Posts by DannyDJ

Quote from MFab

When I turn the key switch, the system does shut down — no issues there. I had just assumed that doing it that way (without using the KCP shutdown menu) might be harmful to the mechanical DOMs or IDE HDDs, since it felt like cutting power without giving Windows a chance to close things properly.

That’s what led me to believe there must be a more “correct” or graceful software-based shutdown method, which is why I came here to ask the great and all-knowing minds of this forum!

Sounds like maybe I was overthinking it — if the key switch is designed to signal a proper shutdown sequence internally, and Windows handles it cleanly, then I guess that really is the intended way.

Thanks for the clarification!

When you turn the key switch OFF at the controller, system should start slowly shutting down,Windows should start closing like on ordinary Windows Xp desktop PC when you pressed ON/OFF button(logg of the current user, etc,...), if the screen on the kuka pendant goes instantly off, usually then this is sign that your backup batteries need replacement, like the Fubini wrote.

Quote from MFab

Thanks for the reply!

Yes, to clarify — right now our shutdown process is:

1. Press the KCP “Shutdown” softkey (not the Windows Start Menu).
2. Wait for Windows XP to fully shut down (black screen, no drive activity).
3. Then we turn the cabinet key switch to OFF (0) to fully cut power.

If we don’t throw the switch quickly enough after shutdown, the KRC2 logic reboots the PC again, which is what we're trying to avoid — mostly to protect the old mechanical HDDs from power cycling stress.

We were hoping to configure the system to stay off after shutdown, so we wouldn't need to rely on someone manually throwing the switch every time. But it sounds like, based on what you said, that’s just how KRC2 systems work until at least KSS V7 — the PLC sees the shutdown and re-applies power unless the key is off.

What happens if you just turn the shutdown switch on the controller without selecting shutdown option from the KSS menu? Usually that is the rightway to shutdown KRC2 controller. There is no need to first choosing the shutdown menu option. With shutdown option choosing from the KSS menu, the controller just reboots. If I can remember correctly, it has been always like that.

Also old mechanical DOM's can be replaced by flash DOM(Disk On Module) or even with SATA by using IDE-SATA adapters. I would recommend to make full HDD clones while system is still in good working condition for later easy restoring on newer disks when old IDE HDD disks will start to fail. It will save you a lot of trouble, time and money.

Quote from socal

The error is resetting itself directly. That's why I can't see the error codes. But I caught one code no: 216 A2 motor cable.

There is oxidation at the connector cable entry in the picture. Could this be why?

Which one here is A2 Motor cable input?

Beside the main motor cable from the robot base to the controller, there is also each motor cable part which goes directly from the motor connector of each axis throught the robot(inside robot cable harness) to the robot base and also inside of the controller from the motor cable connector directly to the KSD.

Quote from RobotRoman

Hello, I am learning KUKA.Sim and i came across this issue that id like to step single movements as in fanuc/abb simulation, not to play whole algorithm.

On this img i think i found where to enable step mode, when i start simulation with play, it stops on first line, but i cant find button for NEXT STEP.
image.png

appreciate any help, thank you

Hello, I think this is for executing the simulation with Officelight or you can try and press play again. You do the same on real robot, you press play, robot stops exactly at point, then you press again play, robot moves to the next point in the program.

Here is the new iiQKA.OS2(based on linux), with iiQWorks.Sim Basic free of charge (one programming environment for KUKA robots Wov, KUKA.Sim and OfficeLight). Only my.KUKA account is needed.

I recommend to try it. It looks like something like ABB's robot studio, everything in one programming environment.

iiQKA.OS2: flexibly customizable operating system for robots | KUKA AG

Scalable OS for all KUKA robotics: excellent user experience, efficient start-up, optimized processes, standard-compliant and cyber-resistant.

www.kuka.com

Quote from Wall-E

Danny, your post just saved my ass.

I had been crippled by endless, random servo and brake errors.

I checked and cleaned most of the big connectors, restored a backup of my project, and nothing worked... until I found your post and saw the exact same burnt A1-X3 connector on my CCU.

Robot back to work.

THANK YOU SO MUCH !!!!

This happens because over that wires there can be currents up to 15A flowing and due to poor connection to the CCU X3 pins (connectors are a bit wider) they get hotter and hotter and burning the brake cable connector pins even more.

Quote from Wall-E

Hi !

I need some help with an error stemming from the encoder of my external axis E1 which is a turntable.

This started happening after I fixed another issue with servo and brake errors that was related to a A1-X3 cable terminal on the CCU that had burnt leads.

Once the leads cleaned up, the robot started working fine again except when it abruptly stopped during the execution of a program involving the rotation of the table ; the errors are attached.

Every time, I need to re-master the table to be able to move the robot, then I un-master and re-master the table in my "origin" position.

Of course, I unplugged, checked and re-plugged the two cables leading to the rotary table, but everything looks normal there.

Any ideas ?

--

Robot : KR 210 R3100

Controller KRC4

Software 8.3.37

Display More

Hello, looking to the generated error messages you have a problem with encoder connection to the E1 motor. This can be either from broken, pinched (sometimes you can't even see it from observing the outside condition of the encoder cable,because the wires inside insulations are broken) wires inside the encoder cable or also it can be like panic mode suggested the encoder cable is not shielded to the RDC box.

