Post by pernockham on Inheritence of struct, ^{CODESYS Forge talk (Post)}
Im looking for a way to define predefined version of the same structures through "extends"/inheritance. What I want to do is best shown with an example: TYPE log_item_val_type : ( BOOL_ := 0, INT_, REAL_, STRING_ ); END_TYPE TYPE LOG_DATA_BASE STRUCT val_type : log_item_val_type; (* value, name etc *) ENDSTRUCT ENDTYPE (* this/below is not possible as I understand from the compiler?? *) TYPE LOG_DATA_BOOL extends LOG_DATA_BASE : STRUCT val_type : log_item_val_type := log_item_val_type.BOOL_; END_STRUCT END_TYPE TYPE LOG_DATA_INT extends LOG_DATA_BASE : STRUCT val_type : log_item_val_type := log_item_val_type.INT_; END_STRUCT END_TYPE etc. for LOG_DATA_REAL and ..STRING. The system will not allow me to "re-define" "val_type" however. Instead I must do the work-around of defining four different types with all fields individually defined. The benefit if I could extend the base item would be that Im (more) sure the structures are identical. In usage I can call in a specific type rather use the base type and assign which type. Usage by pointer/ref to log_data_base would also be easier.. (* Usage: *) log_data : log_data_int := (name, value etc.) (* Instead of: *) log_data : log_data_base := (name, value, val_type).. Is this possible to achieve in some way that I have missed?
Last updated: 2025-03-05

Post by imdatatas on Auto Declare window disappears by itself in the Ladder editor ^{CODESYS Forge talk (Post)}
Hello Everyone, The "Auto Declare" window disappears by itself in the Ladder editor. For example, a new function block is wanted to be added to the Ladder editor. (For example; MC_Jog block and instance name will be fbMcJog). To do this, "Empty box with EN/ENO" is added to the ladder line from the Toolbox on the right and when you type MC_Jog in the empty box and press enter, normally the "Auto Declare" window appears and you are expected to be asked to write the instance name here. However, this "Auto Declare" window closes itself immediately, it does not wait for the user to enter a value or close the window. I often encounter this situation in many IDE versions. Does this seem like a bug? If anyone has encountered a similar problem while creating a new program name, function or function block instance name regarding the "Auto Declare" window and knows a solution, I would be glad if you could share it. Best wishes Imdat
Last updated: 2025-03-06

Post by egau on Enable and Disable Project IO programmatically ^{CODESYS Forge talk (Post)}
Hi @eschwellinger, I tried this with Modbus COM slaves. I was able to disable them (slave becomes greyed out in the device tree), but the DED.Reconfigure "eError" output shows "NOT_SUPPORTED". After this, when I tried to re-enable the slave, it did not work (device stayed greyed out in the device tree), and the DED.Reconfigure "eError" output also showed "NOT_SUPPORTED". Is there something to do about this, or the device just doesn't support reconfiguration? P.S: I tried this running locally on my computer (not in simulation mode). So of course I was not physically connected to the devices. I have two similar projects, but one of them doesn't have the Modbus_COM devices. I know I can "Exclude from build" the devices for that project, but I would really prefer not to do this if possible. I would prefer to dynamically toggle a configuration variable that would enable or disable the slaves.
Last updated: 2025-03-14

Post by pernockham on Persistence Manager: Config File not found? ^{CODESYS Forge talk (Post)}
Back in november I paused this, but wanted to reattempt using the persistent manager, the behavior is still the same. Most "login attempts" s are stopped with the message "One or more errors occurred" but I have no idea what that error might be. The only clue is this message is presented while the lower left status box says "Generating code...: relocation pou". For clarification. Build/Generate code is successful, no errors. Error appears when I try to download/login (physical or simulation) to a device. I did manage to login (in simulation mode) at one point, but then the error message "PersistenceChannel: Config file not found!". And no persistence files were created. I have done device-tree/reset-orgin also clean/clean all etc. Removed/re-inserted persistance manager. Behaviour is just the same. Removing Persistence Manage and everything is working OK. Is there possibly more info somewhere about the "One or more errors occured"-message?
Last updated: 2025-03-20

