Posts by dj_adamix


    Okay that makes more sense. Thank you for your help.

    There is no dedicated input to enable/disable RPS function.

    I think you may need to read up on RPS in the external IO manual and also operations manual.

    In what context are you trying to use RPS?

    Oh I have been reading manuals, posting on a forum is my last resort.

    In the manual it states that it can be enabled via external input. Look at the attachment.


    All my other dedicated signals work fine (Motor On, Cycle start etc)


    All I'm trying to do is to reset pg0 to step 0 when plc cycle starts. I've accomplished this by creating a PC program as stated earlier, but I was under the impression that I can directly control RPS from PLC once I set it to a dedicated input.

    That doesn't seem to make much sense.

    Your title says 'RPS doesn't switch on', but your post says:

    You also need to turn on the system switch RPS in order for external RPS requests to be executed.

    This system switch is usually turned off as default.

    When using this function, it is common to control the system switch inside a background to task to ensure it never gets turned off.


    What I meant was that my EIP communication works and I'm able to monitor input being turned on by the PLC on the robot pendant. This same input is set up as a dedicated input for the RPS.


    So on my input monitor the input turns on when I trigger it from the PLC but the RPS does not. Its almost like they are not mapped internally in the robot, but I do have them set in the aux 601 to be dedicated (I believe its 601, don't have the robot in front of me now).


    I have created a PC program which turns on RPS. I set a dedicated input for "PC program 1 start" and tied it into another input coming from EIP and I'm able to run my PC1 program when I trigger it from the PLC, which turns on the RPS. It works, but it feels more of a workaround.


    If I set RPS to a dedicated input, shouldn't it turn on when I trigger that input from the PLC? It does not, I feel like I'm missing something.

    Hello,


    I have set the RPS to dedicated signal 1039. This signal is routed over EIP to my PLC.

    All my other dedicated signals work fine (motor on, cycle start etc), but RPS wont turn on.

    EIP communication works fine and i can see the input turn on.


    What am I missing?


    Thank you

    Hello, i think that you can do a very simple with only BREAKs, this stops the program pointer, and continue when the robots ends the movement.


    I'm looking for a way to do it, without breaking my CP movement, so robot does not stop or reduce velocity between points.

    You're 100% right. I always try to simplify the system so operators can use it with no issues.

    It doesn't matter how hard i'd try, there is no amount of training that would be enough for some on the maintenance guys to quickly pick up manual robot control. I'd get phone calls on daily basis asking how to home that damn thing. Would get really annoying, so i ended up writing a template for auto homing.


    Yes, if robot is stopped before reaching the position, marker never sets, But that's okay, when auto-homing is executed, it will move to the previous position first. (one it just moved from). You should never set your markers so they set at the beginning of your robot movement.

    You just have to keep in mind to use same motion for your homing (linear or ptp) as the one used in your main program.

    I also write up a detailed homing path layout on paper, then test every highest marker in the sequence and make sure it homes properly from every possible location.



    yes it adds more but i like it. Like you said , every one does it differently.

    I've tweaked my template enough times to be 99% sure it will not cause crashes. Key is to use some fail-safes to prevent people from editing the program etc.

    but... you cant fix stupid.


    anyways, thank you for your answer.

    You have a lot of good resources here for Kawasaki. Thank you , i was super lost few weeks ago when i first started with this project. Your other posts helped me understand it a little bit more.


    Kawasaki tried charging me another $3000 for KTools.... a 19.7MB program, which should be free...., i ended up using KIDE from one of your posts. It's still inferior to other software i've used before (kuka or denso). But it gets the job done.


    That makes sense thank you


    Well, this is just the way i always write my homing sequence on other robots. I used to do it with work zone monitoring signals back in the day, but i had a couple of customers who managed to move the robot into a spot, somewhere in a zone that would cause a collision with an object when they tried to home it in auto. Perhaps i should've had more fail-safes in place, that was years back when i first started programming robots.


    Some robots only offered 8 zone monitors max that i could use, some projects required to have over 10 zones set up in order to properly home it from every position. In that case i would have to start writing my own monitoring code which would monitor x,y,z coordinates in the background.. it was just getting more complicated and unreliable.


    I'm working in tight spaces with robots reaching inside molding machines, and part assembly stations. Some robots would have up to 10 different programs one for different part/ mold/ assembly station. I just feel with my method, there is less error as every position gets tracked by the controller. Every program can have its own homing sequence and its own set of homing points, which are essentially same points used in the main program.


    I found that recording each position step into a variable is a bit more complicated to program at first, but then, it gives me assurance that robot will always home in a specific path, and in a way it was intended.


    I also pair it with a fail-safe preventing users from using auto-homing feature if they ever flicked the switch into teach mode on the robots pendant.

