous.
  • Page 405 M3: Coupled axes 8.5 Generic coupling Configuration The threshold values for the first stage of the synchronous operation monitoring will be adjusted: ● For setpoint / actual value coupling in the machine data: – MD37200 $MA_COUPLE_POS_TOL_COARSE (threshold value for "coarse synchronism") –...
  • Page 406 M3: Coupled axes 8.5 Generic coupling Threshold value of "Coarse" speed synchronous operation Syntax: CPSYNCOV[FAx]= <value> Designation: Coupling Synchronous Difference Coarse Velocity Functionality: Defines the threshold value for the "Coarse'' speed synchronous oper‐ ation. Value: Type: REAL The default value corresponds to the setting in the machine data: MD37220 $MA_COUPLE_VELO_TOL_COARSE [FAx] Threshold value of "Fine"...
  • Page 407: Synchronous Operation Monitoring Stage 2 (Cpsyncop2, Cpsynfip2)

    M3: Coupled axes 8.5 Generic coupling Supplementary conditions ● When considering the synchronous operation difference, an active coupling cascade is not taken into account. This means: if in the considered coupling module, the leading axis is a following axis in another coupling module, the current actual or setpoint position is still used as input variable for the calculation of the synchronous operation difference.
  • Page 408 M3: Coupled axes 8.5 Generic coupling MD37212 $MA_COUPLE_POS_TOL_FINE_2 (second threshold value for "fine synchronous operation") Note If the appropriate threshold value = 0, the associated monitoring is inactive. This is also the default value so that the compatibility with older software versions is retained. Programming CP keywords can also be used to program the threshold values for the second stage of the synchronous operation monitoring:...
  • Page 409 M3: Coupled axes 8.5 Generic coupling Sequence Starting The second stage of the synchronous operation monitoring function starts with active coupling as soon as the following conditions are fulfilled: ● The setpoint synchronous operation is reached: DB31, ... DBX99.4 (synchronization running) = 0 ●...
  • Page 410 M3: Coupled axes 8.5 Generic coupling ● For coupled block changes (CPLNUM, CPLDEN, CPLCTID) in synchronized actions ● Resetting the setpoint synchronous operation because of missing enable signals for the following spindle (emergency stop, alarm responses) The DB31, ... DBX103.4/5 signals are reset when the monitoring is ended. Boundary conditions Exclusion conditions No monitoring is performed in the following cases:...
  • Page 411: Reaction To Stop Signals And Commands (Cpmbrake)

    M3: Coupled axes 8.5 Generic coupling 8.5.5.17 Reaction to stop signals and commands (CPMBRAKE) The response of the following axis to certain stop signals and commands can be defined with the CP keyword CPMBRAKE. Programming Syntax: CPMBRAKE[FAx]= <value> Designation: Coupling Mode Brake Functionality: CPMBRAKE is a bit-coded CP keyword that defines the braking behavior of the following axis FAx for the following events:...
  • Page 412: Response To Certain Nc/Plc Interface Signals (Cpmvdi)

    M3: Coupled axes 8.5 Generic coupling Example 2: Programming Comment CPDEF=(S2) CPLA[S2]=(S1) Definition of a spindle coupling: Leading spindle S1 with following spindle S2 CPON=(S2) CPMBRAKE[S2]=1 Activation of the coupling with following spindle S2. NST "feed stop / spindle stop" or "CP SW limit stop"...
  • Page 413 M3: Coupled axes 8.5 Generic coupling Meaning Reserved. Reserved. Reserved. The effect of NC/PLC interface signal DB31, ... DBX1.3 (axis/spindle disable) on the following axis/spindle can be set via bit 3: Bit 3 = 0 DB31, ... DBX1.3 has no effect on the following axis/ spindle.
  • Page 414 M3: Coupled axes 8.5 Generic coupling Note: When bit 5 is set, the program test state still has an effect on the following axis/spindle, even if the leading axes/spindles have a differ‐ ent state. Bit 6 is used to define the enable for the dependent motion compo‐ nents when the NC/PLC interface signal DB21, …...
  • Page 415 M3: Coupled axes 8.5 Generic coupling A/S disable A/S disable A/S disable CPMVDI CPMVDI Meaning Total Bit 3/5 Bit 4/6 for FA Real move‐ ment, FA spin‐ DEP1 dle disable has DEP2 no effect. Simulated movement, FA DEP1 spindle disable DEP2 has an effect.
  • Page 416: Alarm Suppression (Cpmalarm)

    M3: Coupled axes 8.5 Generic coupling Note The states in the columns for leading axes 1 and 2 also apply if there are several leading axes/ spindles, which have the same state with reference to the axis/spindle disable. 8.5.5.19 Alarm suppression (CPMALARM) The CP keyword CPMALARM can be used to suppress coupling-related alarms.
  • Page 417: Coupling Cascading

    M3: Coupled axes 8.5 Generic coupling Example Program code Comment CPMALARM[X2]='H300' ; The 22025 and 22026 alarms are suppressed for the cou- pling of the X2 following axis. 8.5.6 Coupling cascading Coupling cascades The coupling modules can be connected in series. The following axis/spindle of a coupling module then becomes the leading axis/spindle of another coupling module.
  • Page 418: Compatibility

    M3: Coupled axes 8.5 Generic coupling 8.5.7 Compatibility 8.5.7.1 Adaptive cycles Adaptive cycles The provision of adaptive cycles as fixed component of the NCK software ensures a syntactic and functional compatibility to coupling calls of existing coupling types (coupled motion, master value coupling, electronic gearbox and synchronous spindle).
  • Page 419: Coupling Types (Cpsettype)

    M3: Coupled axes 8.5 Generic coupling User specific adaptive cycles If necessary (functional completion) the user can copy an adaptive cycle to the directory "CMA" or "CUS" and apply changes there. When reading adaptive cycles, the sequence CUS → CMA →...
  • Page 420 M3: Coupled axes 8.5 Generic coupling Example: Programming Comment CPLON[X2]=(X1) CPSETTYPE[X2]="LEAD" ; Creates a coupling module for following axis X2 with leading axis X1 and acti- vates the coupling module. Coupling properties are set such that they cor- respond to the existing master value coupling type.
  • Page 421 M3: Coupled axes 8.5 Generic coupling Keyword Coupling type Default Coupled motion Master value cou‐ Electronic gear Synchronous spin‐ (CP) (TRAIL) pling (EG) (LEAD) (COUP) CPMRESET NONE MD20110 MD20110 MD20110 MD20110 CPMSTART NONE MD20112 MD20112 MD20112 MD20112 CPMPRT NONE MD20112 / MD20112 / MD20112 / MD20112 /...
  • Page 422 M3: Coupled axes 8.5 Generic coupling Additional properties Value ranges or availability of additional properties of a set coupling type (CPSETTYPE) can be found in the following table: Default Coupled motion Master value cou‐ Electronic gear Synchronous spindle (CP) (TRAIL) pling ( (EG) (COUP)
  • Page 423 M3: Coupled axes 8.5 Generic coupling CPSETTYPE= TRAIL LEAD COUP CPLDEF CPLDEL CPON Alarm 16686 Alarm 16686 CPLON CPOF Alarm 16686 Alarm 16686 CPLOF CPRES Alarm 16686 Alarm 16686 Alarm 16686 CPLNUM Alarm 16686 Alarm 16686 CPLDEN Alarm 16686 Alarm 16686 CPLCTID Alarm 16686 Alarm 16686...
  • Page 424: Projected Coupling (Cpres)

    M3: Coupled axes 8.5 Generic coupling 8.5.7.3 Projected coupling (CPRES) If the coupling type "Synchronous spindle" is set, (see CPSETTYPE), the coupling properties contained in machine data can be activated instead of the programmed coupling properties. References: Functions Manual Extension Functions; Synchronous spindles (S3); Chapter "Programming of synchronous spindle couplings"...
  • Page 425: Cross-Channel Coupling, Axis Replacement

    M3: Coupled axes 8.5 Generic coupling 8.5.8 Cross-channel coupling, axis replacement The following and leading axes must be known to the calling channel. Following axis The following axis is requested for replacement in the channel when programming a CP keyword in the part program, depending on the axis replacement projection (MD30552) with the language command GETD.
  • Page 426 M3: Coupled axes 8.5 Generic coupling Modulo reduced rotary axes as leading axes With modulo reduced rotary axes as leading axes, the input variable is not reduced during the reduction of the leading axis. The non-reduced position is still taken as the input variable, i.e. the traversed distance is considered.
  • Page 427: Behavior During Power On

    M3: Coupled axes 8.5 Generic coupling (...) Position indication for X, A Figure 8-12 Example: Modulo reduced rotary axis to linear axis 8.5.10 Behavior during POWER ON, ... Power on No coupling is active at power ON. Coupling modules are not available. RESET The behavior on RESET can be set separately for each coupling module (see CPMRESET).
  • Page 428: Cp Sw Limit Monitoring

    M3: Coupled axes 8.5 Generic coupling Mode change The coupling remains active during a mode change. The coupling is suppressed (not deselected!) only in JOG-REF mode when referencing a following axis. Reference point approach G74 of the following axis is not possible with an active coupling. An alarm is output. If the JOG-REF mode is selected and the following axis is traversed, the coupling is suppressed.
  • Page 429 M3: Coupled axes 8.5 Generic coupling Monitoring and setting the brake The "CP-SW limit monitoring" function checks, in every IPO cycle, as to whether the movement of the following axis/spindle can be enabled for the following IPO cycle, so that the axis can always stop in plenty of time before the software limit switch.
  • Page 430: Parameterization

    M3: Coupled axes 8.5 Generic coupling This means that up to the "final" stopping position, the situation can be somewhat relieved by moving the following axis in the opposite direction. Note For the "CP-SW limit monitoring", coupling involves maintaining synchronous operation. As a consequence, it cannot be guaranteed that the axis stops at precisely the correct position, if the coupling rule is to be maintained.
  • Page 431: Programming

    M3: Coupled axes 8.5 Generic coupling 8.5.11.3 Programming Transferring the brake to the leading axes For generic couplings, type "freely programmable" (CPSETTYPE[FAx] = "CP"), by programming the coupling property CPMBRAKE (see "Reaction to stop signals and commands (CPMBRAKE) (Page 411)") it can be set as to whether the brake of the following axis, initiated using the "CP-SW limit monitoring"...
  • Page 432: Examples

    M3: Coupled axes 8.5 Generic coupling The basic motion is path motion with extremely low velocity in the positive direction. A larger DRF correction motion applies to this in the opposite direction. This results in a travel command in the negative direction. If these movements are now stopped using a CP-SW limit stop, then for standard machine data, this means that DRF motion is canceled, and only path motion is continued after the stop is withdrawn.
  • Page 433: Disturbance Characteristic

    M3: Coupled axes 8.5 Generic coupling EGDEF(Y,X,1) ; Definition of an EG axis group with setpoint coupling from X to Y (following axis). EGON( Y,"FINE",X,1,2) ; Activating coupling. 8.5.12 Disturbance characteristic 8.5.12.1 Rapid stop Function The rapid stop stops the axis / spindle without ramp, i.e. the velocity setpoint value is specified as zero.
  • Page 434: Tracking The Deviation From Synchronism

    M3: Coupled axes 8.5 Generic coupling The start of a rapid stop for a leading axis/spindle is detected across NCUs. Note A simultaneous rapid stop of the leading and following spindle is executed in the synchronized spindle coupling type (CPSETTYPE="COUP") during a servo alarm. 8.5.13 Tracking the deviation from synchronism 8.5.13.1...
  • Page 435: Measuring The Deviation From Synchronism

    M3: Coupled axes 8.5 Generic coupling Versions There are two different options for determining the deviation from synchronism: 1. The deviation from synchronous operation is determined by the NCK (see "Measuring the deviation from synchronism (Page 435)"). 2. The deviation value is already known and entered by the user directly (see "Entering the deviation from synchronism directly (Page 438)").
  • Page 436 M3: Coupled axes 8.5 Generic coupling Requirements The following requirements must be met to enable the controller to calculate the correction value: ● Requirements if the set coupling type is "synchronous spindle" (CPSETTYPE="COUP"): – The coupling has precisely one leading spindle (requirement is met if CPSETTYPE="COUP").
  • Page 437 M3: Coupled axes 8.5 Generic coupling The signal only has an effect on the following spindle. Note In the following cases, signal DB31, ... DBX31.6 (track synchronism) is ignored: ● Axis/spindle disable is active (DB31, ... DBX1.3 = 1). ● Program test is selected. ●...
  • Page 438: Entering The Deviation From Synchronism Directly

    M3: Coupled axes 8.5 Generic coupling $AA_COUP_CORR[S<n>] (following spindle: correction value for synchronous spindle coupling) Note You must ensure that the velocity of the leading and following axes is kept as constant as possible and that no acceleration jump occurs for the duration of the measurement. Example When the coupling of the synchronous spindle [S2] is activated, a position offset of 77 degrees is also programmed:...
  • Page 439: Diagnostics For Synchronism Correction

    M3: Coupled axes 8.5 Generic coupling The correction value is incorporated into the setpoint value calculation for the following spindle, in the coupling module. Resetting the setpoint by the coupling offset relieves the tension between the leading and following spindles. The synchronism signals are produced by comparing the actual values with the corrected setpoints.
  • Page 440: Resetting Synchronism Correction

    M3: Coupled axes 8.5 Generic coupling 8.5.13.6 Resetting synchronism correction Versions Synchronism correction can be reset in the following ways: ● Writing value "0" to variable $AA_COUP_CORR[S<n>]. Synchronism correction is suppressed via a ramp with reduced accelerating power (just as when a correction value is implemented).
  • Page 441: Limitations And Constraints

    M3: Coupled axes 8.5 Generic coupling Figure 8-13 Time diagram for synchronizing and resetting synchronism correction Note If the correction path has not been traversed in full and the NC/PLC interface signal DB31, ... DBX31.7 (reset synchronism correction) has not been reset, writing to variable $AA_COUP_CORR[S<n>] will not have any effect.
  • Page 442 M3: Coupled axes 8.5 Generic coupling Writing variable $AA_COUP_CORR System variable $AA_COUP_CORR is only written from the part program or synchronized actions when a generic machine coupling has been activated for the corresponding axis/ spindle at least once. Correction value If the correction value $AA_COUP_CORR is being written via a part program/synchronized action, as well as being determined due to the "track the deviation from synchronism"...
  • Page 443: Examples

    M3: Coupled axes 8.5 Generic coupling 8.5.14 Examples 8.5.14.1 Programming examples Direct switch on/off with one leading axis A coupling module is created and activated with following axis X2 and leading axis X1. The coupling factor is 2. CPON=(X2) CPLA[X2]=(X1) CPLNUM[X2,X1]=2 CPOF=(X2) ;...
  • Page 444: Adapt Adaptive Cycle

    M3: Coupled axes 8.5 Generic coupling N30 CPOF=(X2) ; All leading axes are deactivated. N40 CPLON[X2]=(X1) ; Leading axis X1 is activated, only this axis supplies a coupling component. Leading axes Z and A remain deactivated. N50 CPLON[X2]=(A) ; Leading axis X1 remains active, leading axis A is deactivated, X1 and A contrib- ute coupling components (→...
  • Page 445: Dynamic Response Of Following Axis

    M3: Coupled axes 8.6 Dynamic response of following axis Figure 8-14 Cycle 700 after adaption. Changes are indicated by a colored bar. Dynamic response of following axis 8.6.1 Parameterized dynamic limits The dynamics of the following axis is limited with the following machine data values: MD32000 $MA_MAX_AX_VELO (maximum axis velocity) MD32300 $MA_MAX_AX_ACCEL (Maximum axis acceleration) Special functions...
  • Page 446: Programmed Dynamic Limits

    M3: Coupled axes 8.6 Dynamic response of following axis 8.6.2 Programmed dynamic limits 8.6.2.1 Programming (VELOLIMA, ACCLIMA) Reducing or increasing dynamics limits The dynamic limits of the following axis (FA) specified through MD32000 and MD32300 can be reduced or increased from the part program: Command Meaning Reducing or increasing the maximum Axis velocity...
  • Page 447 M3: Coupled axes 8.6 Dynamic response of following axis Synchronization between following and leading axes The acceleration characteristics set and the dynamics offsets set change the duration for synchronization between following and leading axes during acceleration operations as follows: Dynamic offset Activation Dynamic reduction Prolongs the synchronism difference.
  • Page 448: Examples

    M3: Coupled axes 8.6 Dynamic response of following axis MD22410 $MC_F_VALUES_ACTIVE_AFTER_RESET (F Function is active even after RESET) Value Meaning The values of VELOLIMA[FA] and ACCLIMA[FA] are set to 100% after RESET. The last programmed values of VELOLIMA[FA] and ACCLIMA[FA] are also active after RESET.
  • Page 449: System Variables

    M3: Coupled axes 8.7 General supplementary conditions Master value coupling with synchronized action Axis 4 is coupled to X via a master value coupling. The acceleration response is limited to position 80% by static synchronized action 2 from position 100. N120 IDS=2 WHENEVER $AA_IM[AX4] >...
  • Page 450: Data Lists

    M3: Coupled axes 8.8 Data lists Data lists 8.8.1 Machine data 8.8.1.1 NC-specific machine data Number Identifier: $MN_ Description 11410 SUPPRESS_ALARM_MASK Screen form for suppressing special alarm outputs 11415 SUPPRESS_ALARM_MASK_2 Suppress alarm outputs 11660 NUM_EG Number of possible electronic gears 11750 NCK_LEAD_FUNCTION_MASK Functions for master value coupling...
  • Page 451: Axis/Spindlespecific Machine Data

    M3: Coupled axes 8.8 Data lists 8.8.1.3 Axis/spindlespecific machine data Number Identifier: $MA_ Description 30130 CTRLOUT_TYPE Setpoint output type 30132 IS_VIRTUAL_AX Axis is virtual axis 30455 MISC_FUNCTION_MASK Axis functions 35040 SPIND_ACTIVE_AFTER_RESET Own spindle RESET 37160 LEAD_FUNCTION_MASK Functions for master value coupling 37200 COUPLE_POS_TOL_COARSE Threshold value for "Coarse synchronous operation"...
  • Page 452 M3: Coupled axes 8.8 Data lists Identifier Meaning $AA_EG_DENOM Numerator of the coupling factor for leading axis b $AA_EG_NUMERA Numerator of the coupling factor for leading axis b $AA_EG_NUMLA Number of leading axes defined with EGDEF $AA_EG_SYN Synchronized position of leading axis b $AA_EG_SYNFA Synchronous position of following axis a $AA_EG_TYPE...
  • Page 453 M3: Coupled axes 8.8 Data lists Identifier Meaning $AA_CPLSETVAL Coupling reference of the leading axis $AA_CPLSTATE State of the coupling $AA_CPSYNCOP Threshold value of position synchronism "Coarse" (main run) $AA_CPSYNCOV Threshold value of velocity synchronism "Coarse" (main run) $AA_CPSYNFIP Threshold value of position synchronism "Fine" (main run) $AA_CPSYNFIV Threshold value of velocity synchronism "Fine"...
  • Page 454: Signals

    Preprocessing acceleration correction set with ACCLIMA $PA_VELOLIMA Preprocessing speed correction set with VELOLIMA 8.8.4 Signals 8.8.4.1 Signals to axis/spindle Signal name SINUMERIK 840D sl SINUMERIK 828D Feedrate override DB31, ... DBX0.0-7 DB380x.DBB0 Axis disable DB31, ... DBX1.3 DB380x.DBX1.3 Controller enable DB31, ...
  • Page 455: Signals From Axis/Spindle

    M3: Coupled axes 8.8 Data lists Signal name SINUMERIK 840D sl SINUMERIK 828D Disable synchronization DB31, ... DBX31.5 DB380x.DBX5007.5 Track synchronism DB31, ... DBX31.6 DB380x.DBX5007.6 Reset synchronism correction DB31, ... DBX31.7 DB380x.DBX5007.7 8.8.4.2 Signals from axis/spindle Signal name SINUMERIK 840D sl...
  • Page 456 M3: Coupled axes 8.8 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 457: R3: Extended Stop And Retract