Post by pistola on C0357: "GetNextClient" is obsolete, use VisuUtils instead ^{CODESYS Forge talk (Post)}
I'm having some troubles with this same issue and I'm wondering if someone can help me out. On a settings visual I allow the operator to enter some values however if they activate the page change button it will change the change with the Numpad dialog open. Since I'm using Visuelems.CURRENTVISU to change the pages I came across this code noted below for determining if the Numpad dialog was active. This code worked great however now in 3.5.19 it's now obsolete. I've tried following the directions in the attached program above however I can't see to get it to work in my program. Can anyone provide some help to determine if a dialog is open? FUNCTION Check_Dialog_Open : Bool VAR_INPUT sDialogName : STRING; // Input variable for the name of the dialog END_VAR VAR pstClientData : POINTER TO VisuElems.VisuStructClientData; // Pointer to the client data structure itfDialogManager : VisuElems.IDialogManager; // Interface for the dialog manager itfMyDialog : VisuElems.IVisualisationDialog; // Interface for the specific visualisation dialog END_VAR // Begin the iteration over the client manager VisuElems.g_ClientManager.BeginIteration(); // Loop through each client until no more clients are found WHILE (pstClientData := VisuElems.VisuElemBase.g_ClientManager.GetNextClient()) <> 0 DO // Get the dialog manager interface itfDialogManager := VisuElems.g_VisuManager.GetDialogManager(); // Get the specific dialog interface using the dialog name itfMyDialog := itfDialogManager.GetDialog(sDialogName); // Check if the dialog is open for the current client Check_Dialog_Open := VisuDialogs.VisuDlgUtil_IsDialogOpen(itfMyDialog, pstClientData, itfDialogManager); // If the dialog is open, exit the loop IF Check_Dialog_Open THEN EXIT; END_IF END_WHILE
Last updated: 2025-03-27

Post by nz-dave on MC_HOMING ^{CODESYS Forge talk (Post)}
Hi Just new to motion and i am building my own 3axis cnc cutter using servo drives. i have every thing working quite well apart from mc_home. when executed (r_trig) it stops immediately and i get fb_wasnt_called_during_motion. if i execute the mc_home fb with a toggle it just runs and runs for a while then errors out. what im unsure of is the way mc_home works. from what i understand it changes the drive mode to 6 (homing) and should then run of the drive homing settings. once origin sensor is made then homing would stop and mc_home.done is true. and the position on the fb is where i what the new home position (zero) to be? or do i feed mc_readposition to that input? so i would think its the settings in the drive that aren't quite right. but i believe they are all correct. has any one got any info to steer me in the right direction? Cheers
Last updated: 2025-04-04

Post by r-c-r on SM3RaspiStepper 0.0.0.6 issue ^{CODESYS Forge talk (Post)}
Ok, so i made some tests in the meantime and i have some conclusions. It apears that the enable bit is activated only when a motion function is executed. I implemented a very simple program with MC_Power and MC_Jog. If i enable everything on MC_Power, the enable GPIO pin stays off. However, as soon as the MC_Jog function is used the enable GPIO pin turns ON and remains ON. In version 0.0.0.4 this is not happening. The GPIO enable pin turns on or off when the MC_Power is enabled or disabled. From this behaviour i understand that in version 0.0.0.6 the GPIO enable bit is activated together with pulse and dir pins when motion starts. When motion stops, enable bit remains with the last state (on in this case) and you cannot turn it off for example. For me this bit is useful because i can turn on and off the stepper drive with MC_Power function. Now i can turn it on, but never off.
Last updated: 2025-04-04

Post by pernockham on Engineering IDE from linux? ^{CODESYS Forge talk (Post)}
I have done the switch (linux mint, virtual windows through QEMU/KVM) now and is generally pleased. Have had problems in two areas initially. To bridge IP to the Virtual unit (solved), I overthought things initially. Solution. Configure manually a bridge in the mint-"network connections". Use that bridge in the Qemu settings as "Bridge device", "Device name": "name of the configured bridge" Figure out where to store the projects. I started out have the data in linux world (sort of) and linking it the windows with "virtio-fs". That had some serious impact though on git-handling. It seems that virtio-fs interpreted any file-path as lowercase only even if there were uppercase, resulting in git not able to sync files between its repository and the source code (it created a new file-path for me with only lowercase). Then I switch to samba-share instead and this works much better, though (git-actions) is a part that still has low performance with lots of waiting for file-actions. What should be the best way of handling the source/git? Edit to answer myself: Save the project "locally" in windows, then a lot of slow performance is avoided. Then a git-remote sync for backup.
Last updated: 2025-05-08