    So if someone in the plant decides to be a robot programmer that day, they will need to manually jog the robot to home and manually avoid any obstacles on it's way. ... I usually get a phone call when that happens... saying... -'robot doesn't work' ... -'well did you home it ?' ...- 'noooo'... lol


    ... but at least it prevents them from moving robot into an unknown position, then running the auto home program, when robot still "thinks" its somewhere else.


    What's your go-to homing sequence?

    Wouldnt that break my CP motion, making robot halt between movements ?

    If yes, any other way I can accomplish that?

    Hi all


    new to Kawasaki, usinr e01 controller


    I have created a real variable named current_position


    every movement robot makes i want to write point number into the variable, which can later be used for homing purposes.

    This is how i do my homing on other robots.


    I'm having problems synchronizing the value written to the robots actual position.


    From manual :




    i'm trying to use PREFETCH.SIGINS to synchronize my code but i think this only works with signal outputs and not variables.


    Basically, for example i want to write a value of 4 to current_pos when robot finishes moving to p4, not when it starts moving to p4.

    Please see code below.


    right now robot will write 4 to current_pos soonest it starts moving to p4 and not at the end of movement.


    Not sure how many lines ahead does the controller read. If its only 1 line, i could play with it and just make it work, but i don't want to risk it as i know other robots go up to 3 and up (kuka), which can be defined.


    Anyone know of a trick i could use to get that done ?






    Seems like you're more concerned about me not providing useless information rather than trying to pinpoint the issue or give an advice.

    I said ive set up profinet many times before, you dont believe me. I checked the PLC error buffer, no error = comms work ok. even without mapping the IO on KRC4.

    On older profinet drivers it was required to map something in order to get the driver to start, on newer stuff (3.3 i believe) you dont have to.

    I said profinet was working just fine without profisafe so why go through all this headache and give you step by step instructions of what i did.


    I provided you with enough information related to profisafe issues i was having. No need to be an a$$ about it.


    Anyways, i came here for help and all i got was insults.... keep running your forum like you do mr moderator.


    Anyways, like i said before, if anyone has the same issue, look into your failsafe Watchdog timer on your PLC. For Siemens it's F_WD_Time parameter. All i had to do was to increase the timer from 150 (GSDML default) to 250ms and everything started working, even without any io mapped.

    Afterwards i successfully mapped IO and got the robot going.


    Thanks

    To Everyone out there having same issues.

    Check your f_WD_Time on your PLC under 64 safe digital in and outputs (KRC4 GSDML section)


    i had it set to 150ms (GSDML default) , once i increased it to 250, problem went away and KRC4 successfully connects to PLC via Profinet.

    Also, i made sure update time under KRC4 Profinet interface in siemens TIA matches update time specified in WoV profinet settings (8ms).


    Those two options fixed my Kss15002 Issue.

    i apologize, here is more info


    KRC v8.3.37

    Options installed : Profinet KRC-Nexxt v3.3.1

    Diagnose Safeyt v2.1.0

    PLC Siemens CPU 1214FC Dc/Dc/Rly

    GSDML used on plc : KRC4-Profinet_3.3

    WoV V 5.0.6


    KRC4 came configured with SIB and x11,

    Please note that ive done profinet communication many times on this type of robot, never used profisafe though.

    I followed my usual set up procedure for Profinet (non profisafe) but this time i selected the number of profisafe outputs to 64, Set the Profinet ID to 7 ( matched f_Dest_address on PLC side). Removed SIB from WoV project, then deployed the project.


    Kuka boots into Local E-stop only , with KSS15002 - Safe device communication error Profinet Device


    I've tried uninstalling Profinet Nexxt, then reinstalling it again as this fixed some issues with profinet in the past.


    i Will include my WoV project file, am i missing anything ?


    Do i need Profinet ProfiSafe Device option installed on KRC4?

    I'm having the same problem


    Ks150002 safe device communication error profinet device

    And safety stop.


    I followed I instructions provided by kuka .

    Set profisafe ID to my plc f destination address.


    Tried all that and nothing works.

    I set up all reserve bits to be true .

    This just gives me plc memory wrote error meaning comms arent working .


    Non profisafe comms are working..


    Profisafe wont work :frowning_face:


    X11 is disabled with a jumper on x311 and x48 unplugged to the SiB


    Any suggestions ?

    And yes, last thing we were thinking of doing is to change the valve on the nozzle to a variable electric actuator-valve. That way we could control the flow rate.
    That's an absolute works case scenario as I dont want to spend the time wiring and reprogramming . Also this machine runs 25 different parts so it would require a lot of recipies, paramater changes for each corner to optimize the flow.


    That's the case if they ever want to upgrade to run faster. For now I think this will do.
    Thank you so much!

    I think rebuilding the end of arm tool will also help as the nozzle is only held with 1 bolt . Whole thing wiggles too much.


    Does changing the load data help a lot? I've entered the center of mass and mass of the end of arm
    Problem is there is a thick glue hose attached to the arm so those values wont be true while robot moves.
    Also my supplementary load for the robot is just set to default. Not sure if changing that would make things better

Advertising from our partners