    R3: Extended stop and retract Brief description The extended stop and retract function - subsequently called ESR - offers the possibility of flexibly responding when a fault situation occurs as a function of the process: ● Extended stop Assuming that the specific fault situation permits it, all of the axes, enabled for extended stopping, are stopped in an orderly fashion.
  • Page 458 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only using the SINAMICS S120 "Vdc control" drive function. Even when the power fails, their kinetic energy is used to maintain the DC link voltage in order to permit NC controlled retraction motion. Note Detailed information on the SINAMICS S120 drive function "Vdc control"...
  • Page 459: Drive-Independent Reactions

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only 9.2.2 Drive-independent reactions Generator operation Generator operation is a drive function. Using the "Vdc control" function, the SINAMICS S120 drive unit can monitor the DC link group for undervoltage. When an adjustable voltage value is fallen below, then the drive intended for the purpose is switched into generator operation.
  • Page 460: Nc-Controlled Extended Stop

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only a resistor module. The operating range of the resistor module (shown highlighted in the diagram) lies below the critical voltage level. Note The pulse power of the resistor module is greater than the infeed power. Monitoring the intermediate circuit minimum voltage limit The DC link voltage can be monitored against a limit value that can be parameterized in the drive:...
  • Page 461 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Time sequence After initiating ESR, the axis continues to traverse with the actual speed. After a delay time has expired (..._TIME1), the axis is braked, interpolating. The set time (..._TIME 2) is the maximum time available for interpolatory braking. After this time has expired, a setpoint of 0 is output for the axis, and the follow-up mode is activated.
  • Page 462: Retract

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Behavior of path axes If NC-controlled extended stopping (MD37500 $MA_ESR_REACTION = 22) is parameterized for a path axis, then the corresponding behavior is also transferred to all other path axes of the channel.
  • Page 463 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only The extended retraction (i.e. LIFTFAST/LFPOS initiated through $AC_ESR_TRIGGER ) cannot be interrupted and can only be terminated prematurely using EMERGENCY OFF. Speed and acceleration limits for the axes involved in the retraction are monitored during the retraction motion.
  • Page 464 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Rapid lift without enabled retraction movement If, for an axis, ESR_REACTION=21 is configured and enabled with $AC_ESR_ENABLE=1, but e.g. no retraction motion is enabled with POLFMASK, then, for this axis, ESR_REACTION=22 is active (extended stopping).
  • Page 465 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Value Description Response to the channel-spec. NC/PLC interface signal DB21 DBB6.0 (feed disable) Retraction motion stop for feed disable in the channel No retraction motion stop for feed disable in the channel Note Influence of the interface signals ●...
  • Page 466 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only For the same axis names of a channel and machine axis, retraction motion is executed in the workpiece coordinate system. Enable for retraction without geometrical interrelationship POLFMASK The program command POLFMASK allows axes to be selected, which, when rapid lift is activated, traverse independently to their position defined with POLF.
  • Page 467: Trigger Sources

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Deactivate rapid lift POLFMASK or POLFMLIN without specifying an axis deactivates rapid lift for all axes activated in the enable call. Interactions, POLFMASK/POLFMLIN The last data entered for a specific axis in one of the two instructions applies. For example: Program code Description N200 POLFLIN(X,Y,Z)
  • Page 468: Logic Gating Functions: Source And Reaction Linking

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Axial trigger sources ● $VA_SYNCDIFF[<following axis>] (synchronism difference, on the actual value side) ● $AA_ESR_STAT (axial return signal word: ESR status) Alarm responses When ESR is active, the alarm responses NOREADY and STOPBYALARM are delayed by one IPO cycle.
  • Page 469: Configuring Aids For Esr

    For the following modules, when the line voltage fails, the power supply must be maintained using suitable measures at least until it is ensured that the axes have come to a stop: ● SINUMERIK 840D sl NCU 7x0 ● SINUMERIK NCU I/O modules ●...
  • Page 470 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only This energy is available for a minimum time of t = E / P * η Energy in Wattseconds [Ws] Buffer time in milliseconds [ms] Power in kilowatt [kW] η: Efficiency of the drive unit Example...
  • Page 471 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Table 9-1 SINAMICS infeeds (ALMs): Nominal and minimum buffer times Max. capacitance Energy content Energy content E Backup time t Buffer time t Pmax Pmin [μF] for C [Ws] for C [Ws]...
  • Page 472: Control System Response

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only When using infeeds with a high power rating (55, 80, 120 kW) for generator operation, it is recommended to use a dedicated axis with additional flywheel effect. For this axis, after the drive has accelerated to the rated speed, energy is only required to compensate for friction losses.
  • Page 473 R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only ● POLF[X]=<retraction position> LFPOS ● POLF[Y]=<retraction position> LFPOS Initial situation Response in the case of ESR Retraction motion in X and Y. ● $AA_ESR_ ENABLE[X] = 1 ● $AA_ESR_ ENABLE[Y] = 1 ●...
  • Page 474: Power Off/Power On

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only Initial situation Response in the case of ESR B is not enabled for ESR. As B is not a path axis, B is immediately stopped ● $AA_ESR_ ENABLE[X] = 1 with a rapid stop.
  • Page 475: Part Program Start, Nc Start

    R3: Extended stop and retract 9.2 Control-managed ESR - 840D sl only 9.2.10.4 Part program start, NC start In order that a defined initial state is available when a part program starts, the programmed absolute retraction positions and the enable signals of the retraction axes are deleted when the part program starts.
  • Page 476: Esr Executed Autonomously In The Drive

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive ESR executed autonomously in the drive 9.3.1 Fundamentals Function Drive-autonomous extended stop and retract (ESR) enables the fast separation of workpiece and tool independent of the higher-level control (NC). For this purpose, the following axial functions can be configured in the drive: ●...
  • Page 477: Configuring Stopping In The Drive

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive ● Control unit drive object, PROFIdrive telegram: 390, 391 ● The 24 V power supply for the modules is buffered via CSM or UPS 9.3.2 Configuring stopping in the drive Drive integrated shutdown is configured via the following drive parameters: Parameter Description...
  • Page 478: Configuring Retraction In The Drive

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Feedback signal The stop status is returned to the control (see Section "Feedback of the ESR status (Page 481)"). References For a detailed description of drive parameters, refer to: SINAMICS S120/S150 Parameter Manual 9.3.3 Configuring retraction in the drive...
  • Page 479: Configuring Generator Operation In The Drive

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Instant in time that retraction was initiated Instant in time when the retraction speed specified in p0893 is reached Instant in time after the time that was configured in p0892 has expired Figure 9-4 Behavior for drive-integrated retraction Feedback signal...
  • Page 480 R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Parameter Description p1248 Lower DC link voltage threshold Setting the lower threshold for the DC link voltage. For p1240 = 2, this threshold is used as setpoint limit for the Vdc_min controller. p1244 Upper DC link voltage threshold p1248 Lower DC link voltage threshold r0208...
  • Page 481: Esr Is Enabled Via System Variable

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive 9.3.5 ESR is enabled via system variable The ESR response of an axis, configured via drive parameter, must be programmed on a user- specific basis in a part program/synchronized action using the following axis-specific system variable: $AA_ESR_ENABLE[<axis>] Value Meaning...
  • Page 482: Acknowledge Esr Reactions

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive System variable The system variables can be evaluated in the part program / synchronized action, for example, in order to trigger the drive-autonomous ESR (see Section "Triggering ESR via system variable (Page 481)") or to acknowledge the ESR reactions triggered in the drive (see Section "Acknowledge ESR reactions (Page 482)").
  • Page 483: Configuring Esr In The Part Program

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive System variable and drive parameters The following diagram shows the relationship between system variables and drive parameters when triggering and acknowledging ESR reactions. ① NC: Enabling the ESR reaction via $AA_ESR_ENABLE = 1 (axis-specific) ②...
  • Page 484: Stopping (Esrs)

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive 9.3.9.1 Stopping (ESRS) Syntax ESRS(<access_1>,<stopping time_1>[,...,<axis_n>,<stopping time_n>]) Meaning Using the function, drive parameters can be changed regarding the drive-au‐ ESRS: tonomous "stop" ESR function. Special situations: ● Must be alone in the block ●...
  • Page 485 R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Meaning Using the function, drive parameters can be changed regarding the drive-au‐ ESRR: tonomous "retract" ESR function. Special situations: ● Must be alone in the block ● Triggers a preprocessing stop ●...
  • Page 486: Boundary Conditions

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Dependencies The programmed values for the retraction path and the retraction velocity refer to the load side. Before writing to the drive parameters these are converted over to the motor side. The transmission ratio effective in the NC at the execution time is applicable for the conversion.
  • Page 487: Esr And Safety Integrated (840D Sl)

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive Block search with calculation in "Program test" (SERUPRO) mode During SERUPRO, the functions ESRS(...) and ESRR( ... ) are executed immediately. Reset behavior The parameter values written using the functions ESRS(...) and ESRR(...) are overwritten with the parameter values saved in the drive when the drive runs up or for a drive warm restart.
  • Page 488: Esr And Safety Integrated (828D)

    R3: Extended stop and retract 9.3 ESR executed autonomously in the drive ● p9561 (SG stop response) ● p9563 (SG-specific stop response) ● p9580 (wait time after which the pulses are safely cancelled after bus failure) ● p9697 (delay time for the pulse suppression after bus failure) ●...
  • Page 489: Boundary Conditions

    R3: Extended stop and retract 9.4 Boundary conditions ESR reactions that are not influenced by safety reactions In order that drive-autonomous ESR reactions in conjunction with Safety Integrated are not influenced by safety reactions as a result of STOP E, STOP F executed in parallel or communication failure, then the appropriate delay times must be configured in the drive regarding safety reactions and the corresponding safety trigger sources to trigger the drive- autonomous ESR reactions.
  • Page 490: Examples

    R3: Extended stop and retract 9.5 Examples Motion synchronous actions Synchronous motion actions are executed in the interpolator clock cycle. Increasing the interpolator clock cycle, e.g. due to a high number of active synchronized motion actions, results in a coarser time grid for evaluating trigger conditions and triggering responses for extended stopping and retraction.
  • Page 491 R3: Extended stop and retract 9.5 Examples Settings Machine data ● MD11602 $MN_ASUP_START_MASK = 'B0101' Ignore stop conditions for ASUB – Bit0: "NC stop, M0 or M01" – Bit3: "ASUB and manual traversing in the JOG mode" ● MD20105 $MC_PROG_EVENT_IGN_REFP_LOCK, Bit<n> = TRUE –...
  • Page 492: Retraction While Thread Cutting

    R3: Extended stop and retract 9.5 Examples Trigger conditions and static synchronized actions Example 1: Trigger condition is the occurrence of alarms, which activate the follow-up (tracking) mode: Program code IDS=02 WHENEVER ($AC_ALARM_STAT B_AND 'H2000') > 0 DO $AC_ESR_TRIGGER=1 Example 2 Trigger condition is when the ELG synchronous monitoring responds, if, e.g.
  • Page 493: Rapid Lift, Absolute And Incremental

    R3: Extended stop and retract 9.5 Examples Program code Comment N50 X0 Y0 G0 N60 POLFMASK(X, Y) ; Enable retraction, axis X Y N70 Z100 G1 F1000 ; Retraction response, axial: X, Y (abs.) N80 POLFMASK(Z) Disable retraction, axis X Y and Enable retraction, axis Z N90 Y10 ;...
  • Page 494: Data Lists

    R3: Extended stop and retract 9.6 Data lists Program code Comment N80 POLFMLIN(X, Y) Enable retraction, axis X Y, linear Relationship N85 POLFMASK(Z) ; Enable retraction, axis Z N90 Z100 G1 F1000 Retraction response: - linear relation: X (inc.), Y (abs.) - axial: Z (abs.) N95 POLF[X]=10 ;...
  • Page 495: Signals

    Alarm status in the channel $AC_ESR_TRIGGER Trigger ESR, channel-specific $AC_STAT Channel status 9.6.3 Signals 9.6.3.1 Signals to channel Signal name SINUMERIK 840D sl SINUMERIK 828D Feedrate disable DB21, ... DBX6.0 9.6.3.2 Signals to axis/spindle Signal name SINUMERIK 840D sl SINUMERIK 828D Feed stop DB31, ...
  • Page 496 R3: Extended stop and retract 9.6 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 497: S9: Setpoint Exchange - 840D Sl Only

    S9: Setpoint exchange - 840D sl only 10.1 Brief description Function The "setpoint exchange" function is used in applications in which the same motor is used to traverse different machine axes. Replacing the technology function "setpoint changeover" (TE5) The "setpoint changeover" function replaces the technology function "setpoint changeover" (TE5).
  • Page 498 S9: Setpoint exchange - 840D sl only 10.2 Startup Figure 10-2 Example 2: 1 motor encoder, separate millhead encoder and spindle encoder Configuration Setpoint exchange enables a number of axes to use the same drive. The same setpoint channel on this drive is assigned a number of times to define the axes participating in setpoint exchange.
  • Page 499 S9: Setpoint exchange - 840D sl only 10.2 Startup Figure 10-3 Setpoint exchange with 2 axes Activation The setpoint is exchanged and the corresponding interface signals are evaluated in the PLC user program. Note An existing PLC user program may need to be modified due to changes in the meaning of interface signals in comparison with the technology card solution.
  • Page 500: Interface Signals

    S9: Setpoint exchange - 840D sl only 10.3 Interface signals Transfer conditions ● Axis standstill of all axes involved. ● Special functions such as reference point approach, measuring, travel to fixed stop, function generator, star/delta changeover, drive parameter set changeover are not active in the axis with drive control.
  • Page 501 S9: Setpoint exchange - 840D sl only 10.3 Interface signals Control signals The control signals contained in the following bytes are only active in the machine axes, which are currently assigned to the drive: DB31, ... DBB20 to DBB21 Controller enable The NC/PLC interface signal: DB31, ...
  • Page 502 S9: Setpoint exchange - 840D sl only 10.3 Interface signals Figure 10-4 Schematic setpoint changeover from machine axes AX1 to AX2 Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 503: Interrupts

    S9: Setpoint exchange - 840D sl only 10.7 Supplementary conditions 10.4 Interrupts Drive alarms are only displayed for axes with drive control 10.5 Position control loop During setpoint exchange, the drive train and therefore the position control loop are isolated. In order to avoid instabilities, exchange only takes place at standstill and once all servo enables have been deleted.
  • Page 504: Data Lists

    S9: Setpoint exchange - 840D sl only 10.8 Data lists Commissioning via SinuCom NC The SinuCom NC commissioning tool can be used to commission the setpoint exchange only via the expert list. Safety Integrated (only 840D sl) A detailed description of the supplementary conditions for setpoint changeover in conjunction with Safety Integrated is available in: References: Manual SINUMERIK Safety Integrated...
  • Page 505: T3: Tangential Control - 840D Sl Only

    T3: Tangential control - 840D sl only 11.1 Brief description Tangential control The tangential control function belongs to the category of NC functions with coupled axes. It is characterized by the following features: ● There are two leading axes which are moved independently by means of normal traversing instructions (leading axes).
  • Page 506: Characteristics Of Tangential Follow-Up Control

    T3: Tangential control - 840D sl only 11.2 Characteristics of tangential follow-up control Applications The tangential control function can be used for example for the following applications: ● Tangential positioning of a rotatable tool for nibbling operations. ● Follow-up control of tool alignment for a bandsaw. ●...
  • Page 507 T3: Tangential control - 840D sl only 11.2 Characteristics of tangential follow-up control Response on follow-up A difference must made between the following cases: ● Without intermediate block (TLIFT) The path velocity of the leading axes is reduced to such an extent that the following axis reaches its target position synchronously with the other axes.
  • Page 508: Using Tangential Follow-Up Control

    T3: Tangential control - 840D sl only 11.3 Using tangential follow-up control 11.3 Using tangential follow-up control Activation The following axis can only be aligned if: ● The assignment between the leading and following axes is declared to the system (TANG) ●...
  • Page 509: Assignment Between Leading Axes And Following Axis

    T3: Tangential control - 840D sl only 11.3 Using tangential follow-up control Further information about the multi-channel block search function SERUPRO, see: References: Function Manual, Basic Functions; Mode Group, Channel, Program Mode, Reset Response (K1), Section: "Program test" 11.3.1 Assignment between leading axes and following axis Programming The programming is carried out using the pre-defined sub-program TANG.
  • Page 510: Switching On Corner Response

    T3: Tangential control - 840D sl only 11.3 Using tangential follow-up control MD37402 $MA_TANG_OFFSET (preselection angle for tangential follow-up) If the angle is zero both in TANGON and in the machine date, the following axis takes the direction of the tangent. Figure 11-2 Tangential control, offset angle of 90 degrees to path tangent Activation is programmed as follows for the above example and an offset angle of 90...
  • Page 511: Termination Of Follow-Up Control

    T3: Tangential control - 840D sl only 11.3 Using tangential follow-up control 11.3.4 Termination of follow-up control Programming The programming is carried out using the pre-defined sub-program TANGOF. The name of the following axis to be decoupled from its leading axes for the remainder of the machining operation must be transferred to the control in conjunction with the subprogram name TANGOF.
  • Page 512 T3: Tangential control - 840D sl only 11.3 Using tangential follow-up control TANGDEL(C) The existing definition in the example of TANG(A, X, Y) is canceled. Example for plane change Program code Comment N10 TANG(A, X, Y, 1) N20 TANGON(A) N30 X10 Y20 N80 TANGOF(A) N90 TANGDEL(A) ;...
  • Page 513: Limit Angle

    T3: Tangential control - 840D sl only 11.4 Limit angle Program code Comment N70 GEOAX(2, Y2) ; Geometry axis switchover permitted N80 TANG(A, X, Y) ; Redef. Follow-up axis group N90 TANGON(A, 90) ; Follow-up group with Y2 is being activated 11.4 Limit angle Description of problem...
  • Page 514: Supplementary Conditions

    T3: Tangential control - 840D sl only 11.5 Supplementary conditions Programming A minimum and a maximum value for the position of the axis made to follow ("C" in example) referred to the base coordinate system are transferred to the control with G25 and G26. These working area limitations are activated with WALIMON and deactivated again with WALIMOF.
  • Page 515: Examples

    T3: Tangential control - 840D sl only 11.6 Examples 11.6 Examples Positioning of workpiece Figure 11-4 Tangential positioning of a workpiece on a bandsaw Positioning of tool Figure 11-5 Positioning of a dressing tool on a grinding wheel Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 516: Data Lists

    T3: Tangential control - 840D sl only 11.7 Data lists Example Corner in area Programming TANG(A,X,Y,1.0,"B") TLIFT(A) G1 G641 X0 Y0 Z0 A0 TANGON(A,0) N4 X10 N5 Z10 N6 Y10 Here, a corner is hidden in the area between N4 and N6. N6 causes a tangent jump. That is why there is no rounding between N5 and N6 and an intermediate block is inserted.
  • Page 517: T4: Automatic Retuning With Ast - Only 840D Sl

    With the "Automatic retuning with AST" function, axes with a changed mechanical system can be retuned from the part program. SIEMENS provides predefined cycles for the functions of the Automatic Servo Tuning (AST) (see "Programming (Page 518)"). They can be used by machine manufacturers to create their own tuning cycles.
  • Page 518: Commissioning

    12.3 Programming 12.2 Commissioning System requirements ● SINUMERIK 840D sl with SINUMERIK Operate ● CNC software as of version 4.7 SP1 Machines with several HMI components On machines with several HMI components, only one HMI must be configured to use the function: 12.3...
  • Page 519: Cycle752 - Add Axis To An Optimization Session

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameter Parameters <S_I_SESSIONCOMMAND> Data type: INT Value: Not defined Open new optimization session Note: Multiple sessions cannot be open concurrently. Opening a new session closes the current session automatically. Close current optimization session Execute optimization functions of the current session with immediate transfer of the optimization results...
  • Page 520 T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters <S_I_ACTIONREQUEST> Meaning: Specifies whether the stored optimization data (optimization strategy and optimization results) should be used or the axis with AST default settings reoptimized for the axis to be added. Note: If the axis has already been optimized and the optimization data has been saved in an optimization file, this data is normally accessed first.
  • Page 521: Cycle753 - Select Optimization Mode