Post by mondinmr on Unexpected behavior in cyclic encoder read method during INT overflow (wrap-around) ^{CODESYS Forge talk (Post)}
Hi everyone, I’m investigating a potential issue with a cyclic method we use for reading incremental encoders in our libraries. I’ve come across two implementations that, at first glance, appear to perform the same operation: motionUnit.vlPositionActualValue is UINT due strange encoder type. Version A METHOD PROTECTED cyclicReadField VAR_INST actPosFieldOld: INT; rI: INT; delta: INT; END_VAR rI := TO_INT(motionUnit.vlPositionActualValue); delta := rI - actPosFieldOld; m_actPosRaw := m_actPosRaw + TO_LREAL(delta); actPosFieldOld := rI; IF settings.velocityFeebackMapped THEN m_actVelRaw := TO_LREAL(motionUnit.velocityActualValue); END_IF Version B METHOD PROTECTED cyclicReadField VAR_INST actPosFieldOld: INT; END_VAR m_actPosRaw := m_actPosRaw + TO_LREAL(TO_INT(motionUnit.vlPositionActualValue) - actPosFieldOld); actPosFieldOld := TO_INT(motionUnit.vlPositionActualValue); IF settings.velocityFeebackMapped THEN m_actVelRaw := TO_LREAL(motionUnit.velocityActualValue); END_IF Both use INT and the same delta logic, so they seem equivalent. However, when an overflow (wrap-around) occurs, version A continues correctly as expected, while version B unexpectedly resets to -32768 without a clear reason. Has anyone experienced similar behavior or can shed light on what might be happening here? It almost seems as if the cast to LREAL is being applied before the calculation — but that would be illogical and potentially very dangerous in many situations! Thanks!
Last updated: 2025-05-13

Post by jschu on Dynamical scaling of bar graph without the bar itself ^{CODESYS Forge talk (Post)}
Hi there, Task: dynamical changing the scaling of an bar graph, linear scaling whitout showing the bar only the scaling values and Lines for Main and partial steps. Define 4 Limits for Hi and LO Values with different colors Option 1 : use the bar graph whitout showing the actual bar whitout showing the bar, only the scaling values and lines for main and partial steps. Then what to do with the 4 Limits ? I could use 4 moveable Rectangles with the start and end value fo the areas but then the whoel bargraph must be transparent. Best way would be if only the scale ist possible to show as it is possible for the Meter 180° but this option does not exist for the bargraph. Option 2 : user defined HMTL5 Widgets wich may be some allreay has created or where is a dome for using this kind of html5 widget. The expanse for use html is in my eyes to big and gives more option for errors. Have some one experience in this topic or a hint how to do it the easiest way? Thanks for your help.
Last updated: 2025-05-15

Post by shaunvdm on Canbus Slave SDO Read from Object ^{CODESYS Forge talk (Post)}
Busy testing a solution to use Canbus Slave to replace a physical device and convert to Ethercat device but keep the Canbus Master in place. Master needs to Read SDO object 5FA2. In the Master PLC if i send the SDO with 8 bytes in message IE: 60B 40 A2 5F 00 00 00 00 00 i get my expected 58B 43 A2 5F 00 00 68 36 02 containing data back. Problem is if we issue a 4 byte command can send 60B 40 A2 5F. Surely the canbus slave should respond with a 58B 43 A2 5F 00 00 68 36 02 but it comes back as 80 A2 5F 00 00 00 00 08 - meaning this OBJECT does not exist. Is there a setting in the canbus slave driver for Canbus to allow the 4 byte command to read the SDO OBJECTS ?
Last updated: 2025-05-27

Post by jy77812 on MC_MoveCircularAbsolute Using Help ^{CODESYS Forge talk (Post)}
Dear People I’m trying to draw a circle using MC_MoveCircularAbsolute, but I keep encountering an error. The error message is SMC_CP_CIRCLE_CENTER_NOT_ON_BISECTOR. What I’m trying to do is simply draw a circle with a diameter of 20, starting from the point X=0, Y=0. I’ve set CircMode to center. It seems like a simple task, but the error keeps occurring, and it’s giving me a headache. I’d appreciate your help. I’ve also attached the program, so please provide any advice you can. I want to use the mode as the center. I've tried changing multiple coordinates and it won't run. Please help. sample stCirculAuxPoint.c.X := 20; stCirculAuxPoint.c.Y := 10; stCirculEndPoint.c.X := 20; stCirculEndPoint.c.Y := 30; fbMoveCircularAbsolute( AxisGroup := AxisGroup, Execute := xGroupCirculEx, CircMode := SMC_CIRC_MODE.CENTER, AuxPoint := stCirculAuxPoint, EndPoint := stCirculEndPoint, PathChoice := 0, Velocity := 50, Acceleration := 50, Deceleration := 50, Jerk := 50, CoordSystem := SMC_COORD_SYSTEM.MCS, BufferMode := 0, TransitionMode := 0, TransitionParameter := , OrientationMode := , VelFactor := , AccFactor := , JerkFactor := , TorqueFactor := , Name := , Done => , Busy => , Active => , CommandAborted => , CommandAccepted => , Error => , ErrorID => , MovementId => );
Last updated: 2025-06-04

Post by christoph--b on Probleme mit Speicheraufteilung nach Wechsel von Wago 750-841 zu 750-891 ^{CODESYS Forge talk (Post)}
Hallo, ich habe in meiner Hausautomation eine Wago 750-841 mit einem doch mittlerweile umfangreichen Programm. Diese wollte ich nun gegen eine 750-891 der 4. Generation austauschen. Aufgrund der Programmgröße hatte ich damals die Speichergröße für Retainvariablen von 16#4000 auf 16#7000 vergrößert und so läuft sie nun seit Jahren fehlerfrei. Allerdings mit der 750-891 bekomme ich offensichtlich eine Überschneidung mit dem Modbusregister, denn die Werte über Modbus spielen verrückt. In der Standardgröße mit 16#4000 tritt das Phänomen nicht auf, reicht dann aber nicht für das vorhandene Programm. Dafür gibt es sicher eine Lösung, denn die 891 hat ja deutlich mehr Speicher. Allerdings kenne ich mich mit der Speicheraufteilung nicht aus, wo man da was optimieren kann, damit ich den Speicherbereich entsprechend erweitern kann, aber es zu keiner Überschneidung kommt. Für Hilfe wäre ich sehr dankbar
Last updated: 2025-07-20

Post by pernockham on AC Persistence - values are not read from file during start? ^{CODESYS Forge talk (Post)}
Im struggling with the AC Persistence. AC persistence seems to work for any variable that is changed online and values are kept during reboot. However, if I try to edit the persistence files offline or replace the persistence files with a previously downloaded older persistence file (with modified values), the online value will not follow, as if the system did not at all read the file at boot? My testing sequence. - Edit a the value of a persistenced variable (with xSaveOnChange = True) - Save/download the associated persistence file, (and confirm the value has changed). - Edit the value online for a second time . - Stop runtime - Replace (upload) the persistence file with my previously downloaded file . - Coldstart/Warmstart/Start runtime - Whatever Im trying the value of my persistence variable will always be equal to that of my second edit as if the file is not read/considered during boot. What am I doing wrong?
Last updated: 2025-09-02

Post by e13740e on Parameterized Property syntax for array element processing ^{CODESYS Forge talk (Post)}
Hello. I'm trying to create a Parameterized Property(CDS V3.5 SP21) to avoid copy-pasting code for processing array elements. I can't seem to get any result. Perhaps I am making a mistake with the syntax. None of the declaration options below are accepted by the compiler. If you know and use them (Parameterized Properties) in your work, please let me know. Thank you in advance. Option 1: PROPERTY <property_name> : <data_type> VAR_INPUT <index_name> : <index_type>; // One or more index parameters are declared here END_VAR</index_type></index_name></data_type></property_name> GET Code that uses <index_name> to return a value END_GET</index_name> SET Code that uses <index_name> and the implicit variable <property_name> to write a value END_SET</property_name></index_name> END_PROPERTY Option 2: PROPERTY <property_name> (<index_name> : <index_type>) : <data_type></data_type></index_type></index_name></property_name> GET Code that uses <index_name> to return a value END_GET</index_name> SET Code that uses <index_name> and the implicit variable <property_name> to write a value END_SET</property_name></index_name> END_PROPERTY
Last updated: 2025-10-09