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Note Premeasurement The decision for performing a premeasurement depends on the available axis optimization data. ● Cases in which the control unit performs a premeasurement for determining the measuring parameters: –...
  • Page 522: Cycle754 - Add/Remove Data Set

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters <S_I_ACTION> Meaning: Specifies the optimization mode for the specified axis. Data type: Value: Not defined Match the path interpolation The axis (path axis) should not be remeasured and reoptimized, but rather considered and adapted only for the optimization of the path interpolation.
  • Page 523: Cycle755 - Backup/Restore Data

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters <S_I_ACTIONREQUEST> Meaning: Specifies whether the specified axis or drive data set should be added to or removed from the data set list. Data type: Value: Not defined Add a data set Remove a data set Remove all data sets from the list Note:...
  • Page 524: Cycle756 - Activate Optimization Results

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters Parameters <S_I_ACTIONREQUEST> Data type: Value: Not defined Backup data for a later restore This setting saves the data values at the time of the cycle call in the specified file. The content of the data backup is determined by the following parameters: - the current valid optimization strategy - the axes currently added to the optimization session...
  • Page 525: Cycle757 - Save Optimization Data

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters <S_I_REGULATOR_ROLE> Meaning: Specifies which optimization results should act for the specified axis in the control unit and in the drive. Data type: Value: Not defined The original data should act for the specified axis. The current optimization results without optimization of the path interpolation should act for the specified axis The current optimization results with optimization of the path interpolation should act for the...
  • Page 526 T4: Automatic retuning with AST - only 840D sl 12.3 Programming Parameters <S_SZ_FILENAME> Meaning: Name of the optimization file to be created (without path details, with file extension ".xml" or ".csv") Examples: ● AXIS.xml ● PATH.xml ● SPEEDCTRL_PLANT.csv ● POSCTRL_MECHRESP.csv Data type: STRING [100] <S_I_CONTENT_TYPE>...
  • Page 527: Cycle758 - Change A Parameter Value

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming 12.3.8 CYCLE758 - Change a parameter value CYCLE758 changes individual strategy and measuring settings prior to the optimization or the optimization results after the optimization. Syntax CYCLE758(<S_I_AXIS>, <S_I_PARAMID>, <S_I_MEASTYPE>, <S_I_MEASINDEX>, <S_SZ_NEWVALUE>) Parameters Parameters...
  • Page 528 T4: Automatic retuning with AST - only 840D sl 12.3 Programming Syntax CYCLE759(<S_I_AXIS>, <S_I_PARAMID>, <S_I_MEASTYPE>, <S_I_MEASINDEX>, <S_SZ_GUDRESULT>) Parameters Parameters <S_I_AXIS> Meaning: Machine axis number (not relevant for parameters that affect the optimization of the path interpolation!) Data type: <S_I_PARAM ID> Meaning: ID number of the parameter whose value should be read Parameter IDs, see "List of the parameters for the automatic servo optimization (Page 529)".
  • Page 529: List Of The Parameters For The Automatic Servo Optimization

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming 12.3.10 List of the parameters for the automatic servo optimization The ID of a parameter for the automatic servo optimization whose value should be changed with CYCLE758 or read with CYCLE759 can be taken from the following table: Meaning Unit Data...
  • Page 530 T4: Automatic retuning with AST - only 840D sl 12.3 Programming Meaning Unit Data Value range Write Read type with with CYCLE758 CYCLE759 Amplitude of the excitation Nm, N, m/ Dynamic s, rad/s, m, rad (de‐ pending on the measur‐ ing type) Velocity offset m/s or rad/...
  • Page 531 T4: Automatic retuning with AST - only 840D sl 12.3 Programming Meaning Unit Data Value range Write Read type with with CYCLE758 CYCLE759 Equivalent time constant of the current control loop as parameter for the torque precontrol Adapted parameter value 108 for considering the following effects: ●...
  • Page 532: Manufacturer-Defined Identifier For Cycle Calls And Parameters

    T4: Automatic retuning with AST - only 840D sl 12.3 Programming Meaning Unit Data Value range Write Read type with with CYCLE758 CYCLE759 Optimization aggressiveness of the speed con‐ 0 ... 1 troller A higher value produces a better optimization re‐ sult coupled with a reduction of the robustness Kp upper limit based on the bandwidth of a PT1- equivalent proportionally-controlled system...
  • Page 533 T4: Automatic retuning with AST - only 840D sl 12.3 Programming ; cycles DEFINE ASTSESSION AS CYCLE751 DEFINE ASTADDAXIS AS CYCLE752 DEFINE ASTSETTUNEACTION AS CYCLE753 DEFINE ASTDATASETS AS CYCLE754 DEFINE ASTRESTOREPOINT AS CYCLE755 DEFINE ASTACTIVATERESULT AS CYCLE756 DEFINE ASTSAVETOFILE AS CYCLE757 DEFINE ASTSETPARVAL AS CYCLE758 DEFINE ASTGETPARVAL...
  • Page 534 T4: Automatic retuning with AST - only 840D sl 12.3 Programming DEFINE _AST_REG_AXIS_OPTIMAL AS 2 DEFINE _AST_REG_PATH_OPTIMAL AS 3 ; cycle757 DEFINE _AST_FILECONTENT_AXIS_XML AS 1 DEFINE _AST_FILECONTENT_PATH_XML AS 2 DEFINE _AST_FILECONTENT_SPEEDPLANT AS 3 DEFINE _AST_FILECONTENT_POSMECHRESP AS 4 ; cycle758, cycle759 DEFINE PAR_RSPTST_INITPOS AS 1 DEFINE PAR_RSPTST_DIRSEQ...
  • Page 535 T4: Automatic retuning with AST - only 840D sl 12.3 Programming DEFINE PAR_PCTL_EQTIMESPEEDCTL AS 106 DEFINE PAR_PCTL_EQTIMESPEEDCTLMD AS 107 DEFINE PAR_PCTL_EQTIMCURRCTL AS 108 DEFINE PAR_PCTL_EQTIMECURRCTLMD AS 109 DEFINE PAR_STRAT_PCTL_KVREDFACTOR AS 151 DEFINE PAR_STRAT_PCTL_ENUMKVMAXMETH AS 153 DEFINE PAR_STRAT_PCTL_DSCACTIVATION AS 154 DEFINE PAR_STRAT_PCTL_FFWMODE AS 155 DEFINE PAR_STRAT_PCTL_MAXKV AS 156...
  • Page 536: Supplementary Conditions

    T4: Automatic retuning with AST - only 840D sl 12.5 Examples 12.4 Supplementary conditions Coupling function active The "Automatic retuning with AST" cannot be used on axes of an active coupling function (gantry axes, master-slave coupling, coupled motion, master value coupling, electronic gear, synchronous spindle).
  • Page 537: Example 2: Reoptimize The Speed Controller Of An Axis

    T4: Automatic retuning with AST - only 840D sl 12.5 Examples CYCLE751(4) ; Start the optimization without acti- vating the results. CYCLE757(myaxiswithnewload,"ANEWINERTIA.XML",1,,) ; Save the optimization data in an XML file. CYCLE757(myaxiswithnewload,"ANEWFREQRESP.CSV",3,,) ; Export the speed-controlled system. CYCLE759(myaxiswithnewload,34,,,"_AST_R_ESTINERTIA") ; Fetch the inertia. ;...
  • Page 538 T4: Automatic retuning with AST - only 840D sl 12.5 Examples Preparation ● Select optimization strategy 105 from the user interface: ● If necessary, change the optimization objective (fast, moderate, robust) with the "Speed" softkey. Programming Note In the following program example, those cycle calls required for the reoptimization are highlighted bold.
  • Page 539: Example 3: Reoptimize The Speed Controller And The Position Controller Of An Axis

    T4: Automatic retuning with AST - only 840D sl 12.5 Examples CYCLE752(myaxiswithnewload,3,false) ; Add axis 4, accept optimization strat- egy from the standard optimization file (XML file), do not consider the axis for an optimization of the path interpola- tion. CYCLE754(myaxiswithnewload,2,2,-1) ;...
  • Page 540 T4: Automatic retuning with AST - only 840D sl 12.5 Examples Preparation ● Select optimization strategy 102 from the user interface: ● Press the "Position" softkey to display the window for adaptation of the position controller strategy and select "User-defined strategy 209": Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 541 T4: Automatic retuning with AST - only 840D sl 12.5 Examples ● Select the "Force equivalent time target value" option. Note The "Force equivalent time target value" option is set, when in addition to the speed controller, the position controller is also reoptimized, and despite an equivalent time of the precontrol specified, the path interpolation should not be reoptimized.
  • Page 542: Example 4: Reoptimization Of The Path Interpolation

    T4: Automatic retuning with AST - only 840D sl 12.5 Examples CYCLE754(myaxiswithnewload,1,2,3) ; Add DDS3 to the data set list and over- write after optimization. ; CYCLE754(myaxiswithnewload,1,1,3) ; Axis: Add parameter set 4 to the data set list and overwrite after optimiza- tion.
  • Page 543 T4: Automatic retuning with AST - only 840D sl 12.5 Examples axes involved by being optimized by AST and now, after the reoptimization of the axis, the path interpolation optimization is performed again by AST. The optimization strategy for axis and speed controller selected from the user interface is deployed and stored in the XML file for this axis (\user\sinumerik\data\optimization\AST_AX4_A1….xml).
  • Page 544 T4: Automatic retuning with AST - only 840D sl 12.5 Examples Preparation ● No equivalent time is forced, namely, the predefined strategies for the position controller optimization are deployed: ● For the path interpolation, the strategy (all equivalent times identical or use MD32895 $MA_DESVAL_DELAY_TIME) was already chosen for the first optimization.
  • Page 545 T4: Automatic retuning with AST - only 840D sl 12.5 Examples The following selection sets the balancing times (in MD32800 $MA_EQUIV_CURRCTRL_TIME or MD32810 $MA_EQUIV_SPEEDCTRL_TIME) of all involved axes to the largest time constant: ● The following questions must be answered: –...
  • Page 546: Example 5: Reoptimize The Speed Control Loop To Eliminate Known Periodic Disturbance Frequencies

    T4: Automatic retuning with AST - only 840D sl 12.5 Examples CYCLE752(2.3,true) ; Add axis 2 because of the path inter- polation. CYCLE752(3.3,true) ; Add axis 3 because of the path inter- polation. CYCLE752(5.3,true) ; Add axis 5 because of the path inter- polation.
  • Page 547 T4: Automatic retuning with AST - only 840D sl 12.5 Examples Preparation In the simplest case, only a speed controller optimization is necessary: ● Select optimization strategy 105 from the user interface: Programming Program code Comment DEF INT myAxis=4 CYCLE751(1) ;...
  • Page 548 T4: Automatic retuning with AST - only 840D sl 12.5 Examples Program code Comment CYCLE758(myAxis,208,,,"false" ) ; Deactivate the use of the gain filter. CYCLE751(3) ; Perform the optimization. CYCLE751(2) ; Close the optimization session. ; For normal broadband faults. Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 549: Te01: Installation And Activation Of Loadable Compile Cycles

    Technology and special functions provided by Siemens in the form of compile cycles New compile cycle files (*.ELF) for Siemens technology and special functions for CNC software versions as of SW 4.5 can be obtained from your regional Siemens sales office.
  • Page 550: Loading Compile Cycles

    Compile cycles are functional expansions of the NCK system software that can be created by the machine manufacturer and/or by Siemens and then imported in the control later. As part of the open NCK system architecture, compile cycles have comprehensive access to data and functions of the NCK system level via defined software interfaces.
  • Page 551: Loading A Compile Cycle With Hmi Advanced

    TE01: Installation and activation of loadable compile cycles 13.1 Loading compile cycles Execution Perform the following steps to load a compile cycle from a USB-FlashDrive, for example: 1. Insert the USB-FlashDrive in the USB interface on the front of the operator panel. 2.
  • Page 552: Loading A Compile Cycle From An External Computer With Winscp3

    TE01: Installation and activation of loadable compile cycles 13.2 Interface version compatibility 13.1.3 Loading a compile cycle from an external computer with WinSCP3 Precondition To transfer a compile cycle to the control, the following requirements must be met: ● The external computer (programming device / PC) which the compile cycle is loaded onto is linked to the PCU via a network (TCP / IP).
  • Page 553 TE01: Installation and activation of loadable compile cycles 13.2 Interface version compatibility Interface versions The relevant interface versions are displayed under: ● Interface version of the NCK system software HMI Advanced: Diagnosis > Service Display > Version > NCU Version Display (excerpt) ------------------------------------------- CC Interface Version:...
  • Page 554: Software Version Of A Compile Cycle

    TE01: Installation and activation of loadable compile cycles 13.4 Activating the technological functions in the NCK 13.3 Software version of a compile cycle The SW version of a compile cycle is displayed under: HMI Advanced: Diagnosis > Service Display > Version > NCU Version Display (excerpt) ------------------------------------------- CC Interface Version:...
  • Page 555: Function-Specific Startup

    After the NCK is booted up next, the activated technology functions are integrated into the system software. CAUTION SINUMERIK 840D sl The following alarm is displayed when a bit is set for the first time in the function-specific NCK machine data: $MN_CC_ACTIVE_IN_CHAN_XXXX[0] Alarm 4400 "MD modification causes reorganization of the buffered memory (data loss)"...
  • Page 556: Creating Alarm Texts With Hmi Advanced

    9. Restart SINUMERIK Operate. More information about creating alarm texts with SINUMERIK Operate can be found in: References: SINUMERIK 840D sl Base Software and Operating Software Commissioning Manual; SINUMERIK Operate (IM9) Commissioning Manual, Section: Configuring machine data/alarms 13.6.2 Creating alarm texts with HMI Advanced The following alarms should be added to the alarm texts of the technology functions: 075999 0 0 "Channel %1 Sentence %2 Call parameter is invalid"...
  • Page 557: Upgrading A Compile Cycle

    TE01: Installation and activation of loadable compile cycles 13.7 Upgrading a compile cycle Proceed as follows 1. Copy the "mbdde.ini" file from the "F:\mmc2" directory to the "F:\oem" directory. 2. Add the following two lines in the "mbdde.ini" file: [TextFiles] UserCZYK=F:\oem\alc_ 3.
  • Page 558: Deleting A Compile Cycle

    TE01: Installation and activation of loadable compile cycles 13.8 Deleting a compile cycle 4. Initiate a power on reset with an NCK general reset (NCK commissioning switch: position "1"). See: References: Commissioning Manual IBN CNC: NCK, PLC, Drive; Section: General tips > Separate NCK and PLC general reset >...
  • Page 559: Data Lists

    TE01: Installation and activation of loadable compile cycles 13.9 Data lists 3. Initiate a power on reset with an NCK general reset (NCK commissioning switch: position "1"). See: References: Commissioning Manual IBN CNC: NCK, PLC, drive; Chapter: General tips > Separate NCK and PLC general reset >...
  • Page 560 TE01: Installation and activation of loadable compile cycles 13.9 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 561: Te02: Simulation Of Compile Cycles (Only Hmi Advanced)

    TE02: Simulation of Compile Cycles (only HMI Advanced) 14.1 Brief description 14.1.1 Function If, at the SINUMERIK user interface "HMI Advanced" part programs are simulated that use compile cycles, depending on the compile cycle being used, general specific conditions must be established.
  • Page 562 TE02: Simulation of Compile Cycles (only HMI Advanced) 14.2 OEM transformations 4. Close the HMI application. 5. Launch the HMI application. 6. In the directory for the manufacturer cycles, create the file "TRAORI.SPF" with the following content: PROC TRAORI(INT II) 7.
  • Page 563: Te1: Clearance Control - 840D Sl Only

    Note The "clearance control" technological function is only available in the first NC channel! Availability The "clearance control" technological function is available in SINUMERIK 840D sl. Compile cycle The "clearance control" technological function is a compile cycle. For a description of the system-specific availability and use of compile cycles (see Section "TE01: Installation and activation of loadable compile cycles (Page 549)").
  • Page 564: Function Description

    System overview (840D sl) An overview of the system components required for clearance control in conjunction with SINUMERIK 840D sl is provided in the following diagram. Figure 15-1 System components for clearance control with SINUMERIK 840D sl...
  • Page 565: Clearance Control

    TE1: Clearance control - 840D sl only 15.2 Clearance control 1D/ 3D machining Clearance control can be used for 1D and 3D machining with up to five interpolatory axes. ● 1D machining For the 1D machining, only one axis is affected by the clearance control. For example, axis Z, as shown in the machine configuration described in the system overview (see previous figure).
  • Page 566 TE1: Clearance control - 840D sl only 15.2 Clearance control Clearance control characteristics Clearance control is based on the two characteristics shown in the following diagram: ● Clearance sensor characteristic (sensor property) ● Clearance control characteristic (can be parameterized via machine data) Figure 15-2 Correlation between characteristics: Clearance sensor and clearance control ●...
  • Page 567: Velocity Feedforward Control

    In order to maximize the dynamics of the control response, clearance control takes place on the highest priority position controller level of the NCK. SINUMERIK 840D sl with I/O modules and drives connected via PROFIBUS DP produces a deadtime T...
  • Page 568: Control Loop Structure

    TE1: Clearance control - 840D sl only 15.2 Clearance control The speed filters of the SINAMICS S120 drive provide additional damping capabilities: ● Parameter 1414 and following: (time constant for speed setpoint filter 1, 2) Note Every damping measure implemented contributes to increasing the overall time constant of the control loop! You will find a complete description of the velocity feedforward control in: References...
  • Page 569: Compensation Vector

    TE1: Clearance control - 840D sl only 15.2 Clearance control Figure 15-4 Control structure, clearance control (principle) 15.2.4 Compensation vector Standard compensation vector The compensation vector of the clearance control and the tool orientation vector are normally identical. Consequently, the compensation movement of the clearance control is normally always in the direction of the tool orientation.
  • Page 570 TE1: Clearance control - 840D sl only 15.2 Clearance control Note In all the figures in this section, the traversing movement of the machining head needed in order to machine the workpiece is in the direction of the Y coordinate, i.e. perpendicular to the drawing plane.
  • Page 571 TE1: Clearance control - 840D sl only 15.2 Clearance control Figure 15-7 Programmable compensation vector Changes in orientation Based on the above observations, a different behavior also results when the orientation of the machining head is changed while the clearance control is active. In the following diagram the normal case is shown on the left (compensation vector == tool orientation vector);...
  • Page 572: Technological Features Of Clearance Control

    TE1: Clearance control - 840D sl only 15.3 Technological features of clearance control The meaning of the individual positions of the machining head is as follows: 1. Programmed position of the machining head 2. Actual position of the machining head with clearance control active before the orientation change 3.
  • Page 573: Sensor Collision Monitoring

    TE1: Clearance control - 840D sl only 15.4 Sensor collision monitoring ● Control options via the PLC interface The following signals are available at the PLC interface: Status signals: – Closed-loop control active – Overlaying movement at standstill – Lower limit reached –...
  • Page 574: Startup

    TE1: Clearance control - 840D sl only 15.5 Startup 15.5 Startup Compile cycle Before commissioning the technological function, ensure that the corresponding compile cycle has been loaded and activated (see Section "TE01: Installation and activation of loadable compile cycles (Page 549)"). 15.5.1 Activating the technological function The technological function is activated via the machine data:...
  • Page 575: Parameters Of The Programmable Compensation Vector

    TE1: Clearance control - 840D sl only 15.5 Startup Analog input The following machine data must be parameterized for the analog input: ● MD10300 $MN_FASTIO_ANA_NUM_INPUTS (number of active analog NCK inputs) ● MD10362 $MN_HW_ASSIGN_ANA_FASTIN (per analog module) (hardware assignment for the fast analog NCK inputs) Specifying the physical address activates the analog input module Digital input The following machine data must be parameterized for the digital input:...
  • Page 576 TE1: Clearance control - 840D sl only 15.5 Startup 4. The direction axes may not participate in an axis coupling, e.g. transformation, electronic gear, etc. 5. To ensure that the dynamic response of the path is not limited by the dynamic response of the direction axes, the following machine data for the direction axes must be set equal to or more than the corresponding values of the geometry axes of the channel: –...
  • Page 577: Parameter Settings For Clearance Control