Post by e13740e on Parameterized Property syntax for array element processing ^{CODESYS Forge talk (Post)}
Hello. I'm trying to create a Parameterized Property(CDS V3.5 SP21) to avoid copy-pasting code for processing array elements. I can't seem to get any result. Perhaps I am making a mistake with the syntax. None of the declaration options below are accepted by the compiler. If you know and use them (Parameterized Properties) in your work, please let me know. Thank you in advance. Option 1: PROPERTY <property_name> : <data_type> VAR_INPUT // One or more index parameters are declared here <index_name> : <index_type>; END_VAR</index_type></index_name></data_type></property_name> // Accessor for reading GET // Code that uses <Index_Name> to return a value END_GET // Accessor for writing SET // Code that uses <Index_Name> and // the implicit variable <Property_Name> to write a value END_SET END_PROPERTY Option 2: PROPERTY <property_name> (<index_name> : <index_type>) : <data_type></data_type></index_type></index_name></property_name> // Accessor for reading GET // Code that uses <Index_Name> to return a value END_GET // Accessor for writing SET // Code that uses <Index_Name> and // the implicit variable <Property_Name> to write a value END_SET END_PROPERTY
Last updated: 2025-10-09

Post by cihangur on C0138: No matching 'FB_Init' method found for instantiation of VisuClientObjectMgr ^{CODESYS Forge talk (Post)}
This "FB_Init" error has become very annoying lately. It's usually related to webvisu and is resolved by deleting all objects and related libraries. Of course, you need to re-add the deleted objects. Lately, we've generally been managing by not upgrading, but this problem persists in every version since SP20 and should be resolved. I updated a project from SP 19 to SP 21 and found 116 similar errors. [ERROR] alarmmanager, 4.7.0.0 (intern): ReplaceAlarmPlaceholdersConverted: C0138: No matching 'FB_Init' method found for instantiation of CharBufferString [ERROR] visuelemsalarm, 4.7.0.0 (system): protDrawRowHeader [VisuFbElemAlarmTable]: C0138: No matching 'FB_Init' method found for instantiation of CharBufferString [ERROR] alarmmanager, 4.7.0.0 (intern): AnalyzePlaceholders2 [Alarm]: C0138: No matching 'FB_Init' method found for instantiation of CharBufferString If you have an ERROR related VisuClientObjectMgr you can try to disable "Visualization Manager - Settings - Support client animations and overlay of native elements " selection "C0138: No matching 'FB_Init' method found for instantiation of VisuClientObjectMgr" etc
Last updated: 2025-10-10

Post by ph0010421 on How to create a stopwatch? ^{CODESYS Forge talk (Post)}
Do you need an 'hours-run' counter? And 1 second resolution is ok? I think you're over-thinking it. (The task time needs to be < 1second) Have a look at the LAD... Then, the time in seconds can be made into hh:mm:ss with this FUNction Declarations: FUNCTION funSecondsToStringTime: string VAR_INPUT InSeconds: UDINT; END_VAR VAR AsString: STRING; Minutes: UDINT; Hours: UDINT; Seconds: UDINT; MinutesAsString: STRING(2); HoursAsString: STRING(2); SecondsAsString: STRING(2); END_VAR and the code: Hours := InSeconds / 60 / 60; //Derive hours Minutes := (InSeconds - (Hours * 60 * 60)) / 60; //Derive minutes Seconds := InSeconds - ((Hours * 60 * 60) + (Minutes * 60));//Derive seconds HoursAsString := UDINT_TO_STRING(Hours); MinutesAsString := UDINT_TO_STRING(Minutes); SecondsAsString := UDINT_TO_STRING(Seconds); IF LEN(HoursAsString) = 1 THEN HoursAsString := CONCAT('0',HoursAsString); END_IF; IF LEN(MinutesAsString) = 1 THEN MinutesAsString := CONCAT('0',MinutesAsString); END_IF; IF LEN(SecondsAsString) = 1 THEN SecondsAsString := CONCAT('0',SecondsAsString); END_IF; AsString := CONCAT(HoursAsString, ':'); //assemble string AsString := CONCAT(AsString,MinutesAsString); AsString := CONCAT(AsString,':'); AsString := CONCAT(AsString, SecondsAsString); funSecondsToStringTime := AsString;
Last updated: 2023-12-08