    TE1: Clearance control - 840D sl only 15.5 Startup 15.5.5 Parameter settings for clearance control The part program name The following machine data must be parameterized for the declaration of the function-specific part program name CLC_GAIN and CLC_VOFF: ● MD10712 $MN_NC_USER_CODE_CONF_NAME_TAB[0] = "OMA1" (list of re-configured NC codes) ●...
  • Page 578: Starting Up Clearance Control

    TE1: Clearance control - 840D sl only 15.5 Startup ● MD36040 $MA_STANDSTILL_DELAY_TIME[<x>] (standstill monitoring time delay) ● MD36060 $MA_STANDSTILL_VELO_TOL[<x>] ("axis/ spindle stopped" velocity/speed threshold) <x> = axis number of clearance-controlled machine axis 15.5.6 Starting up clearance control Clearance sensor The clearance sensor outputs should be connected to the I/O modules that were activated using the following machine data: ●...
  • Page 579 TE1: Clearance control - 840D sl only 15.5 Startup The voltage specification for the analog output $A_OUTA[6] used in the synchronized action is subtracted from the clearance sensor input voltage by the clearance control function and therefore has the opposite polarity to the input signal. Set the following machine data to induce the clearance control function to use analog output 6 ($A_OUTA[6]) as an additional input overlaid on the sensor input: MD10366 $MN_CLC_OFFSET_ASSIGN_ANAOUT = 6 (hardware assignment for the external...
  • Page 580: Programming

    TE1: Clearance control - 840D sl only 15.6 Programming Completion A data backup is recommended once the start-up procedure has been completed. References: Commissioning Manual IBN CNC: NCK, PLC, drive Note A data backup is recommended once the start-up procedure has been completed. 15.6 Programming 15.6.1...
  • Page 581 TE1: Clearance control - 840D sl only 15.6 Programming ● CLC(0) Deactivation of clearance control without canceling the position offset. If the clearance-controlled axes are still moving at the instant of deactivation due to the sensor signal, they are stopped. The workpiece coordinate system (WCS) is then synchronized with the corresponding standstill positions.
  • Page 582 TE1: Clearance control - 840D sl only 15.6 Programming Block change with exact stop If exact stop is active at the end of the block (G60/G09 with G601/G602) the block change may be delayed due to axis movements induced by the clearance control sensor signal. Sensor collision monitoring A digital input for an additional collision signal can be configured by the sensor using the following machine data:...
  • Page 583 TE1: Clearance control - 840D sl only 15.6 Programming Compensation vector Actual position of the direction axes If the clearance control is activated with a programmable compensation vector at a position of 0 on all 3 direction axes, a compensation vector cannot be calculated from this information. The following alarm is then displayed: number , error ID: 1, angle 0.0"...
  • Page 584 TE1: Clearance control - 840D sl only 15.6 Programming Figure 15-10 Interpolation of the compensation vector The compensation vector must be oriented by programming the direction axes at [1, 0, 0] before part program block N100. In part program block N100, the end position of the compensation vector is oriented by programming the direction axes at [0, 0, -1].
  • Page 585: Closed-Loop Control Gain (Clc_Gain)

    TE1: Clearance control - 840D sl only 15.6 Programming In the case of a re-orientation (rotation) of the compensation vector, it is also necessary to note the ratio between the programmed traversing path and the configured dynamic response of the direction axes. The ratio should be chosen such that the programmed traversing path is not traversed in one or a small number of interpolation cycles, due to the dynamic response of the axis.
  • Page 586 TE1: Clearance control - 840D sl only 15.6 Programming Functionality The current closed-loop control gain for clearance control is produced by the active characteristic specified via machine data: ● MD62510 $MC_CLC_SENSOR_VOLTAGE_TABLE1 (coordinate voltage of interpolation points sensor characteristic 1) ● MD62511 $MC_CLC_SENSOR_VELO_TABLE1 (coordinate velocity of interpolation points sensor characteristic 1) ●...
  • Page 587: Limiting The Control Range (Clc_Lim)

    TE1: Clearance control - 840D sl only 15.6 Programming Figure 15-11 Response of the CLC offset vector when CLC_GAIN=0.0 Reset Within a part program, a modified gain factor must be reset by means of explicitly programming CLC_GAIN=1.0. RESET behavior CLC_GAIN=1.0 becomes effective after a power on reset, NC RESET or end of program. 15.6.3 Limiting the control range (CLC_LIM) Syntax...
  • Page 588 TE1: Clearance control - 840D sl only 15.6 Programming Functionality The maximum control range for clearance control can be modified on a block-specific basis using CLC_LIM. The maximum programmable lower/upper limit is limited by the limit value preset in the relevant machine data: ●...
  • Page 589: Direction-Dependent Traversing Motion Disable

    TE1: Clearance control - 840D sl only 15.6 Programming Error messages The following programming errors are displayed with an alarm: ● Programming more than 2 arguments number Block number CLC_LIM: general programming – CLC alarm "75005 Channel error" ● Programming arguments outside the permissible limits number Block number CLC_LIM Value greater than MD –...
  • Page 590 TE1: Clearance control - 840D sl only 15.6 Programming Parameterization The following machine data is used to parameterize the digital outputs: ● MD62523 $MC_CLC_LOCK_DIR_ASSIGN_DIGOUT[n] (assignment of the digital outputs for disabling the CLC movement) n = 0 → Digital output for disabling the negative traversing direction n = 1 →...
  • Page 591: Voltage Offset, Can Be Set On A Block-Specific Basis (Clc_Voff)

    TE1: Clearance control - 840D sl only 15.6 Programming 15.6.5 Voltage offset, can be set on a block-specific basis (CLC_VOFF) Syntax oltage offset CLC_VOFF = V Voltage offset ● Format: Real ● Unit: Volts ● Range of values: No restrictions CLC_VOFF is an NC address and can therefore be written together with other instructions in a part program block.
  • Page 592: Selection Of The Active Sensor Characteristic (Clc_Sel)

    TE1: Clearance control - 840D sl only 15.6 Programming Number of the parameterized analog output (see "Parameter Assignment" section) ● Format: Integer ● Range of values: 1, 2, . . .max. number of analog outputs oltage offset Just like the voltage offset for CLC_VOFF (see Section "Voltage offset, can be set on a block- specific basis (CLC_VOFF) (Page 591)").
  • Page 593: Function-Specific Display Data

    TE1: Clearance control - 840D sl only 15.7 Function-specific display data CLC_SEL(...) is a procedure call and must therefore be programmed in a dedicated part program block. Characteristic number = 2 selects characteristic 2. Any other value selects characteristic 1 without alarm.
  • Page 594: Channel-Specific Gud Variables

    TE1: Clearance control - 840D sl only 15.7 Function-specific display data Types of variable The display data is available both as channel-specific GUD (Global User Data) variables and as OPI variables. 15.7.1 Channel-specific GUD variables The "clearance control" technological function provides the following channel-specific GUD variables for HMI applications: ●...
  • Page 595 TE1: Clearance control - 840D sl only 15.7 Function-specific display data 1. Edit the GUD variable definitions DEF CHAN REAL CLC_DISTANCE[3] ; Array of real, 3 elements DEF CHAN REAL CLC_VOLTAGE[3] ; Array of real, 3 elements 2. Save the file and close the editor. 3.
  • Page 596: Opi Variable

    TE1: Clearance control - 840D sl only 15.7 Function-specific display data 15.7.2 OPI variable The "clearance control" technological function provides the following channel-specific OPI variables as display data for the HMI application: ● SINUMERIK HMI Advanced Table 15-2 Channel-specific OPI variable OPI variable Description Unit...
  • Page 597: Function-Specific Alarm Texts

    NC using an I/O module with an analog input. Connection options The SIMATIC ET 200S I/O for SINUMERIK 840D sl is connected via PROFIBUS DP. The clearance sensor is connected via an analog S7 I/O module. Figure 15-13 I/O modules connection for SINUMERIK 840D sl...
  • Page 598: External Smoothing Filters

    TE1: Clearance control - 840D sl only 15.9 Supplementary conditions Suitable I/O modules As the A/D conversion time directly affects the deadtime of the clearance control servo loop, only one I/O module may be used with low conversion time. Suitable SIMATIC S7 I/O modules for the clearance control are: ●...
  • Page 599 TE1: Clearance control - 840D sl only 15.9 Supplementary conditions Followup If a clearance-controlled axis is to be switched as an alarm response or via the corresponding interface signal from the NC/PLC in "follow-up" mode, setpoint output will cease for clearance control on this axis.
  • Page 600 TE1: Clearance control - 840D sl only 15.9 Supplementary conditions Displaying the axis position The actual current axis position of a clearance-controlled axis as the sum of an interpolatory axis position and the current position offset of clearance control is not displayed in the main machine screen: ●...
  • Page 601: Data Lists

    TE1: Clearance control - 840D sl only 15.10 Data lists Frame rotations If the Z axis is not a geometry axis, the CRPL(1,0) command rather than the CROT(Z,0) command must used be for frame rotations around the Z axis. 15.10 Data lists 15.10.1 Machine data...
  • Page 602: Axis/Spindlespecific Machine Data

    TE1: Clearance control - 840D sl only 15.10 Data lists Number Identifier: $MC_ Description 62513 CLC_SENSOR_VELO_TABLE_2 Coordinate velocity of interpolation points sensor char‐ acteristic 2 62516 CLC_SENSOR_VELO_LIMIT Clearance control movement velocity 62516 CLC_SENSOR_ACCEL_LIMIT Clearance control movement acceleration 62520 CLC_SENSOR_STOP_POS_TOL Positional tolerance for status message "Clearance control zero speed"...
  • Page 603: Signals

    TE1: Clearance control - 840D sl only 15.10 Data lists Number Short name Long name p1417[0...n] n_soll_filt 1 fn_n Speed setpoint filter 1 denominator natural frequency p1418[0...n] n_soll_filt 1 D_n Speed setpoint filter 1 denominator damping p1419[0...n] n_soll_filt 1 fn_z Speed setpoint filter 1 numerator natural frequency p1420[0...n] n_soll_filt 1 D_z...
  • Page 604 TE1: Clearance control - 840D sl only 15.10 Data lists Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 605: Te3: Speed/Torque Coupling, Master-Slave

    TE3: Speed/torque coupling, master-slave 16.1 Brief description A master-slave coupling is a speed setpoint coupling between a master and any number of slave axes - performed at the position controller level - with and without torque equalization control. The coupling can be permanently switched on, dynamically switched on/off and reconfigured.
  • Page 606: Coupling Diagram

    TE3: Speed/torque coupling, master-slave 16.2 Coupling diagram 16.2 Coupling diagram If the coupling is closed, the slave axis is traversed only with the load-side setpoint speed of the master axis. It is therefore only speed-controlled, not position-controlled. There is no differential position control between the master and slave axis.
  • Page 607: Configuring A Coupling

    TE3: Speed/torque coupling, master-slave 16.3 Configuring a coupling 16.3 Configuring a coupling Static assignment For a speed setpoint coupling and torque equalization control, the static assignment of master and slave axis is defined separately in the following machine data: ● Speed setpoint coupling MD37250 $MA_MS_ASSIGN_MASTER_SPEED_CMD[<slave axis>] = <machine axis number of the master axis for the speed setpoint coupling>...
  • Page 608 TE3: Speed/torque coupling, master-slave 16.3 Configuring a coupling (See Section "Tension torque (Page 611)") Supplementary conditions The following supplementary conditions must be observed for the dynamic assignment: ● When the coupling is switched on, a change to the assignment using MASLDEF has no effect.
  • Page 609: Torque Compensatory Controller

    TE3: Speed/torque coupling, master-slave 16.4 Torque compensatory controller General supplementary conditions Please note the following general supplementary conditions: ● A slave axis can only be assigned to one master axis ● A master axis can be assigned several slave axes ●...
  • Page 610 TE3: Speed/torque coupling, master-slave 16.4 Torque compensatory controller Scaling Scaling the machine data: ● MD37256 $MA_MS_TORQUE_CTRL_P_GAIN (gain factor (P component)) ● MD37260 $MA_MS_MAX_CTRL_VELO (speed setpoint limiting) Must be entered using the following machine data: MD37253 $MA_MS_FUNCTION_MASK[<slave axis>], bit 0 = <value> <value>...
  • Page 611: Tension Torque

    TE3: Speed/torque coupling, master-slave 16.5 Tension torque Deactivating the torque equalization controller If the following settings are made, the torque compensatory controller will be inactive: ● MD37254 $MA_MS_TORQUE_CTRL_MODE[<slave axis>] = 3 ● MD37256 $MA_MS_TORQUE_CTRL_P_GAIN[<slave axis>] = 0 Torque weighting The percentage of the torque generated by the slave axis of the total torque can be set using the torque weighting.
  • Page 612 TE3: Speed/torque coupling, master-slave 16.5 Tension torque Setting The tension torque is entered as a percentage of the rated torque of the slave axis and is active immediately: MD37264 $MA_MS_TENSION_TORQUE[<slave axis>] = <tension torque> As can be seen from the structure of the torque equalization controller (Chapter "Torque compensatory controller (Page 609)"), the tension torque is entered via a PT1 filter.
  • Page 613 TE3: Speed/torque coupling, master-slave 16.5 Tension torque Axis Reference axis of the speed set‐ Reference axis of the torque equal‐ Input of the torque equalization point coupling ization controller contr. MD37250 = value MD37252 = value MD37254 = value Value Description Value Description...
  • Page 614 TE3: Speed/torque coupling, master-slave 16.5 Tension torque Program code Comment PROC MASL_SWITCH IPRTLOCK DISPLOF IF MASL_REQUEST==4 ; Activate "1x4 axes" MASLOF(AX2, AX4) ; Switch off coupling MASLDEL(AX2, AX4) ; Delete coupling $MA_MS_FUNCTION_MASK[AX4]= ; MD37253, bit1 = 1: $MA_MS_FUNCTION_MASK[AX4] B_OR 'B10' ;...
  • Page 615: Closing/Opening A Coupling

    TE3: Speed/torque coupling, master-slave 16.6 Closing/opening a coupling Figure 16-4 Example 2: Alternating coupling with 1x4 and 2x2 axes 16.6 Closing/opening a coupling Default setting After the control has booted, the following machine data defines whether the coupling is permanently switched on (static) or can be dynamically switched on/off and reconfigured: MD37262 $MA_MS_COUPLING_ALWAYS_ACTIVE[<slave axis>] = <switch on mode>...
  • Page 616 TE3: Speed/torque coupling, master-slave 16.6 Closing/opening a coupling Program code Comment N300 $MA_MS_COUPLING_ALWAYS_ACTIVE[AX2] = 1 ; Coupling type: dynamic -> static, switch on cou- pling. Note A statically switched on coupling can neither be switched on/off nor reconfigured using the master-slave-specific NC/PLC interface signals and/or program commands.
  • Page 617: Response On Activation/Deactivation

    TE3: Speed/torque coupling, master-slave 16.7 Response on activation/deactivation $AA_MASL_STAT[<slave axis>] Value Description 1) The coupling of the slave axis is not active. 2) The specific axis is not a slave axis > 0 The coupling is active. <value> == Machine axis number of the master axis NC/PLC interface signal The current coupling state of a slave axis can be read using the following axis-specific NC/PLC interface signal:...
  • Page 618 TE3: Speed/torque coupling, master-slave 16.7 Response on activation/deactivation When switching on the coupling using the program command MASLON, the system waits until the coupling is closed before the block is changed. The "Master-slave switchover active" message will be displayed on the user interface during this time. Switching on/off during motion (spindle) Note Switching on/off during motion...
  • Page 619 TE3: Speed/torque coupling, master-slave 16.7 Response on activation/deactivation Switching on When switching on during the motion, the coupling operation divides itself at different speeds in two phases. ● Phase 1 The PLC user program must request the switch on of the coupling with: DB31, ...
  • Page 620 TE3: Speed/torque coupling, master-slave 16.7 Response on activation/deactivation – MD37272 $MA_MS_VELO_TOL_FINE ("Tolerance fine"). Note The "Speed tolerance coarse" signal can be used to implement a monitoring function on the PLC side that checks a coupled master-slave grouping for loss of speed synchronism.
  • Page 621: Supplementary Conditions

    TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions Switch off with braking If the coupling is switched off with the MASLOFS program command, the coupling will be switched off immediately for spindles in speed control mode and the slave spindles braked. Note The implicit preprocessing stop is omitted for MASLON and MASLOF.
  • Page 622 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions ● When the coupling is switched-in (closed) via the slave axis, the master axis is braked automatically, if the channel axis is in the same channel: ⇒ asymmetrical behavior when activating and deactivating the coupling: –...
  • Page 623: Axial Nc/Plc Interface Signals

    TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions ● The maximum master spindle chuck speed must be configured in the following machine data to be lower than or equal to that of the following spindles: MD35100 $MA_SPIND_VELO_LIMIT[<master spindle>] ● The axial velocity monitoring function should be adapted to the chuck speed: MD36200 $MA_AX_VELO_LIMIT[<master spindle>] 16.8.2 Axial NC/PLC interface signals...
  • Page 624: Interaction With Other Functions

    TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions ● If, for the master or slave axis, one of the following drive status signals is not set: DB31, ... DBX61.7 (current controller active) == 0 OR DB31, ... DBX61.6 (speed controller active) == 0 then, when the slave axis is at a standstill, the status signal is reset: DB31, ...
  • Page 625 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions Dynamic stiffness control (DSC) The "Dynamic stiffness control (DSC)" function must either be active or not active for all axes of a master-slave grouping. MD32640 $MA_STIFFNESS_CONTROL_ENABLE Speed/torque feedforward control (FFW) The "Speed/torque feedforward control (FFW)" function does not have to be activated explicitly in the slave axis.
  • Page 626 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions Gear stage change with activated master-slave coupling An automatic gear stage change in a coupled slave spindle is not possible and can only be implemented indirectly using the master spindle. The point in time at which the gear stage is changed is then derived from the master spindle.
  • Page 627 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions Figure 16-7 Coupling between container spindle S3 and auxiliary motor AUX (prior to rotation) Figure 16-8 Coupling between container spindle S3 and auxiliary motor AUX (after to rotation) Hardware and software limit switches If the software or hardware limit switch is overtraveled by a slave axis, the master-slave grouping is stopped via the master axis.
  • Page 628 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions The movement away of the slave axis from the limit switch must be performed via the traversing of the master axis. Switching off the coupling (MASLOF) and traversing the slave axis is not possible.
  • Page 629 TE3: Speed/torque coupling, master-slave 16.8 Supplementary conditions ● The system ASUB "PROGEVENT.SPF" must be saved under the following path: / _N_CMA_DIR/_N_PROG_EVENT_SPF ● The following machine data should be parameterized so that PROGEVENT.SPF is started. NC-specific machine data: – MD11450 $MN_SEARCH_RUN_MODE = 'H02' –...
  • Page 630: Examples

    TE3: Speed/torque coupling, master-slave 16.9 Examples For more application examples (see Chapter "Examples (Page 630)). Note For an activated coupling, it is recommended to only use block search type 5, "Block search via program test" (SERUPRO) for a block search. More information about event-driven program calls and "block search using the program test"...
  • Page 631: Closing/Separating The Coupling Via The Part Program For The Sinumerik 840D Sl

    TE3: Speed/torque coupling, master-slave 16.9 Examples Preconditions ● One configured master axis MD37250 $MA_MS_ASSIGN_MASTER_SPEED_CMD ≠ 0 ● Activation of a master-slave coupling via MD37262 $MA_MS_COUPLING_ALWAYS_ACTIVE=0 ● The coupling is open. Typical sequence of operations Action Effect/comment Each axis moves to the coupling position. ●...
  • Page 632: Release The Mechanical Brake

    TE3: Speed/torque coupling, master-slave 16.9 Examples Program code Comment N40 MASLOF (AX2) Open the coupling. (Mechanically separate the axes) N50 AX1=200 AX2=200 ; Move the axes separately. N60 M30 16.9.4 Release the mechanical brake This application allows implementation of a brake control for machine axes AX1=Master axis and AX2=Slave axis in a master-slave coupling.
  • Page 633: Data Lists

    TE3: Speed/torque coupling, master-slave 16.10 Data lists 16.10 Data lists 16.10.1 Machine data 16.10.1.1 Axis/spindlespecific machine data Number Identifier: $MA_ Description 37250 MS_ASSIGN_MASTER_SPEED_CMD Leading axis for speed setpoint coupling 37252 MS_ASSIGN_MASTER_TORQUE_CTR Leading axis for torque distribution 37254 MS_TORQUE_CTRL_MODE Connection of torque control output 37255 MS_TORQUE_CTRL_ACTIVATION Activate torque compensatory control...
  • Page 634: Signals

    TE3: Speed/torque coupling, master-slave 16.10 Data lists 16.10.3 Signals 16.10.3.1 Signals to axis/spindle Signal name SINUMERIK 840D sl SINUMERIK 828D Request master-slave torque compensatory controller DB31, ..DBX24.4 DB380x.DBX5000.4 On==1 / Off==0 Request master-slave coupling On==1 / Off==0 DB31, ..DBX24.7 DB380x.DBX5000.7...
  • Page 635: Te4: Handling Transformation Package - 840D Sl Only

    TE4: Handling transformation package - 840D sl only 17.1 Brief description Functionality The handling transformation package has been designed for use on manipulators and robots. The package is a type of modular system, which enables the customer to configure the transformation for his machine by setting machine data (provided that the relevant kinematics are included in the handling transformation package).
  • Page 636: Kinematic Transformation

    TE4: Handling transformation package - 840D sl only 17.3 Definition of terms 17.2 Kinematic transformation Task of a transformation The purpose of a transformation is to transform movements in the tool tip, which are programmed in a Cartesian coordinate system, into machine axis positions. Fields of application The handling transformation package described here has been designed to cover the largest possible number of kinematic transformations implemented solely via parameter settings in...
  • Page 637: Definition Of Positions And Orientations Using Frames

    TE4: Handling transformation package - 840D sl only 17.3 Definition of terms 17.3.2 Definition of positions and orientations using frames In order to make a clear distinction from the term "frame" as it is used in the NC language, the following description explains the meaning of the term "frame"...
  • Page 638: Definition Of A Joint

    TE4: Handling transformation package - 840D sl only 17.3 Definition of terms Figure 17-1 Example of rotation through RPY angles 17.3.3 Definition of a joint Meaning A sliding joint is implemented using a translatory axis, and a swivel joint, using a rotary axis. The basic axis identifiers result from the arrangement and sequence of the individual joints.
  • Page 639: Configuration Of A Kinematic Transformation

    TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Figure 17-2 Joint identifying letters 17.4 Configuration of a kinematic transformation Meaning In order to ensure that the kinematic transformation can convert the programmed values into axis motions, it must have access to some information about the mechanical construction of the machine.
  • Page 640: General Machine Data

    TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation 17.4.1 General machine data MD24100 $MC_TRAFO_TYPE_1 (definition of transformation 1 in the channel) The value 4100 must be entered in this data for the handling transformation package. MD24110 $MC_TRAFO_AXES_IN_1 (axis assignment for transformation) The axis assignment at the transformation input defines which transformation axis is mapped internally onto a channel axis.
  • Page 641: Parameterization Using Geometry Data

    TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation 17.4.2 Parameterization using geometry data Modular principle The machine geometry is parameterized according to a type of modular principle. With this method, the machine is successively configured in geometry parameters from its base center point to the tool tip, thereby producing a closed kinematic loop.
  • Page 642 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Frame between base center point and internal coordinate system The frame T_IRO_RO links the base center point of the machine (BCS = RO) with the first internal coordinate system (IRO) determined by the transformation.
  • Page 643 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation ① SS: MD62603 = 1, gantry (3 linear axes, rectangular) ② CC: MD62603 = 2, SCARA (1 linear axis, 2 rotary axes (in parallel)) ③ CS: MD62603 = 6, SCARA (2 linear axes, 1 rotary axis (spin axis)) ④...
  • Page 644 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Position of the 4th axis Whether the 4th axis is mounted parallel, anti-parallel or perpendicular to the last rotary basic axis is specified in machine data: ●...
  • Page 645 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Figure 17-5 Overview of wrist axis configuration Parameterization of wrist axes With the following machine data, using a special frame type, the geometry of the wrist axis and/or the position of the coordinate system in the wrist axis with respect to one another is defined.
  • Page 646 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Table 17-1 Configuring data for a central wrist axis Machine data Value MD62604 $MC_TRAFO6_WRIST_AXES MD62614 $MC_TRAFO6_DHPAR4_5A [0.0, 0.0] MD62615 $MC_TRAFO6_DHPAR4_5D [0.0, 0.0] MD62616 $MC_TRAFO6_DHPAR4_5ALPHA [-90.0, 90.0] Inclined wrist axis The inclined wrist axis differs from the central wrist axis in two respects, i.e.
  • Page 647 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Figure 17-8 Link frames Frame: T_IRO_RO Frame T_IRO_RO provides the link between the base center point coordinate system (RO) defined by the user and the internal robot coordinate system (IRO). The internal robot coordinate system is predefined in the handling transformation package for each basic axis type and included in the kinematic diagrams for the basic axis arrangements.
  • Page 648 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Frame: T_X3_P3 Frame T_X3_P3 describes the method used to attach the wrist axis to the basic axes. Frame T_X3_P3 is used to link the coordinate system of the last basic axis (p3_q3_r3 coordinate system) with the coordinate system of the first wrist axis (x3_y3_z3 coordinate system).
  • Page 649 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation Changing the axis sequence Rearrangement of axes: MD62620 Note With certain types of kinematics, it is possible to transpose axes without changing the behavior of the kinematic transformation. Machine data: MD62620 $MC_TRAFO6_AXIS_SEQ (rearrangement of axes) The axes on the machine are numbered consecutively from 1 to 6 and must be entered in the internal sequence in machine data:...
  • Page 650 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation ① Kinematics 1 ② Kinematics 2 Figure 17-9 Rearrangement of axes 1 Example 2 This example involves a SCARA kinematic transformation as illustrated in Fig. "Rearrangement of axes 2", in which the axes can be freely transposed. Kinematic 1 is directly included in the handling transformation package.
  • Page 651 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation The mathematical zero points of axes are preset in the handling transformation package. However, the mathematical zero point does not always correspond to the mechanical zero point (calibration point) of axes.
  • Page 652 TE4: Handling transformation package - 840D sl only 17.4 Configuration of a kinematic transformation The transformation package distinguishes between the following axis types: ● Linear axis: MD62601 = 1 ● Rotary axis: MD62601 = 3 Velocities and acceleration rates Separate velocities for the Cartesian movement components are introduced for axes that are traversed with G00 and active transformation.
  • Page 653: Descriptions Of Kinematics

    TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Orientation angle acceleration rates The acceleration rates for individual directions of orientation for axis traversal with G00 can be preset in machine data: ● MD62632 $MC_TRAFO6_ACCORI[i] (orientation angle acceleration rates [no.]: 0...2) –...
  • Page 654 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3. Compare the basic axes with the basic axes contained in the handling transformation package. → Enter the basic axis identifier in machine data: MD62603 $MC_TRAFO6_MAIN_AXES (basic axis identifier) 4.
  • Page 655 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3-axis CC kinematics Figure 17-12 3-axis CC kinematics Table 17-4 Configuration data for 3-axis CC kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 1, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [2, 1, 3, 4, 5, 6]...
  • Page 656 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3-axis SC kinematics Figure 17-13 3-axis SC kinematics Table 17-5 Configuration data for 3-axis SC kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [1, 1, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6]...
  • Page 657 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3-axis CS kinematic Figure 17-14 3-axis CS kinematic Table 17-6 Configuration data for 3-axis CS kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 1, 1, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6]...
  • Page 658 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Articulated-arm kinematics 3-axis NR kinematics Figure 17-15 3-axis NR kinematics Table 17-7 Configuration data 3-axis NR kinematic Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ...
  • Page 659 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3-axis RR kinematics Figure 17-16 3-axis RR kinematics Table 17-8 Configuration data for 3-axis RR kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 1, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6]...
  • Page 660: 4-Axis Kinematics

    TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 3-axis NN kinematics Figure 17-17 3-axis NN kinematics Table 17-9 Configuration data for 3-axis NN kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6]...
  • Page 661 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Restrictions The following restrictions apply to 4-axis kinematics: The frame T_FL_WP is subject to the following condition: ● MD62611 $MC_TRAFO6_TFLWP_RPY = [ 0.0, 90.0, 0.0 ] (frame between wrist point and flange (rotation component)) ●...
  • Page 662 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 11.Define frame T_IRO_RO and enter the offset in the machine data: MD62612 $MC_TRAFO6_TIRORO_POS (frame between base center point and internal system (position component)) Enter the rotation in the machine data: MD62613 $MC_TRAFO6_TIRORO_RPY (frame between base center point and internal system (rotation component)) 12.Specification of frame T_X3_P3 to attach wrist axis.
  • Page 663 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Table 17-10 Configuration data for 4-axis CC kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6_AXES_TYPE [3, 1, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [2, 1, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR)
  • Page 664 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Table 17-11 Configuration data for 4-axis SC kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6_AXES_TYPE [1, 1, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR...
  • Page 665 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Table 17-12 Configuration data for 4-axis CS kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6_AXES_TYPE [3, 1, 1, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR...
  • Page 666: 5-Axis Kinematics

    TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Table 17-13 Configuration data 4-axis NR kinematic Machine data Value MD62600 $MC_TRAFO6_ KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR...
  • Page 667 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 4. There are restrictions for the tool as far as 5-axis Scara kinematics are concerned: – 4. Axis perpendicular to the 3rd axis: 1-dimensional tool is possible [X, 0.0, 0.0] 5.
  • Page 668 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 12.Specification of frame T_X3_P3 to attach hand. The offset is entered in machine data: MD62608 $MC_TRAFO6_TX3P3_POS (attachment of hand (position component)) The rotation is entered in the machine data: MD62609 $MC_TRAFO6_TX3P3_RPY (attachment of hand (rotation component)) 13.Specification of wrist axes parameters.
  • Page 669 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Machine data Value MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6 _AXES_TYPE [3, 1, 3, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [2, 1, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR [1, 1, 1, 1, 1, 1] MD62617 $MC_TRAFO6_MAMES [0.0, 0.0, 0.0, 0.0, 0.0, 0.0]...
  • Page 670: 6-Axis Kinematics

    TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Table 17-15 Configuration data 5-axis NR kinematic Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62606 $MC_TRAFO6_A4PAR MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, 3, 3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR...
  • Page 671: Special Kinematics

    TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics 17.5.5 Special kinematics MD62602 $MC_TRAFO6_SPECIAL_KIN (special kinematic type) Special kinematics are kinematics that are not directly included in the building block system of the Handling transformation package. They are frequently missing a degree of freedom or are characterized by mechanical links between the axes or with the tool.
  • Page 672 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Machine data Value MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR [1, 1, 1, 1, 1, 1] MD62617 $MC_TRAFO6_MAMES [0.0, 0.0, 0.0, 0.0, 0.0, 0.0] MD62607 $MC_TRAFO6_MAIN_LENGTH_AB [400.0, 500.0] MD62612 $MC_TRAFO6_TIRORO_POS [0.0, 0.0, 300.0]...
  • Page 673 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Machine data Value MD62618 $MC_TRAFO6_AXES_DIR [1, 1, 1, 1, 1, 1] MD62617 $MC_TRAFO6_MAMES [0.0, 0.0, 0.0, 0.0, 0.0, 0.0] MD62607 $MC_TRAFO6_MAIN_LENGTH_AB [0.0, 0.0] MD62612 $MC_TRAFO6_TIRORO_POS [0.0, 0.0, 400.0] MD62613 $MC_TRAFO6_TIRORO_RPY [0.0, 0.0, 0.0] MD62608 $MC_TRAFO6_TX3P3_POS...
  • Page 674 TE4: Handling transformation package - 840D sl only 17.5 Descriptions of kinematics Machine data Value MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, 1, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR [1, 1, 1, 1, 1, 1] MD62617 $MC_TRAFO6_MAMES [0.0, 0.0, 0.0, 0.0, 0.0, 0.0] MD62607 $MC_TRAFO6_MAIN_LENGTH_AB...
  • Page 675: Tool Orientation

    TE4: Handling transformation package - 840D sl only 17.6 Tool orientation Table 17-19 Configuration data for special 2-axis NR kinematics Machine data Value MD62600 $MC_TRAFO6_KINCLASS MD62602 $MC_TRAFO6_SPECIAL_KIN 5 (8) MD62605 $MC_TRAFO6_NUM_AXES MD62603 $MC_TRAFO6_MAIN_AXES MD62604 $MC_TRAFO6_WRIST_AXES MD62601 $MC_TRAFO6_AXES_TYPE [3, 3, ...] MD62620 $MC_TRAFO6_AXIS_SEQ [1, 2, 3, 4, 5, 6] MD62618 $MC_TRAFO6_AXES_DIR...
  • Page 676 TE4: Handling transformation package - 840D sl only 17.6 Tool orientation Machine data Identifiers of the Euler angles The identifiers with which the Euler angles are programmed in the NC program can be set using: MD10620 $MN_EULER_ANGLE_NAME_TAB (name of Euler angles) Standard identifier: "A2", "B2", "C2"...
  • Page 677: Orientation Programming For 4-Axis Kinematics

    TE4: Handling transformation package - 840D sl only 17.6 Tool orientation The coordinate system in which motion is executed is defined using the ORIWKS and ORIMKS commands. Syntax ORIWKS ORIMKS Meaning Tool orientation in the workpiece coordinate system (WCS) ORIWKS: In the case of a change in orientation with the tool tip at a fixed point in space, the tool moves along a large arc on the plane extending from the start vector to the end vector.
  • Page 678: Orientation Programming For 5-Axis Kinematics

    TE4: Handling transformation package - 840D sl only 17.6 Tool orientation Figure 17-29 Orientation angle for 4-axis kinematic 17.6.2 Orientation programming for 5-axis kinematics Tool orientation for 5-axis kinematics For 5-axis kinematics, when programming via orientation vector, it is assumed that the orientation vector corresponds to the x component of the tool.
  • Page 679: Singular Positions And How They Are Handled

    TE4: Handling transformation package - 840D sl only 17.7 Singular positions and how they are handled Figure 17-30 Orientation angle for 5-axis kinematic The following machine data can be used to set the flange coordinate system on the user side so that Z can be set as tool direction for 5-axis kinematics: MD62636 $MC_TRAFO6_TFL_EXT_RPY (adaptation of the flange coordinate system) The following machine data can be used to specify whether the tool direction should be set...
  • Page 680: Call And Application Of The Transformation

    TE4: Handling transformation package - 840D sl only 17.8 Call and application of the transformation Singular positions ● A singular position is, for example, characterized by the fact that the fifth axis is positioned at 0°. In this case, the singular position does not depend on a specified orientation. The fourth axis is not specified in this position, i.e., the fourth axis does not have any influence on the position or the orientation.
  • Page 681: Actual Value Display

    TE4: Handling transformation package - 840D sl only 17.9 Actual value display Deactivating The currently active transformation is deactivated by means of TRAFOOF or TRAFOOF(). Note When the "Handling transformation package" transformation is deactivated, no preprocessing stop and no synchronization of the preprocessing with the main run is performed. RESET/end of program The control behavior in terms of transformation following run-up, end of program or RESET depends on machine data:...
  • Page 682: Tool Programming

    TE4: Handling transformation package - 840D sl only 17.10 Tool programming 17.10 Tool programming Meaning The tool lengths are specified in relation to the flange coordinate system. Only 3-dimensional tool compensations are possible. Depending on the kinematic type, there are additional tool restrictions for 5-axis and 4-axis kinematics.
  • Page 683: Cartesian Ptp Travel With Handling Transformation Package

    TE4: Handling transformation package - 840D sl only 17.12 Startup 17.11 Cartesian PTP travel with handling transformation package It is possible to use the "Cartesian PTP travel" function with the handling transformation package (see Section "Cartesian PTP travel (Page 38)"). For this, the following machine data must be set: MD24100 $MC_TRAFO_TYPE_1 = 4100 (definition of transformation 1 in the channel) References...
  • Page 684 TE4: Handling transformation package - 840D sl only 17.12 Startup Parameterizing transformation 1. Enter the transformation type into the machine data: – MD24100 $MC_TRAFO_TYPE_1 = 4100 (definition of transformation 1 in the channel) 2. Enter the assignment of the channel axes involved in the transformation in the machine data (axis numbers, starting at 1): –...
  • Page 685: Supplementary Conditions

    TE4: Handling transformation package - 840D sl only 17.13 Supplementary conditions 12.Enter the position in relation to the calibration point in the machine data: – MD62617 $MC_TRAFO6_MAMES (offset between mathematical and mechanical zero points) 13.Enter the Cartesian velocities and acceleration rates. 17.13 Supplementary conditions Clearance control...
  • Page 686: Data Lists

    TE4: Handling transformation package - 840D sl only 17.14 Data lists Singularities A pole cannot be directly passed through when a transformation is active. NOTICE Traversing close to the pole When traversing close to the pole, the axes involved can be overloaded. This is because the control does not automatically reduce or limit the feedrate, acceleration or jerk.
  • Page 687 TE4: Handling transformation package - 840D sl only 17.14 Data lists Number Identifier: $MC_ Description 62602 TRAFO6_SPECIAL_KIN Special kinematic type 62603 TRAFO6_MAIN_AXES Basic axis identifier 62604 TRAFO6_WRIST_AXES Wrist axis identifier 62605 TRAFO6_NUM_AXES Number of transformed axes 62606 TRAFO6_A4PAR Axis 4 is parallel/anti-parallel to last basic axis 62607 TRAFO6_MAIN_LENGTH_AB Basic axis lengths A and B...
  • Page 688: Signals

    TE4: Handling transformation package - 840D sl only 17.14 Data lists 17.14.2 Signals 17.14.2.1 Signals from channel DB number Byte.bit Description 21, … 29.4 Activate PTP traversal 21, … 33.6 Transformation active 21, … Number of active G function of G function group 25 (ORIWKS, ORIMKS, ORIPATH) 21, …...
  • Page 689: Te6: Mcs Coupling - 840D Sl Only

    TE6: MCS coupling - 840D sl only 18.1 Brief description If there are two or more separately traversable machining heads on a machine tool and a transformation is required for machining, the orientation axes of the machining heads cannot be coupled via the standard coupling functions COPON, TRAILON. The coupling is performed in the workpiece coordinate system (WCS).
  • Page 690: Description Of Mcs Coupling Functions

    TE6: MCS coupling - 840D sl only 18.2 Description of MCS coupling functions The following functions are not possible for a CC_Slave: ● To be a PLC axis ● To be a command axis ● To be traversed separately from the CC_Master in JOG mode Tolerance window When the coupling is active, the actual values of the CC_Master and CC_Slave are monitored for compliance with a parameterizable tolerance window.
  • Page 691: Switching The Coupling On/Off

    TE6: MCS coupling - 840D sl only 18.2 Description of MCS coupling functions Requirements ● The CC_Master and CC_Slave axes must be either both rotary axes or both linear axes. ● Spindles cannot be coupled by this function. ● Neither the CC_Master nor CC_Slave axis may be an exchange axis ($MA_MASTER_CHAN[AXn]=0) 18.2.2 Switching the coupling ON/OFF...
  • Page 692: Tolerance Window

    TE6: MCS coupling - 840D sl only 18.2 Description of MCS coupling functions Boundary condition A coupling can only be switched on or off when the axes are at standstill. 18.2.3 Tolerance window A monitoring window is specified via axial machine data: MD63541 $MA_CC_POSITION_TOL (monitoring window) The absolute difference between the actual values of CC_Slave axis and CC_Master axis must never be greater than this value.
  • Page 693: Description Of Collision Protection