Post by andrax on Meine Gerätesammlung basierend auf TCA9548 ^{CODESYS Forge talk (Post)}
Hallo zusammen, ich programmiere öfters mal kleinere Messgeräte und Datenlogger. Je nach Einsatzzweck musste ich mir ein paar Gerätetreiber selber schreiben. Da es eine recht große Sammlung ist, stelle ich die hier einfach mal ein. Die Sammlung basiert auf den Gerätetreiber TCA9548 Multiplexer von Steffen Dreyer. Codesysversion: V3.5 SP18 Patch 4 Inhalt: -4channel Dimmer (Stefan Dreyer) -BMA456 (Andre Klien) -BME280 (Stefan Dreyer) -BMP280 (Stefan Dreyer) -EEPROM_24C512 (Stefan Dreyer) -ADS1115 (Andre Klien) -ICP10125 (Andre Klien) -PCA9555 (Stefan Dreyer) -PCA9685 (Stefan Dreyer) -PCF8575 (Stefan Dreyer) -SDP810 (Andre klien) -SSD1306 (Stefan Dreyer) -TCA9548 (Stefan Dreyer) -TSL2561 (Stefan Dreyer) Alle Gerätetreiber laufen über den Multiplexer, können aber auch ohne betrieben werden. Einschränkung: BMA456: Die Konfigurationsdaten werden erfolgreich in den Chip geschrieben, dieser meldet aber config wrong. Warum auch immer, der Sensor funktioniert und bringt hochempfindlich Beschleunigungsdaten. ADS1115: Hie habe ich nur die wichtigsten funktionen implementiert. 1-4 Kanäle, PGA, kontinuierlicher Modus, Datenrate. Alle anderen Funktionen (single-shot, komparator,alm) sind nicht implementiert. Das macht auch keinen Sinn, da diese Funktionen besser über die Steuerung realisiert werden können. Achtung beim ADS1115 ist zwingend ein Levelshifter notwendig, wenn ihr diesen mit 5V betreiben wollt. Ihr grillt sonst euren Raspi (SDA, SCL sind dann 5V) Gruß Andre
Last updated: 2023-12-18

Post by spiessli on Softmotion axis to CANopen Maxon IDX drive ^{CODESYS Forge talk (Post)}
Thanks for the advise, indeed, this was somewhat, what my intention was. I tried around again with EDSes and DCFs, anyway. It appears that the DCFs end up in a subfolder in the attach device dialog. Astonishingly, when attaching the DCF version I get a device in the device tree, where I can add a SoftMotion axis, whereas when I attach the device not in the subfolder, then I cannot. Further astonishingly, when looking in the CANopen parameters in supported profiles, "402" is listed in the case of the DCF variant, whereas in the EDS variant "0" is shown. Comparing EDS and DCF files I cannot guess, where the difference is coming from. I cannot tell either why I have always attached the EDS variant and apparantly never the DCF variant. As I am not in the lab right now, I cannot test if the connection actually is working. Thanks again, spiessli
Last updated: 2024-01-07

Post by edepalos on Text list from sd card ^{CODESYS Forge talk (Post)}
Hi guys, I have a textlist in my project which is working just as expected in a drop down list. I would like to move this text list to a sd card and access it from there, because I want to ease the editing of the contents of this textlist. I'm imagining a solution like the operator shuts down the plant, takes the sd card out, edits the textlist on the card with a windows machines notepad, then reinserts the card in the PLC and restarts the plant, having the new content of the textlist populating the dropdown list. I'm sorry that I'm asking you to spoonfeed me, but I did not find anything straightforward to do this... I saw in the Project>Project settings>Visualisation a relative path thingy, but I have the plant in operation and I would not like to make 100 trial and errors right now... I'm also not interested in SysLib file read commands and stuff, I would just like to access the file from the cards part, I don't want to replace its contents dynamically or such... nor I have language files or recipes... Thank You in advance! Br, Ede
Last updated: 2024-01-16