    TE6: MCS coupling - 840D sl only 18.3 Description of collision protection 18.3 Description of collision protection 18.3.1 Defining protection pairs A ProtecSlave axis (PSlave) is matched to its ProtecMaster (PMaster) axis via the following axial machine data: MD63542 $MA_CC_PROTECT_MASTER (specifies the PMaster axis assigned to a PSlave axis) The protection pairs can thus be defined independently of the coupling pairs.
  • Page 694: Configuring Example

    TE6: MCS coupling - 840D sl only 18.3 Description of collision protection An alarm is output as soon as the axes reach zero speed. WARNING Risk of collision when starting If the axes are forced to brake, the positions displayed in the workpiece coordinate system are incorrect! These are not re-synchronized again until a system RESET.
  • Page 695: User-Specific Configurations

    TE6: MCS coupling - 840D sl only 18.4 User-specific configurations Note Since the collision protection function extrapolates the target positions from the "current velocity + maximum acceleration (or +20%)", the monitoring alarm may be activated unexpectedly at reduced acceleration rates: Example: PMaster = X, PSlave = X2, $MA_CC_COLLISION_WIN = 10mm Starting point in part program: X=0.0 X2=20.0...
  • Page 696: Special Operating States

    TE6: MCS coupling - 840D sl only 18.5 Special operating states 18.5 Special operating states Reset The couplings can remain active after a RESET. Reorg No non-standard functionalities. Block search During a block search, the last block containing an OEM-specific language command is always stored and then output with the last action block.
  • Page 697: Boundary Conditions

    TE6: MCS coupling - 840D sl only 18.7 Data lists Single block There are no nonstandard functionalities. 18.6 Boundary conditions Validity The function is configured only for the first channel. Braking behavior Braking behavior at the SW limit with path axes The programmable acceleration factor ACC for braking at the SW limit corresponds to the path axes.
  • Page 698: Axis/Spindlespecific Machine Data

    TE6: MCS coupling - 840D sl only 18.7 Data lists 18.7.1.2 Axis/spindlespecific machine data Number Identifier: $MA_ Description 63540 CC_MASTER_AXIS Specifies the CC_Master axis assigned to a CC_Slave axis. 63541 CC_POSITION_TOL Monitoring window 63542 CC_PROTEC_MASTER Specifies the PMaster axis assigned to a PSlave axis. 63543 CC_PROTEC_OPTIONS 63544...
  • Page 699: Te7: Continue Machining At The Contour (Retrace Support) - 840D Sl Only

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.1 Brief description Function The "Continue machining - Retrace support (RESU)" technological function supports the retracing of uncompleted 2-dimensional machining processes such as laser cutting, water jet cutting, etc. In the event of a fault during the machining process, e.g.
  • Page 700: Function Description

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description compensation) are mapped as straight lines through the start and end points of the corresponding contour element, thereby preventing precise retracing of contours. Function code The code for the "Continue Machining - Retrace support"...
  • Page 701 TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description Continue machining might be required for example in a laser cutting application if the laser is lost during the machining operation and machining needs to resume at the point at which it was interrupted.
  • Page 702: Definition Of Terms

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description 19.2.2 Definition of terms Interruption point The interruption point is the point of the contour at which the traversing movement comes to a standstill following an NC stop and reverse travel is activated. Program continuation point The program continuation point is the point of the contour at which reverse travel terminates and program continuation is activated.
  • Page 703 TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description 5. Select forward travel (optional): For forward travel, reverse travel must be deselected via PLC interface signal: DB21, … DBX0.1 = 0 6. Forward travel (optional): The contour is traversed in the RESU working plane in the forward direction with NC start.
  • Page 704 TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description Signal chart for interface signals The principle sequence of the RESU function is illustrated in the following figure as a signal chart of the interface signals involved: ①...
  • Page 705: Maximum Retraceable Contour Area

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.2 Function description 19.2.4 Maximum retraceable contour area In multiple machining continuation within a contour area, the reverse travel on the contour is always possible only up to the last machining continuation point (W). In first-time reverse travel after RESU start, reverse travel up to the start of the contour range is possible.
  • Page 706: Startup

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.3 Startup 19.3 Startup 19.3.1 Activation Before commissioning the technological function, ensure that the corresponding compile cycle has been loaded and activated (see also Section "TE01: Installation and activation of loadable compile cycles (Page 549)").
  • Page 707: Memory Configuration: Heap Memory

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.3 Startup 19.3.4 Memory configuration: Heap memory Memory requirements RESU requires compile cycles heap memory for the following function-specific buffers: ● Block buffer The larger the block buffer (see "Figure 19-6 RESU-specific part programs (Page 713)"), the larger the number of part program blocks that can be traversed backwards.
  • Page 708: Resu Main Program Memory Area

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.3 Startup The size of the block buffer is adjusted via the machine data: MD62571 $MC_RESU_RING_BUFFER_SIZE Default setting: MD62571 $MC_RESU_RING_BUFFER_SIZE = 1000 RESU portion of the total heap memory The RESU portion of the total heap memory that can be used for compile cycles is set via the machine data: MD62572 $MC_RESU_SHARE_OF_CC_HEAP_MEM...
  • Page 709: Storage Of The Resu Subroutines

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.3 Startup Storage in the static NC memory If the RESU main program is created in the static memory area of the NC, it is retained even after a POWER OFF. However, since RESU regenerates the RESU main program every time the retrace support function is used, this parameter setting is not recommended.
  • Page 710: Plc User Program

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.3 Startup The following machine data must be set for the start enable for the RESU-specific ASUB CC_RESU_ASUP.SPF while the channel is in the NC STOP state: MD11602 $MN_ASUP_START_MASK, bit 0 = 1 (ignore stop reason for ASUB) MD11604 $MN_ASUP_START_PRIO_LEVEL = 1 (priorities from which MD11602 is effective) 19.3.8...
  • Page 711: Programming

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.4 Programming DB21, … DBX0.1 // IF "Forward/Reverse" == 1 DB21, … DBX0.2 // THEN "Start retrace support" = 0 19.4 Programming 19.4.1 RESU Start/Stop/Reset (CC_PREPRE) Start / stop / reset / of RESU is done with the program instruction: CC_PREPRE (Prepare Retrace) Programming Syntax:...
  • Page 712: Resu-Specific Part Programs

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.5 RESU-specific part programs RESET response In reset events: ● NCK POWER ON RESET (warm start) ● NCK Reset ● End of program (M30) CC_PREPRE(-1) is executed implicitly. Error messages The following programming errors are detected and displayed with alarms: ●...
  • Page 713: Main Program (Cc_Resu.mpf)

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.5 RESU-specific part programs Figure 19-6 RESU-specific part programs 19.5.2 Main program (CC_RESU.MPF) Function In addition to the calls for the RESU-specific subprograms, the RESU main program "CC_RESU.MPF" contains the traversing blocks generated from the traversing blocks logged in the block buffer for reverse/forward travel along the contour.
  • Page 714: Ini Program (Cc_Resu_Ini.spf)

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.5 RESU-specific part programs If the RESU main program is created in the static user memory, the following system alarm appears at the same time as the RESU alarm: Alarm 6500 "NC memory full"...
  • Page 715 TE7: Continue machining at the contour (retrace support) - 840D sl only 19.5 RESU-specific part programs G71 G90 G500 T0 G40 F200 ;system frames that are present are deactivated ;actual value and scratching if $MC_MM_SYSTEM_FRAME_MASK B_AND 'H01' $P_SETFRAME = ctrans() endif ;external zero point offset if $MC_MM_SYSTEM_FRAME_MASK B_AND 'H02'...
  • Page 716: End Program (Cc_Resu_End.spf)

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.5 RESU-specific part programs 19.5.4 END program (CC_RESU_END.SPF) Function The task of the RESU-specific subroutine "CC_RESU_END.SPF" is to stop reverse travel once the end of the retraceable contour is reached. If the RESU function is parameterized appropriately, this scenario will not arise under normal circumstances.
  • Page 717: Resu Asub (Cc_Resu_Asup.spf)

    ASUB is initiated if the following RESU interface signal is switched over in the NC stop state: DB21, … DBX0.1 (Forward/Reverse) Program structure CC_RESU_ASUP.SPF has the following content: PROC CC_RESU_ASUP ; siemens system asub - do not change G4 F0.001 REPOSA Note CC_RESU_ASUP.SPF must not be changed.
  • Page 718: Retrace Support

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.6 Retrace support 19.6 Retrace support 19.6.1 General Retrace support refers to the entire operation from initiation of retracing through the interface signal DB21, … DBX0.2 = 1 (start machining continuation) up to to the continuation of the part program processing of the programmed contour.
  • Page 719: Reposition

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.6 Retrace support All part program instructions which are not executed in the action block but are required for retrace support in the part program must be entered manually in the RESU-specific retrace support ASUB CC_RESU_BS_ASUP.SPF, e.g.: ●...
  • Page 720: Temporal Conditions Concerning Nc Start

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.6 Retrace support Geometry axes In the approach block, the geometry axes of the RESU working plane (e.g. 1st and 2nd geometry axes of the channel) traverse the shortest route along the contour to the program continuation point.
  • Page 721: Block Search From Last Main Block

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.6 Retrace support 19.6.5 Block search from last main block The block search with calculation on the contour executed within the framework of the machining continuation via the use of the most powerful NCU in very large part programs can itself lead to computation times of several minutes up to the reaching of the target block.
  • Page 722: Function-Specific Display Data

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.7 Function-specific display data Supplementary conditions In order that a new retrace support operation can take place following a retrace support operation with block search from last main block, the RESU start CC_PREPRE(1) must be programmed in the retrace support ASUB "CC_RESU_BS_ASUP.SPF".
  • Page 723: Function-Specific Alarm Texts

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.9 Boundary conditions The GUD variable is not displayed on the operator panel. Note The new GUD variable, which is already being displayed, will be detected by the RESU function and supplied with an up-to-date value only following an NCK POWER ON Reset.
  • Page 724: Continue Machining Within Program Loops

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.9 Boundary conditions 19.9.1.2 Continue machining within program loops In NC high-level language, program loops can be programmed using: ● LOOP ENDLOOP ● FOR ENDFOR ● WHILE ENDWHILE ●...
  • Page 725: Supplementary Conditions For Standard Functions

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.9 Boundary conditions 19.9.2 Supplementary conditions for standard functions 19.9.2.1 Axis replacement As long as RESU is active, the two geometry axes of the RESU working plane (e.g. the 1st and 2nd geometry axes of the channel) must not be transferred to another channel via axis replacement ( RELEASE(x)/GET(x)).
  • Page 726: Block Search

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.9 Boundary conditions 19.9.2.4 Block search Block search with calculation RESU is subject to the following constraints in the context of the "block search with calculation (on contour/at end of block") standard function: ●...
  • Page 727: Frames

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.9 Boundary conditions A full description of the compensations can be found in: References: Function Manual, Extended Functions; Compensations (K3) 19.9.2.7 Frames RESU can be used in conjunction with frames. However, as the traversing movements of the two geometry axes of the RESU working plane are recorded in the basic coordinate system (BCS) and therefore after the frames have been taken into account, the frame offsets must be deactivated during retrace support (reverse /...
  • Page 728: Data Lists

    TE7: Continue machining at the contour (retrace support) - 840D sl only 19.10 Data lists 19.10 Data lists 19.10.1 Machine data 19.10.1.1 General machine data Number Identifier: $MN_ Meaning 11602 ASUP_START_MASK Ignore stop reasons if an ASUB is running. 11604 ASUP_START_PRIO_LEVEL Defines the ASUB priority from which MD11602 is effec‐...
  • Page 729: Te8: Cycle-Independent Path-Synchronous Switching Signal Output - 840D Sl Only

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.1 Brief description Function The "Cycle-independent path synchronized switching signal output" technological function serves the purpose of switching time-critical, position-based machining processes on and off quickly, e.g. high speed laser cutting (HSLC; High Speed Laser Cutting). The switching signal output can be block-related or path length-related: ●...
  • Page 730: Functional Description

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description 20.2 Functional description 20.2.1 General Note The functionality is described with examples, with the help of the "High speed laser cutting technology (HSLC, High Speed Laser Cutting). 20.2.2 Calculating the switching positions 20.2.2.1...
  • Page 731 TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description The following block positions function as switching positions: ● Position X30 for G0-edge change from N10 to N20 ● Position X100 for G0-edge change from N30 to N40 Freely programmable velocity threshold value as switching criterion A freely programmable velocity threshold value is used to define the setpoint velocity programmed in the part program block at and above which the switching signal is activated/...
  • Page 732: Path Length-Related Switching Signal Output

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description The following block positions function as switching positions: ● Position X30 for edge change from N10 to N20 ● Position X70 for edge change from N20 to N30 Note G0 always deactivates the switching signal, regardless of the threshold value.
  • Page 733: Calculating The Switching Instants

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description 20.2.3 Calculating the switching instants In order for the switching to be as precise as possible at the switching positions calculated, the control calculates the positional difference between the actual position of the geometry axes involved and the switching difference in every position controller cycle.
  • Page 734: Approaching Switching Position

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description Value below minimum switching position distance For path length-related switching signal output, the value may fall below the minimum switching position distance, e.g. due to: ● Increase in feed rate ●...
  • Page 735: Programmed Switching Position Offset

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.2 Functional description 20.2.6 Programmed switching position offset Programmed switching position offset For block-related switching signal output, a positional offset of the switching position can be programmed : ● Offset distance negative = lead With a negative offset distance, the switching position is offset before the set point position programmed in the part program block.
  • Page 736: Startup

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.3 Startup 20.3 Startup 20.3.1 Activation Before commissioning the technological function, ensure that the corresponding compile cycle has been loaded and activated (see also Section "TE01: Installation and activation of loadable compile cycles (Page 549)").
  • Page 737: Parameterizing The Switching Signal

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.3 Startup 20.3.4 Parameterizing the switching signal Output number of the switching signal Once the compile cycle has started up, the following function-specific machine data appears in the channel-specific machine data: MD62560 $MC_FASTON_NUM_DIG_OUTPUT (number of the digital output of the switching signal) The number n of the on-board digital output through which the switching signal is to be output,...
  • Page 738: Programming

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.4 Programming Changing the default setting For a deviating machine configuration (e.g. definition of a third geometry axis), the default setting can be adjusted via the following machine data: MD60948 $MN_CC_ACTIVE_IN_CHAN_HSCL[1] Value Meaning $MN_CC_ACTIVE_IN_CHAN_HSCL[1]='H3'...
  • Page 739: Activating The Path Length-Related Switching Signal Output (Cc_Faston_Cont)

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.4 Programming Programming example Programming Comment DEF REAL DIFFON= -0.08 ; Length* of the offset distance for activating the switching signal = -0.08 DEF REAL DIFFOFF= 0.08 ; Length* of the offset distance for deactivating the switching signal = 0.08 DEF REAL FEEDTOSWITCH= 20000...
  • Page 740: Deactivation (Cc_Fastoff)

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.5 Function-specific alarm texts Programming example Programming Comment DEF REAL PATH_DISTANCE_ON = 0.5 ; Length* of the stretch section with machining = 0.5 DEF REAL PATH_DISTANCE_OFF = 1.0 ; Length* of the stretch section without machining = 1.0 CC_FASTON_CONT (PATH_DISTANCE_ON, PATH_DISTANCE_OFF) ;...
  • Page 741: Supplementary Conditions

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.6 Supplementary conditions 20.6 Supplementary conditions 20.6.1 Block search Switching signal output for block search If a block search is on a part program block which lies after a CC_FASTON() procedure call for activating the technology function, then the switching signal is activated with the next traversing motion.
  • Page 742: Transformations

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.6 Supplementary conditions Figure 20-5 Switching signal after block search Suppressing the switching signal output The user (machine manufacturer) must take appropriate measures, e.g. disable the switching signal, in order to suppress the activation of the switching signal in the REPOS block in the constellation described above.
  • Page 743: Compensations

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.6 Supplementary conditions 20.6.3 Compensations The following compensations are considered while calculating the switching positions: ● Temperature compensation ● Sag compensation A description of the compensations can be found in: Reference: Function Manual, Extended Functions;...
  • Page 744: Software Cams

    TE8: Cycle-independent path-synchronous switching signal output - 840D sl only 20.7 Data lists 20.6.6 Software cams Because the hardware timer is also used for the "software cam" function, it is not possible to use the "clock-independent switching signal output" function with software cams at the same time.
  • Page 745: Te9: Axis Pair Collision Protection

    TE9: Axis pair collision protection 21.1 Brief description 21.1.1 Brief description Function The "axis pair collision protection" function enables machine axes, which are arranged on the same guide element of a machine, to be monitored in pairs to ensure that no collisions occur and that the maximum distance between the two axes is not exceeded.
  • Page 746: Startup

    TE9: Axis pair collision protection 21.3 Startup Distance monitoring If the offset vector is selected accordingly, the function can also be used to monitor the maximum distance between the two machine axes (maximum distance vector). Figure 21-1 Basic design Monitoring status The actual status of an axis pair can be read out of _PROTECT_STATUS (Page 751), optionally defined global user variables in the NC program (GUD).
  • Page 747: Definition Of An Axis Pair

    TE9: Axis pair collision protection 21.3 Startup Channel-specific activation The function must be activated in the following NC channels: ● Always in the 1st channel of the NC ● All channels that are assigned – with the appropriate machine data parameterization – to machine axes to be monitored ●...
  • Page 748: Retraction Direction

    TE9: Axis pair collision protection 21.3 Startup 21.3.4 Retraction direction The direction of travel for retracting the corresponding machine axis is entered in the following item of machine data: MD61517 $MN_CC_PROTECT_SAFE_DIR[<axis pair>] = <yyxx> <yyxx> Meaning Retraction direction for the 1st axis pair Retraction direction for the 2nd axis pair Retraction in the positive direction of travel of the machine axis: xx or yy >...
  • Page 749: Protection Window

    TE9: Axis pair collision protection 21.3 Startup Note Machine data change Machine data $MN_...OFFSET[<axis pair>] may only be made changed for an axis pair, if the function for the axis pair is not active ($MN_...PAIRS[<axis pair>] = 0). 21.3.6 Protection window Using the machine data, the minimum clearance is defined, which the axes of an axis pair must not fall below: MD61519 $MN_CC_PROTECT_WINDOW[<axis pair>] = <minimum clearance>...
  • Page 750: Protection Window Extension

    TE9: Axis pair collision protection 21.3 Startup 21.3.8 Protection window extension The protection window can be expanded using the protection window (Page 749) extension function. The protection window extension is set in the following item of machine data: MD61533 $MN_CC_PROTECT_WINDOW_EXTENSION[<axis pair>] = <extension> The resulting effective protection window of an axis pair is as follows: Effective protection window = $MN_...WINDOW (protection window) + $MN_...WINDOW_EXTENSION (protection window extension)
  • Page 751: Monitoring Status (Gud)

    TE9: Axis pair collision protection 21.3 Startup 21.3.11 Monitoring status (GUD) The actual status of one axis pair is displayed using the global user variable _PROTECT_STATUS. The system variable is not available as standard default setting. When required, it must be defined in the GUD.DEF definition window. Definition DEF NCK INT _PROTECT_STATUS[<number of parameterized axis pairs>...
  • Page 752: Limitations And Constraints

    TE9: Axis pair collision protection 21.4 Limitations and constraints Bit n → (n+1)th machine axis, with n = 0, 1, 2, ... Bit n Meaning Machine axis (n+1) is not braked Machine axis (n+1) is braked 21.4 Limitations and constraints 21.4.1 Precedence of function-specific acceleration The function only uses the function-specific acceleration of the machine axes MD63514...
  • Page 753 TE9: Axis pair collision protection 21.4 Limitations and constraints Assigning parameters for the NC Logical machine axes: axis numbers 1 and 13 ● MD10002 $MN_AXCONF_LOGIC_MACHAX_TAB [ 0 ] = "CT1_SL1" (log. mach. axis 1) ● MD10002 $MN_AXCONF_LOGIC_MACHAX_TAB [ 12 ] = "CT1_SL2" (log. mach. axis 13) Axis container CT1, slot 1 and slot 2 ●...
  • Page 754: Interpolatory Couplings

    TE9: Axis pair collision protection 21.5 Examples 21.4.3 Interpolatory couplings Assumption 1. A machine axis is part of an interpolatory coupling, e.g.: – Generic coupling (CP) – Coupled motion(TRAIL) – Master value coupling (LEAD) – Electronic gear (EG) – Synchronous spindle(COUP) 2.
  • Page 755 TE9: Axis pair collision protection 21.5 Examples Parameter assignment: Protection function 1 Axis pair: 1st machine axis A3, 2nd machine axis A1 ● MD61516 $MN_CC_PROTECT_PAIRS[0] = 01 03 Retraction direction: A1 in negative direction, A3 in positive direction ● MD61517 $MN_CC_PROTECT_SAFE_DIR[0] = 00 01 Offset vector from machine coordinate system machine_A1 to machine_A3 with reference to machine_A3 ●...
  • Page 756: Collision Protection And Distance Limiter

    TE9: Axis pair collision protection 21.5 Examples 21.5.2 Collision protection and distance limiter The figure shows the arrangement of two machine axes, the offset and orientation of the machine coordinate systems (machine), and the minimum and maximum distance vectors. Figure 21-3 Collision protection and distance limiter for one axis pair Parameter assignment: Protection function 1 - Collision protection Axis pair: 1st machine axis A1, 2nd machine axis A3...
  • Page 757: Data Lists

    TE9: Axis pair collision protection 21.6 Data lists Parameter assignment: Protection function 2 - Distance limiter Axis pair: 1st machine axis A1, 2nd machine axis A3 ● MD61516 $MN_CC_PROTECT_PAIRS[1] = 03 01 Retraction direction: A1 in positive direction, A3 in negative direction ●...
  • Page 758: Machine Data

    TE9: Axis pair collision protection 21.6 Data lists 21.6.2 Machine data 21.6.2.1 NC-specific machine data Number Identifier: $MN_ Description 60972 CC_ACTIVE_IN_CHAN_PROT[n] Channel-specific activation of the technology function BIT0 = 1 => activation in the 1st NC channel BIT1 = 1 => activation in the 2nd NC channel, etc.
  • Page 759: V2: Preprocessing

    V2: Preprocessing 22.1 Brief description Preprocessing The programs stored in the directories for standard and user cycles can be preprocessed to reduce runtimes. Preprocessing is activated via machine data. Standard and user cycles are preprocessed when the power is switched on, i.e. as an internal control function, the part program is translated (compiled) into a binary intermediate code optimized for processing purposes.
  • Page 760 V2: Preprocessing 22.1 Brief description Memory is required for preprocessing cycles. You can optimize your memory in two ways: ● The program to be executed can be shortened with the command DISPLOF (display off). ● MD10700 $MN_PREPROCESSING_LEVEL has been expanded by bit 2 and 3. This allows selective cycle preprocessing of the individual directories (e.g.
  • Page 761: Program Handling

    This is a sensible setting when no cycles with call parameters are used. During control power-up, the call description of the cycles is generated. All user cycles (_N_CUS_DIR directory) and Siemens cycles (_N_CST_DIR directory) with transfer parameters can be called up without external statement. Changes to the cycle-call interface do not take effect until the next POWER ON.
  • Page 762 V2: Preprocessing 22.2 Program handling Compiling In the directories for standard cycles: _N_CST_DIR, _N_CMA_DIR and user cycles: Subprograms (_SPF file extension) located in _N_CUS_DIR, and, if necessary, the subprograms marked with PREPRO are compiled. The name of the compilation corresponds to the original cycle with extension _CYC.
  • Page 763 V2: Preprocessing 22.2 Program handling Program code Comment G1 F10 X=VARIABLE*10-56/86EX4+4*SIN(VARIABLE/3) ENDFOR ; 1 name, only for preprocessing In order to execute this program normally, the following machine data must specify at least 3 names: MD28020 $MC_MM_NUM_LUD_NAMES_TOTAL Six names are required to compile this program after POWER ON. Preprocessed programs/cycles are stored in the dynamic NC memory.
  • Page 764: Program Call

    V2: Preprocessing 22.3 Program call 22.3 Program call Overview Figure 22-1 Generation and call of preprocessed cycles without parameters Figure 22-2 Generation and call of preprocessed cycles with parameters Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 765 ● The change to an external language mode with G291 is rejected and an alarm issued. When the pre-compiled cycle is called, an explicit change is made to the Siemens language mode. ● When the subroutine is called, it is checked whether the compiled file is older than the cycle.
  • Page 766: Constraints

    V2: Preprocessing 22.4 Constraints After the errors detected during preprocessing have been corrected, preprocessing must be started again by means of an NCK POWER ON. 22.4 Constraints Availability of the "preprocessing" function The function is an option ("Program pre-processing"), which must be assigned to the hardware via the license management.
  • Page 767: Examples

    V2: Preprocessing 22.5 Examples The axes to be traversed are addressed indirectly via machine data or transferred as parameters: ● Indirect axis programming: – IF $AA_IM[AXNAME($MC_AXCONF_CHANAX_NAME_TAB[4])] > 5 ; This branch will pass through if the actual value of the 5th channel axis ;...
  • Page 768 V2: Preprocessing 22.5 Examples Program code Comment PROC UP2 N2000 DEF INT VARIABLE, ARRAY[2] N2010 IF $AN_NCK_Version < 3.4 N2020 SETAL(61000) N2030 ENDIF N2040 START: N2050 FOR VARIABLE = 1 TO 5 N2060 G1 F1000 X=VARIABLE*10-56/86EX4+4*SIN(VARIABLE/3) N2070 ENDFOR N2080 M17 PROC MAIN N10 G0 X0 Y0 Z0 N20 UP1...
  • Page 769: Preprocessing In The Dynamic Nc Memory

    V2: Preprocessing 22.6 Data lists 22.5.2 Preprocessing in the dynamic NC memory Machine data for preprocessing only in the dynamic NC memory with selective selection: Program code Comment ; Bit 5 = 1 Selective program selection ; Bit 6 =0 no switch to ;...
  • Page 770: Channelspecific Machine Data

    V2: Preprocessing 22.6 Data lists 22.6.1.2 Channelspecific machine data Number Identifier: $MC_ Description 28010 MM_NUM_REORG_LUD_MODULES Number of blocks for local user variables for REORG (DRAM) 28020 MM_NUM_LUD_NAMES_PER_PROG Number of local user variables (DRAM) 28040 MM_LUD_VALUES_MEM Memory size for local user variables (DRAM) Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 771: W5: 3D Tool Radius Compensation - 840D Sl Only

    W5: 3D tool radius compensation - 840D sl only 23.1 Brief description 23.1.1 General Why 3D TRC? 3D tool radius compensation is used to machine contours with tools that can be controlled in their orientation independently of the tool path and shape. Note This description is based on the specifications for 2D tool radius compensation.
  • Page 772 W5: 3D tool radius compensation - 840D sl only 23.1 Brief description Circumferential milling, face milling The following diagram shows the differences between 2 D and 3D TRC with respect to circumferential milling operations. Figure 23-1 21/2D and 3D tool radius compensation The parameters for display in the "Face milling"...
  • Page 773: Machining Modes

    W5: 3D tool radius compensation - 840D sl only 23.1 Brief description Orientation With 3D TRC, a distinction must be drawn between: ● Tools with space-bound orientation ● Tools with variable orientation 23.1.2 Machining modes There are two modes for milling spatial contours: ●...
  • Page 774: Circumferential Milling

    W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling 23.2 Circumferential milling Circumferential milling The variant of circumferential milling used here is implemented through the definition of a path (directrix) and the associated orientation. In this machining mode, the tool shape is irrelevant on the path and at the outside corners.
  • Page 775: Corners For Circumferential Milling

    W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Figure 23-4 Insertion depth 23.2.1 Corners for circumferential milling Outside corners/inside corners Outside corners and inside corners must be treated separately. The terms inside corner and outside corner are dependent on the tool orientation. When the orientation changes at a corner, for example, the corner type may change while machining is in progress.
  • Page 776: Behavior At Outer Corners

    W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Figure 23-6 Change of corner type during machining 23.2.2 Behavior at outer corners In the same manner as 21/2D tool radius compensation procedures, a circle is inserted at outer corners for G450 and the intersection of the offset curves is approached for G451.
  • Page 777 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling When the orientation needs to be changed at outside corners, the change can be implemented in parallel to interpolation or separately from the path motion. When ORID is programmed, the inserted blocks are executed first without a path motion (blocks with changes in orientation, auxiliary function outputs, etc.).
  • Page 778 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Program code Comment N40 CUT3DC N50 ORIC ; TRC selection N60 G42 X10 Y10 N70 X60 ; Change in orientation to the for N70 N80 A3=1 B3=0 C3=1 N90Y60 ;...
  • Page 779 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling ORID If ORID is active, then all blocks between the two traversing blocks are executed at the end of the first traversing block. The circle block with constant orientation is executed immediately before the second traversing block.
  • Page 780: Behavior At Inside Corners

    W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Note The command DISC is not evaluated. G451 The intersection is determined by extending the offset curves of the two participating blocks and defining the intersection of the two blocks at the corner in the plane perpendicular to the tool orientation.
  • Page 781 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Figure 23-9 The contact points of the tool must not cross the limits of block N70 or N90 as a result of the change in orientation in block N80 Example: Program code Comment...
  • Page 782 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Figure 23-10 Path end position and change in orientation at inside corners Change in insertion depth Generally speaking, the contour elements that form an inside corner are not positioned on the plane perpendicular to the tool.
  • Page 783 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Figure 23-11 Change in insertion depth Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 784 W5: 3D tool radius compensation - 840D sl only 23.2 Circumferential milling Example of inside corners Figure 23-12 Change in orientation at an inside corner Example: Program code Comment N10 A0 B0 X0 Y0 Z0 F5000 N20 T1 D1 ; Radius=5 N30 TRAORI(1) ;...
  • Page 785: Face Milling

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling 23.3 Face milling The face milling function allows surfaces with any degree or form of curvature to be machined. In this case, the longitudinal axis of the tool and the surface normal vector are more or less parallel.
  • Page 786 W5: 3D tool radius compensation - 840D sl only 23.3 Face milling The shaft characteristics are not taken into account on any of the tool types. For this reason, the two tool types 120 (end mill) and 155 (bevel cutter), for example, have an identical machining action since only the section at the tool tip is taken into account.
  • Page 787: Orientation

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling A change in tool involving only a change in other tool data (e.g. tool length) is permitted provided that no other restrictions apply. An alarm is output if a tool is changed illegally. 23.3.2 Orientation The options for programming the orientation have been extended for 3D face milling.
  • Page 788: Compensation On Path

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling would be with other interpolation quantities such as, for example, the position of an additional synchronized axis. In addition to the usual methods of programming orientation, it is also possible to refer the tool orientation to the surface normal vector and path tangent vector using the addresses LEAD (lead or camber angle) and TILT (side angle).
  • Page 789: Corners For Face Milling

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling Figure 23-15 Change in the machining point on the tool surface close to a point in which surface normal vector and tool orientation are parallel The problem is basically solved as follows: If the angle d between the surface normal vector and tool orientation w is smaller than a limit value (machine data) δ...
  • Page 790: Behavior At Outer Corners

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling The surface normals of the two surfaces forming the edge may point in opposite directions of the overall surface (the front edge of one surface is continued on the rear edge of the second surface), see also the following Figure.
  • Page 791: Behavior At Inside Corners

    W5: 3D tool radius compensation - 840D sl only 23.3 Face milling Programming ● ORIC: Change in orientation and path movement in parallel (ORIentation Change Continuously) ● ORID: Change in orientation and path movement consecutively (ORIentation Change Discontinuously) The ORIC and ORID program commands are used to determine whether changes in orientation programmed between two blocks are executed before the inserted circle block is processed or at the same time.
  • Page 792 W5: 3D tool radius compensation - 840D sl only 23.3 Face milling Figure 23-17 Inside corner with face milling (view in direction of longitudinal axis of tool) Note The amount by which the contact points deviate from the programmed contour will generally be small since the explanatory example shown in the Figure, in which the machining point "changes"...
  • Page 793: Monitoring Of Path Curvature

    W5: 3D tool radius compensation - 840D sl only 23.4 Selection/deselection of 3D TRC A variable tool orientation in a block that is shortened owing to an inside corner is also treated in the same way as described in Subsection "Behavior at inside corners" for 3D milling, i.e. the entire change in orientation is executed in the shortened block.
  • Page 794: Selection Of 3D Trc

    W5: 3D tool radius compensation - 840D sl only 23.4 Selection/deselection of 3D TRC 23.4.1 Selection of 3D TRC CUT3DC 3D radius compensation for circumferential milling (only when 5-axis transformation is active). CUT3DFS 3D tool offset for face milling with constant orientation. The tool orientation is determined by G17 - G19 and is not influenced by frames.
  • Page 795: Deselection Of 3D Trc

    W5: 3D tool radius compensation - 840D sl only 23.5 Constraints 23.4.2 Deselection of 3D TRC Deselection The 3D tool radius compensation function is deselected in a linear block G0/G1 with geometry axes by means of G40. Example: Program code Comment N10 A0 B0 X0 Y0 Z0 F500 N20 T1 D1...
  • Page 796: Examples

    W5: 3D tool radius compensation - 840D sl only 23.6 Examples 23.6 Examples Example program for 3D circumferential milling: Program code Comment ; Definition of tool D1 $TC_DP1[1,1]=120 ; Type (end mill) $TC_DP3[1,1] = ; Length offset vector $TC_DP6[1.1] =8 ;...
  • Page 797: Data Lists

    W5: 3D tool radius compensation - 840D sl only 23.7 Data lists Program code Comment N170 A3=-2 C3=1 ; Change in orientation with ORIC N180 A3=-1 C3=1 ; Change in orientation with ORIC N190 Y-10 A4=-1 C4=3 ; New plane definition N200 X-20 Y-20 Z10 ;...
  • Page 798: Channelspecific Machine Data

    W5: 3D tool radius compensation - 840D sl only 23.7 Data lists 23.7.2 Channelspecific machine data Number Identifier: $MC_ Description 20110 RESET_MODE_MASK Definition of control basic setting after powerup and RE‐ SET / part program end 20120 TOOL_RESET_VALUE Definition of tool for which tool length compensation is selected during powerup or on reset or parts program end as a function of MD 20110 20130...
  • Page 799: W6: Path Length Evaluation - 840D Sl Only

    W6: Path length evaluation - 840D sl only 24.1 Brief description Function With the "Path length evaluation" function, the NCK specific machine axis data is made available as the system and OPI variables, with whose help it is possible to assess the strain on the machine axes and thereby make an evaluation on the state of the machine's maintenance.
  • Page 800: Data

    W6: Path length evaluation - 840D sl only 24.3 Parameterization 24.2 Data The following data is available: System variable OPI variable Meaning $AA_TRAVEL_DIST aaTravelDist Total traverse path: sum of all set position changes in MCS in [mm] or [deg.]. $AA_TRAVEL_TIME aaTravelTime Total travel time: sum of IPO clock cycles from set position changes in MCS...
  • Page 801: Examples

    W6: Path length evaluation - 840D sl only 24.4 Examples MD33060 $MA_MAINTENANCE_DATA (configuration to record maintenance data) Bit Value Activation of the following data: System variable / OPI variable ● Total traverse path: $AA_TRAVEL_DIST / aaTravelDist ● Total travel time: $AA_TRAVEL_TIME / aaTravelTime ●...
  • Page 802: Data Lists

    W6: Path length evaluation - 840D sl only 24.5 Data lists Program code Comment R10=$AA_TRAVEL_DIST[AX1] ; Total traversing distance AX1 R11=$AA_TRAVEL_TIME[AX1] ; Total traversing time AX1 R12=$AA_TRAVEL_COUNT[AX1] ; Number of traversing operations AX1 ; Traversing motion of axes ; Generating the difference: R10=R10-$AA_TRAVEL_DIST[AX1] ;...
  • Page 803: Z3: Nc/Plc Interface Signals

    Z3: NC/PLC interface signals 25.1 F2: 3 to 5-axis transformation 25.1.1 Signals from channel (DB21, ...) DB21, ... DBX29.4 activate PTP traversal Edge evaluation: Yes Signal(s) updated: Signal state 1 or edge activate PTP traversal. change 0 → 1 Signal state 0 or edge Activate CP traversal.
  • Page 804 Z3: NC/PLC interface signals 25.1 F2: 3 to 5-axis transformation DB21, … DBX317.6 PTP traversal active Signal state 0 or edge CP traversal active. change 1 → 0 Signal irrelevant for ... No handling transformations active. Additional references Function Manual, Special Functions; 3-Axis to 5-Axis Transformation (F2) DB21, …...
  • Page 805: G1: Gantry Axes

    Z3: NC/PLC interface signals 25.2 G1: Gantry axes 25.2 G1: Gantry axes 25.2.1 Signals to axis/spindle (DB31, ...) DB31, ... DBX29.4 Start gantry synchronization Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 or edge Request from PLC user program to synchronize the leading axis with the assigned synchron‐ change 0 →...
  • Page 806: Signals From Axis/Spindle (Db31

    Z3: NC/PLC interface signals 25.2 G1: Gantry axes DB31, ... DBX29.5 Disable automatic synchronization Signal irrelevant for ... Gantry synchronized axis Application example(s) The automatic synchronization process can be disabled by sending a VDI signal to the axial PLC → NC interface of the master axis. This always makes sense when the axes are not activated by default.
  • Page 807 Z3: NC/PLC interface signals 25.2 G1: Gantry axes DB31, ... DBX101.3 Gantry warning limit exceeded Application example(s) When the gantry warning limit is exceeded, the necessary measures (e.g. program interruption at block end) can be initiated by the PLC user program. Special cases, errors, ...
  • Page 808 Z3: NC/PLC interface signals 25.2 G1: Gantry axes DB31, ... DBX101.5 Gantry grouping is synchronized Signal state 0 or edge The gantry axis grouping defined with the following machine data is not synchronized: change 1 → 0 MD37100 $MA_GANTRY_AXIS_TYPE (gantry axis definition) which means that the positions of the leading and synchronized axes may not be ideally aligned (e.g.
  • Page 809: K9: Collision Avoidance

    Z3: NC/PLC interface signals 25.3 K9: Collision avoidance DB31, ... DBX101.7 Gantry axis Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 or edge The axis is defined as a gantry axis within a gantry axis grouping (see MD37100). change 0 → 1 The PLC user program can read IS "Gantry leading axis"...
  • Page 810: Signals From The Nc To The Plc (Db10)

    Z3: NC/PLC interface signals 25.3 K9: Collision avoidance DB10 DBX58.0 - 7 Collision avoidance: Deactivate protection area group Remark 1) Type of a protection area: System variable $NP_PROT_TYPE (Page 224) Note ● The protection area groups can be deactivated through the SINUMERIK Operate user interface.
  • Page 811: Signals From Channel (Db21)

    Z3: NC/PLC interface signals 25.3 K9: Collision avoidance DB10 DBX226.0 - DBX233.7 Collision avoidance: Protection area active Remark A protection area is connected with the interface signal via parameter assign‐ ment of the Bit number corresponding with the interface signal, in the system variables $NP_BIT_NO (Page 228) of the protection area.
  • Page 812: M3: Axis Couplings

    Z3: NC/PLC interface signals 25.4 M3: Axis couplings 25.4 M3: Axis couplings 25.4.1 Signals to axis (DB31, ...) DB31, … Active following axis overlay DBX26.4 Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 or edge An additional traversing motion can be overlaid on the following axis. change 0 →...
  • Page 813: R3: Extended Stop And Retract

    Z3: NC/PLC interface signals 25.5 R3: Extended stop and retract DB31, … DBX98.6 Acceleration warning threshold Signal irrelevant ... Without electronic gear. Corresponding to ... The following machine data: MD37550 $MA_EG_VEL_WARNING (threshold value velocity alarm threshold) MD32300 $MA_MAX_AX_ACCEL (axis acceleration) DB31, …...
  • Page 814: S9: Setpoint Changeover