Post by axilleas on SMC_Interpolator + SMC_controAxislbyPos ^{CODESYS Forge talk (Post)}
dear all I am working on a very generic gantry system (virtual drives + win v3 control) reading gcode from external file. In order to stop the interpolator function from running when gaps are detected, SMC_controlAxisByPos FB has the bStopIpo output which according to the manual is connected to the Interpolator's bEmergencyStop. Lets assume that the axis are parked in position X1000 Y1000 and the initial G code movement is G00 X500 Y500. The normal behavior is to travel from the actual x1000 y1000 to G00 X500 Y500. The SMC_Interpolator always gets the X0 Y0 as the initial position, meaning that the bStopIpo is active until the axis are in X0 Y0 and the release the interpolator to set the new positions to X500 Y500. Is there a suggested method to solve this behavior? bonus question: I get a lot of bStopIpo during standard interpolation movements (not gaps). While working with virtual axis this is not a problem but when real axis are installed this will be an issue. What exactly triggers the bStopIpo output? Slow axis accelerations compare to Gcode defaults? Kind Regards
Last updated: 2024-02-05

Post by mondinmr on Direct Pointers in IOMapping for EtherCAT with IoDrvEthercatLib.ETCSlave_Dia ^{CODESYS Forge talk (Post)}
Using SDO, I can read the EtherCAT mappings and offsets from various registers like 0x1c12, 0x1c13, etc. When I obtain registers mapped in the PDOs and various offsets, I could technically access directly to the statusword, controlword, etc., if they are mapped. I have noticed that on IoDrvEthercatLib.ETCSlave_Diag I can find pointers to the input and output buffers. However, although the input buffer can be easily read by referring to what has been obtained from the SDOs, it is not possible to write to the output buffer, as it is overwritten in each cycle by the data from the IOMapping task. Is there a way, knowing an instance of IoDrvEthercatLib.ETCSlave_Diag, to obtain the pointer to the first data in the IOMapping? The offsets are identical to those of the PDOs, but obviously the data is a copy. For me, the cleanest thing would be to access the pointers on IOMapping, for DS402a devices to retrieve the key data and point them to internal references. I need to know this as, having developed our own motion library, I would like to simplify the initial setup by eliminating unnecessary and nonsensical mappings that can lead users to unnecessary errors. In the current version that we have been using for years, we have to map everything manually. I would like to pass only the slave reference to the FB_init constructors and eliminate the mapping.
Last updated: 2024-02-12

Post by dtamm on How to read string from CharBufferPtr? ^{CODESYS Forge talk (Post)}
I am trying to implement a mechanism to send alarm SMS. For this, I have created an Alarm Class with a Notification Action which calls a POU. This POU implements IAlarmNotifiable making it selectable in the Alarm Class configuration. IAlarmNotifiable has an Execute method which is called by the alarm class. Now, I want to extract the alarm message of the triggered alarm. The Execute method comes with a VAR_INPUT itfAlarm: AlarmManager.IAlarm which supplies a method called GetMessage. The help text says: Returns the message text, that has been configured for the alarm. The result is a pointer to the buffer, where the message text is stored The type of the result is CharBufferPtr, and at some place I read that it can be interpreted as a pointer to string. But I cannot manage to extract the alarm message, all I get are numbers. I have tried the following: psAlarmText := itfAlarm.GetMessage(__SYSTEM.TYPE_CLASS.TYPE_STRING)^; sAlarmText := TO_STRING( itfAlarm.GetMessage(__SYSTEM.TYPE_CLASS.TYPE_STRING)^ ); So how do I get access to the alarm message as clear text? (In this case, "Larm 1")
Last updated: 2024-02-21

Post by mtho on JSON Utilities - JSON file written with invalid structure ^{CODESYS Forge talk (Post)}
I'm using the JSON Utilities SL library to write configurations in my program to a file. I'm generating the JSON data using the builder function block and all appears to work correctly. When I write the jsondata to file, I'm finding that some of the JSON elements are not being written in the right location in the file. Rather than writing they keys under the parent object, they get written outside the root JSON object at the end of the file. If I then try to read the file back into the program, I get an invalid_structure error. I don't get any errors when populating the JSON data or when writing the file. I attached an example of the JSON writer output. The problems begin with key P200. Each P200 should go under MBR1 through MBR3. Among the rest of the keys, they are supposed to be under different parent objects. I am using default values for wsLineBreak and encoding.
Last updated: 2024-03-06

Sort by relevance or date