    Z3: NC/PLC interface signals 25.6 S9: Setpoint changeover DB31, ... DBX95.2 ESR: Reaction triggered or generator operation active Edge evaluation: No Signal(s) updated: Cyclically Signal state 1 The drive signals that the configured ESR response has been triggered or generator operation is active.
  • Page 815: Signals From Axis/Spindle (Db31

    Z3: NC/PLC interface signals 25.8 TE1: Clearance control 25.6.2 Signals from axis/spindle (DB31, ...) DB31, ... DBX96.5 Status of setpoint exchange Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 The axis has taken over control of the drive. Signal state 0 The axis has relinquished control of the drive.
  • Page 816: Signals From Channel (Db21

    Z3: NC/PLC interface signals 25.8 TE1: Clearance control DB21, ... DBX1.5 CLC_Override Edge evaluation: no Signal(s) updated: Cyclic Signal state 1 or The channel-specific override DB21, … DBB4 (feed rate override) also applies to the clearance edge change 0 → 1 control Override settings <...
  • Page 817 Z3: NC/PLC interface signals 25.8 TE1: Clearance control DB21, ... DBX37.5 CLC motion at upper motion limit Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 or The traversing movement of the clearance-controlled axes based on clearance control has edge change 0 → 1 been stopped at the upper movement limit set in MD62506 $MC_CLC_SENSOR_UPPER_LIMIT (upper clearance control motion limit) or programmed with CLC_LIM(..).
  • Page 818: Te3: Speed/Torque Coupling, Master-Slave

    Z3: NC/PLC interface signals 25.9 TE3: Speed/torque coupling, master-slave 25.9 TE3: Speed/torque coupling, master-slave 25.9.1 Signals to axis/spindle (DB31, ...) DB31, ... DBX24.4 Torque compensatory controller on Edge evaluation: Yes Signal(s) updated: Cyclically Signal state 1 or edge Torque compensatory controller is to be activated. change 0 →...
  • Page 819: Te4: Transformation Package, Handling

    Z3: NC/PLC interface signals 25.10 TE4: Transformation package, handling DB31, ... DBX96.3 Master/slave coarse Edge evaluation: No Signal(s) updated: Cyclic Signal state 1 or edge The differential speed is in the range defined by the following item of machine data: change 0 →...
  • Page 820: Te6: Mcs Coupling

    Z3: NC/PLC interface signals 25.11 TE6: MCS coupling DB21, ... DBX33.6 Transformation active Edge evaluation: Yes Signal(s) updated: Signal state 1 or edge Active transformation. change 0 → 1 Signal state 0 or edge Transformation not (no longer) active. change 1 → 0 Signal irrelevant for ...
  • Page 821: Signals From Axis/Spindle (Db31

    Z3: NC/PLC interface signals 25.11 TE6: MCS coupling DB31, … DBX24.2 Deactivate or disable coupling Signal irrelevant for ... Application example(s) Evaluated only on the CC_Slave axis. DB31, … DBX24.3 Switch on collision protection Edge evaluation: Yes Signal(s) updated: Signal state 1 Collision protection ON Signal state 0 Collision protection OFF...
  • Page 822: Te7: Continue Machining At The Contour - Retrace Support

    Z3: NC/PLC interface signals 25.12 TE7: Continue machining at the contour – retrace support DB31, … DBX97.2 Activate mirroring Edge evaluation: No Signal(s) updated: Signal state 1 Mirroring active (1:-1) Signal state 0 1:1 coupling active Signal irrelevant for ... Relevant only if coupling is active (DBX97.1 = 1) Application example(s) Displayed only for the CC_Slave axis.
  • Page 823: Signals From Channel

    Z3: NC/PLC interface signals 25.12 TE7: Continue machining at the contour – retrace support DB21, ... DBX0.2 Start retrace support Edge evaluation: No Signal(s) updated: Signal state 1 Start retrace support: The original machining program is reselected and a block search is carried out to locate the program continuation point.
  • Page 824 Z3: NC/PLC interface signals 25.12 TE7: Continue machining at the contour – retrace support Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 825: Appendix

    Appendix List of abbreviations Output ADI4 (Analog drive interface for 4 axes) Adaptive Control Active Line Module Rotating induction motor Automation system ASCII American Standard Code for Information Interchange: American coding standard for the exchange of information ASIC Application-Specific Integrated Circuit: User switching circuit ASUB Asynchronous subprogram AUXFU...
  • Page 826 Appendix A.1 List of abbreviations Connector Output Certificate of License Communication Compiler Projecting Data: Configuring data of the compiler Cathode Ray Tube: picture tube Central Service Board: PLC module Control Unit Communication Processor Central Processing Unit: Central processing unit Carriage Return Clear To Send: Ready to send signal for serial data interfaces CUTCOM Cutter radius Compensation: Tool radius compensation...
  • Page 827 Appendix A.1 List of abbreviations Input/Output Encoder: Actual value encoder Compact I/O module (PLC I/O module) Electrostatic Sensitive Devices ElectroMagnetic Compatibility European standard Encoder: Actual value encoder EnDat Encoder interface EPROM Erasable Programmable Read Only Memory: Erasable, electrically programmable read-only memory ePS Network Services Services for Internet-based remote machine maintenance Designation for an absolute encoder with 2048 sine signals per revolution...
  • Page 828 Appendix A.1 List of abbreviations Abbreviation for hexadecimal number AuxF Auxiliary function Hydraulic linear drive Human Machine Interface: SINUMERIK user interface Main Spindle Drive Hardware Commissioning Interpolatory compensation Interface Module: Interconnection module Interface Module Receive: Interface module for receiving data Interface Module Send: Interface module for sending data Increment: Increment Initializing Data: Initializing data...
  • Page 829 Appendix A.1 List of abbreviations Media Access Control MAIN Main program: Main program (OB1, PLC) Megabyte Motion Control Interface MCIS Motion Control Information System Machine Control Panel: Machine control panel Machine Data Manual Data Automatic: Manual input Motor Data Set: Motor data set MSGW Message Word Machine Coordinate System...
  • Page 830 Appendix A.1 List of abbreviations PCMCIA Personal Computer Memory Card International Association: Plug-in memory card standardization PC Unit: PC box (computer unit) Programming device Parameter identification: Part of a PIV Parameter identification: Value (parameterizing part of a PPO) Programmable Logic Control: Adaptation control PROFINET PROFIBUS user organization POWER ON...
  • Page 831 Appendix A.1 List of abbreviations Request To Send: Control signal of serial data interfaces RTCP Real Time Control Protocol Synchronized Action Safe Brake Control: Safe Brake Control Single Block: Single block Subroutine: Subprogram (PLC) Setting Data System Data Block Setting Data Active: Identifier (file type) for setting data SERUPRO SEarch RUn by PROgram test: Search run by program test System Function Block...
  • Page 832 Appendix A.1 List of abbreviations Terminal Board (SINAMICS) Tool Center Point: Tool tip TCP/IP Transport Control Protocol / Internet Protocol Thin Client Unit Testing Data Active: Identifier for machine data Totally Integrated Automation Terminal Module (SINAMICS) Tool Offset: Tool offset Tool Offset Active: Identifier (file type) for tool offsets TRANSMIT Transform Milling Into Turning: Coordination transformation for milling operations on...
  • Page 833 Appendix A.1 List of abbreviations Extensible Markup Language Work Offset Active: Identifier for work offsets Status word (of drive) Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 834: Overview

    Appendix A.2 Overview Overview Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 835: Glossary

    Glossary Absolute dimensions A destination for an axis motion is defined by a dimension that refers to the origin of the currently active coordinate system. See → Incremental dimension Acceleration with jerk limitation In order to optimize the acceleration response of the machine whilst simultaneously protecting the mechanical components, it is possible to switch over in the machining program between abrupt acceleration and continuous (jerk-free) acceleration.
  • Page 836 Glossary Auxiliary functions Auxiliary functions enable → part programs to transfer → parameters to the → PLC, which then trigger reactions defined by the machine manufacturer. Axes In accordance with their functional scope, the CNC axes are subdivided into: ● Axes: Interpolating path axes ●...
  • Page 837 Glossary Baud rate Rate of data transfer (bits/s). Blank Workpiece as it is before it is machined. Block "Block" is the term given to any files required for creating and processing programs. Block search For debugging purposes or following a program abort, the "Block search" function can be used to select any location in the part program at which the program is to be started or resumed.
  • Page 838 Glossary See → NC Computerized Numerical Control: includes the components → NCK, → PLC, HMI, → COM. Component of the NC for the implementation and coordination of communication. Compensation axis Axis with a setpoint or actual value modified by the compensation value Compensation table Table containing interpolation points.
  • Page 839 Glossary Curvature The curvature k of a contour is the inverse of radius r of the nestling circle in a contour point (k = 1/r). Cycles Protected subprograms for execution of repetitive machining operations on the → workpiece. Data block 1.
  • Page 840 Glossary Editor The editor makes it possible to create, edit, extend, join, and import programs/texts/program blocks. Exact stop When an exact stop statement is programmed, the position specified in a block is approached exactly and, if necessary, very slowly. To reduce the approach time, → exact stop limits are defined for rapid traverse and feed.
  • Page 841 Glossary Frame A frame is an arithmetic rule that transforms one Cartesian coordinate system into another Cartesian coordinate system. A frame contains the following components: → zero offset, → rotation, → scaling, → mirroring. Geometry Description of a → workpiece in the → workpiece coordinate system. Geometry axis The geometry axes form the 2 or 3-dimensional →...
  • Page 842 Glossary HW Config SIMATIC S7 tool for the configuration and parameterization of hardware components within an S7 project Identifier In accordance with DIN 66025, words are supplemented using identifiers (names) for variables (arithmetic variables, system variables, user variables), subprograms, key words, and words with multiple address letters.
  • Page 843 Glossary Interpolatory compensation Mechanical deviations of the machine are compensated for by means of interpolatory compensation functions, such as → leadscrew error, sag, angularity, and temperature compensation. Interrupt routine Interrupt routines are special → subprograms that can be started by events (external signals) in the machining process.
  • Page 844 Glossary Leadscrew error compensation Compensation for the mechanical inaccuracies of a leadscrew participating in the feed. The controller uses stored deviation values for the compensation. Limit speed Maximum/minimum (spindle) speed: The maximum speed of a spindle can be limited by specifying machine data, the →...
  • Page 845 Glossary Machining channel A channel structure can be used to shorten idle times by means of parallel motion sequences, e.g. moving a loading gantry simultaneously with machining. Here, a CNC channel must be regarded as a separate CNC control system with decoding, block preparation and interpolation. Macro techniques Grouping of a set of statements under a single identifier.
  • Page 846 Glossary Mode group Axes and spindles that are technologically related can be combined into one mode group. Axes/spindles of a mode group can be controlled by one or more → channels. The same → mode type is always assigned to the channels of the mode group. Numerical Control component of the →...
  • Page 847 Glossary Overall reset In the event of an overall reset, the following memories of the → CPU are deleted: ● → Working memory ● Read/write area of → load memory ● → System memory ● → Backup memory Override Manual or programmable control feature which enables the user to override programmed feedrates or speeds in order to adapt them to a specific workpiece or material.
  • Page 848 Programmable Logic Controller: → Programmable logic controller. Component of → NC: Programmable control for processing the control logic of the machine tool. PLC program memory SINUMERIK 840D sl: The PLC user program, the user data and the basic PLC program are stored together in the PLC user memory. PLC programming The PLC is programmed using the STEP 7 software.
  • Page 849 Glossary Pre-coincidence Block change occurs already when the path distance approaches an amount equal to a specifiable delta of the end position. Program block Program blocks contain the main program and subprograms of → part programs. Program level A part program started in the channel runs as a → main program on program level 0 (main program level).
  • Page 850 Glossary R parameters Arithmetic parameter that can be set or queried by the programmer of the → part program for any purpose in the program. Rapid traverse The highest traverse rate of an axis. For example, rapid traverse is used when the tool approaches the →...
  • Page 851 Glossary Setting data Data which communicates the properties of the machine tool to the NC as defined by the system software. Softkey A key, whose name appears on an area of the screen. The choice of softkeys displayed is dynamically adapted to the operating situation. The freely assignable function keys (softkeys) are assigned defined functions in the software.
  • Page 852 Glossary Synchronized actions 1. Auxiliary function output During workpiece machining, technological functions (→ auxiliary functions) can be output from the CNC program to the PLC. For example, these auxiliary functions are used to control additional equipment for the machine tool, such as quills, grabbers, clamping chucks, etc.
  • Page 853 Glossary specify that multiple channels share one → TOA unit so that common tool management data is then available to these channels. TOA unit Each → TOA area can have more than one TOA unit. The number of possible TOA units is limited by the maximum number of active →...
  • Page 854 Glossary User memory All programs and data, such as part programs, subprograms, comments, tool offsets, and zero offsets/frames, as well as channel and program user data, can be stored in the shared CNC user memory. User program User programs for the S7-300 automation systems are created using the programming language STEP 7.
  • Page 855 Glossary Working memory The working memory is a RAM in the → CPU that the processor accesses when processing the application program. Workpiece Part to be made/machined by the machine tool. Workpiece contour Set contour of the → workpiece to be created or machined. Workpiece coordinate system The workpiece coordinate system has its starting point in the →...
  • Page 856 Glossary Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 857: Index

    Index $AC_TOOLR_END, 101 $AN_ACTIVATE_COLL_CHECK, 273 $AN_COLL_CHECK_OFF, 273 $AN_COLL_IPO_ACTIVE, 273 $AA_BRAKE_CONDB, 430 $AN_COLL_IPO_LIMIT, 273 $AA_BRAKE_CONDM, 430 $AN_COLL_LOAD, 273 $AA_BRAKE_STATE, 430 $AN_COLL_MEM_AVAILABLE, 274 $AA_COLLPOS, 273 $AN_COLL_MEM_USE_ACT, 274 $AA_COUP_ACT, 302 $AN_COLL_MEM_USE_MAX, 274 $AA_COUP_CORR, 438 $AN_COLL_MEM_USE_MIN, 274 $AA_COUP_CORR_DIST, 439 $AN_COLL_STATE, 273 $AA_DTBREB, 274 $AN_COLL_STATE_COND, 273 $AA_DTBREB_CMD, 274 $AN_ESR_TRIGGER, 481 $AA_DTBREB_CORR, 274...
  • Page 858 Index $VA_SYNCDIFF, 403 $VC_TOOL_O, 102 $VC_TOOL_O_DIFF, 102 $VC_TOOL_R, 103 aaJerkCount, 800 $VC_TOOL_R_DIFF, 103 aaJerkTime, 800 $VC_TOOLO, 101 aaJerkTotal, 800 $VC_TOOLO_DIFF, 101 aaTravelCount, 800 $VC_TOOLO_STAT, 101 aaTravelCountHS, 800 $VC_TOOLR, 102 aaTravelDist, 800 $VC_TOOLR_DIFF, 102 aaTravelDistHS, 800 $VC_TOOLR_STAT, 102 aaTravelTime, 800 aaTravelTimeHS, 800 Acceleration Channel-specific, 752 2-axis swivel head, 84...
  • Page 859 Index Block change criterion, 388 Software version, 554 Block search SW version, 554 Master-slave, speed coupling, 628 Compile cycles, 550 Boundary conditions, 766 Compiling, 762 Compression mode, 90 Compressor, 88 Continue machining ASUB, 716 Contour accuracy Call condition, 765 Programmable, 180 Cardan milling head, 60 Contour tunnel Applications, 60...
  • Page 860 Index CPFRS, 388 CUT3DF, 794 CPLDEF, 378 CUT3DFF, 794 CPLDEL, 379 CUT3DFS, 794 CPLDEN, 385 Cutter shapes, 785 CPLINSC, 401 CYCLE751 CPLINTR, 401 External programming, 518 CPLNUM, 385 CYCLE752 CPLOF, 383 External programming, 519 CPLON, 382 CYCLE753 CPLOUTSC, 402 External programming, 521 CPLOUTTR, 401 CYCLE754 CPLPOS, 391...
  • Page 861 Index DB21, ... DBX0.0, 301 DBB232, 803, 820 DBX0.1, 301 DBX0.1, 822 DBX0.2, 301 DBX0.2, 823 DBX0.3, 301 DBX0.3, 301 DBX0.4, 301 DBX1.4, 815 DBX0.5, 301 DBX1.5, 816 DBX0.6, 301 DBX29.4, 803, 819 DBX0.7, 301 DBX317.6, 820 DBX1.3, 301 DBX318.2, 124, 125 DBX1.4, 166, 302 DBX318.3, 124 DBX1.5, 157, 166, 302, 323, 339...
  • Page 862 Index DBX98.0, 346, 347, 404, 434 DBX98.1, 346, 347, 404, 434 DBX98.4, 348 DBX98.5, 348 Effects on orientations, 68 DBX98.6, 348 EG axis group DBX98.7, 475 - activating, 350 DBX99.0, 302 - Deactivating, 355 DBX99.1, 302 - defining, 349 DBX99.2, 439 - delete, 355 DBX99.3, 348 EG axis groupings, 343...
  • Page 863 Index Gear stage change with activated master-slave coupling, 626 General functionality, 759 General information, 759 Language scope, 766 Generator operation, 458, 471 Large circle interpolation, 79 Generic 5-axis transformation and variants, 65 Laser cutting Generic orientation transformation variants, 67 Clearance control, 564 GUD variables, 722 Leading axes Defining, 378...
  • Page 864 Index MD10622, 267 MD20470, 181 MD10640, 47, 676 MD20482, 90 MD10674, 111 MD20610, 124 MD10700, 759, 761, 763 MD20900, 310, 316 MD10712, 577 MD20905, 310, 313, 315, 324 MD11410, 298, 416 MD21050, 179 MD11410 $MN_SUPPRESS_ALARM_MASK MD21060, 179 Bit31, 346 MD21070, 180 MD11450, 629 MD21084, 787 MD11602, 491, 629, 710...
  • Page 865 Index MD24567, 75 MD32900, 162 MD2457, 68 MD32910, 162 MD24570, 66, 70 MD33000, 161 MD24572, 67, 70 MD33060, 801 MD24573, 74 MD33100, 89 MD24574, 68, 70, 74 MD34040, 152 MD24576, 74 MD34070, 152 MD24580, 104 MD34080, 156, 162, 172 MD24582, 69 MD34090, 156, 162, 172 MD24585, 104, 105 MD34100, 150, 152, 154, 156, 158, 173...
  • Page 866 Index MD37402, 510 MD62611, 644, 654, 661, 666, 668 MD37500, 491 MD62612, 642, 654, 662, 667 MD37500 $MA_ESR_REACTION, 461 MD62613, 642, 654, 662, 667 MD37550 $MA_EG_VEL_WARNING, 348 MD62614, 668, 684 MD37560 $MA_EG_ACC_TOL, 348 MD62615, 684 MD43108, 364 MD62616, 645, 668, 684 MD51160, 269 MD62617, 651, 654, 661, 667, 685 MD51161, 269...
  • Page 867 Index Online tool length offset, 122 Parallel subchains, 187 Opening angle, 120 PI controller, 447 Opening angle of the cone, 119 Plane, 706 Operating modes Pole, 51 JOG, 80 Polynomial interpolation, 112 Option, 746 Orientation vector, 110 ORIC, 777, 791 Polynomial of the 5th degree, 116 ORICONCCW, 119, 121 Polynomials for 2 angles, 112...
  • Page 868 Series machine startup, 709 Setup Collision avoidance, 255 Shutting down, 457 Tangential control, 505 Extended, 458 Applications, 506 Siemens compile cycles, 550 Examples, 630 Simulated master value, 330 TANGON, 510 Singular positions, 50 Technology functions, 549 Singularities, 83 Activate, 555...
  • Page 869 Index Two-axis swivel head, 40 Universal milling head, 36 V2 preprocessing Brief description, 759 Velocity threshold value, 731 Velocity warning threshold, 812 VELOLIMA[FA], 446 Versions, 368 Virtual leading axis, 330 WAITC, 389 World coordinate system, 187 Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...
  • Page 870 Index Special functions Function Manual, 01/2015, 6FC5397-2BP40-5BA2...

This manual is also suitable for:

Sinumerik 828dSinumerik 840de sl

Table of Contents