In your case this I guess happens just for a brief moment, that is why you are experiencing abruptly stopped E1 because it happens during the movement of the turn table and effect of that is also the E1 losses it's mastering.

This would also happen for example if you would disconnect one of the encoder cable from the robot axis during movement along side other errors.

Hi, in your generated program in RoboDK, you are using $BASE which is not attached with DKP400.

With command $BASE={FRAME: } you are just setting ordinary BASE which is offset from WORLD of the robot.

With $BASE=EK(.... command you should be working with defined external kinematic in KUKA.Sim or on real robot.

Usually on real robot you calibrate the DKP400 by moving the tcp of the robot to 4 different positions of the DKP400 axes(cross mark) so that robot knows where DKP400 is located according to the robot.During calibration procces you pick one base number(reference coordinate system for DKP), and this BASE is no longer "attached" to robot, but to External Kinematic and when you move DKP also this BASE moves along with DKP and if you choose attached BASE also the robot will follow the DKP.

The same behaviour by running the generated program you would get on real robot. In your roboDK generated program you should call a base coordinated system which is attached to external kinematics(in your case DKP400) on a real robot or in KUKA.Sim.

Also I noticed that you have commented lines in roboDK generated program

;using external axes ----

$BASE=EK(....

Did you tried to uncomment those lines and run the code in KUKA.Sim and comment $BASE={FRAME: ....} line?

Also on real robot when you have done EK calibration and choose the base number for the attached coordinate system in your program, it is then also enough to call just for example BAS(#BASE,17) if BASE_DATA[17] is coordinate system attached to EK.

You can see which BASE is attached in config.dat file.

Also it is better to call BAS(#TOOL, number of your calibrated tool) instead of directly assigning the system variable $TOOL because you are also taking into consideration the LOAD_DATA of the tool, which can have big impact on the accuracy of the robot.

In that way your roboDK post proccesor should be also generating the $LOAD command.

You are calling exact position in the FOR LOOP

Did you tried to run with C_DIS approximation parameter?

With $APO.CDIS before motion you set approximation distance in mm.

$VEL.CP = 0.01

FOR INDEX = 1 TO 114   

SLIN P1_POS[INDEX] C_DIS

 TRIGGER WHEN PATH = 2 ONSTART DELAY=0 DO SIGNAL_LOG (PASS, INDEX) PRIO=1

ENDFOR

Quote from Fummelknabe

Thank you for the responses.

Indeed, after i updated my WorkVisual Version to 6.0.31_Build3247, the EL6xxx are being recognised and i can configure them without issues!

The trouble is, that I want to control the I/Os using a background task. For this I need to export the I/O configuration to the sunrise workbench. I am getting an error when trying to export my I/O configuration to the SunriseWorkbench (V.2.6.5). The Sunrise Optionpackage (V. 1.6.0.0, included in robot delivery) is installed for WoV.

It is the same result for different I/O configurations. I suspect that this WoV version might not be compatible with my Sunrise version, as the KUKA ecosystem seems to be very picky when it comes to different software versions.

Do you have any suggestions, what I could try to resolve this issue?

But you should be able also to import the EL6xx in previous versions of WorkVisual. I had the same Device Description files imported also in previous versions of WorkVisual. When I make an upgrade of WoV, the catalog of Device Description files stays the same.

Maybe you could try again with correct version of WoV for your Sunrise Workbench.

Quote from panic mode

have not tried it recently but as i recall, using array points in a loop is ok with legacy motions but.. you get mixed results with new motions

so using LIN will approximate and using SLIN ... well ...sometimes....

In my case with new SCIRC array commands running them through FOR LOOP like in the video to generate spirals for the miling didn't have any issues with approximation nor with advance run pointer stopping but pre-calculated points on the circles are evenly distributed.

In fact I needed to use SCIRC because of the circular path accuracy which with old CIRC I couldn't achieve. Circles in the video are only 18mm in diameter and motions are smooth.

Quote from Fummelknabe

Thanks for the replies. It is good to hear that it is possible to use EL6xxx. This will make things much easier for me.

Though, I am wondering how you managed to get it to work. Work Visual (6.0) does not seem to recognise the Device Description files for the Beckhoff adapter. Atleast, they are not visible in the DMT catalog after importing.

As this was also the topic of another discussion on this forum, I concluded that these adapter are not supported.

In my case in WorkVisual 6.0.31 the Device Description files for the EL6xx Beckhoff adapters are recognized without issues.

Quote from panic mode

years ago i borrowed RS232 and RS485 modules to see if i could convince them to work. it did not work. later on KUKA did add support on newer KSS but i have not tried it.

https://xpert.kuka.com/ID/PB13730

Hi, I can confirm that Beckhoff EL6021 works ok on KRC4 KSS 8.6. I had a project in past where I needed to implement communication directly with OnRobot gripper and I've used EL6021 for that.

Quote from Jones1403

Thank you all for the valuable contributions.

Indeed the CMOS battery on the KPC motherboard died. The KPC was not booting at all. No Fan, no power to the CDROM,...

I replaced the CMOS battery and jump started the ATX PSU as per directions from Hermann.

The KPC booted right up after pressing F1 to re-initialise the BIOS settings.

Thanks again, I'm glad this forum exists!

If you needed to replace CMOS battery, the BIOS setting also got reset and maybe now everytime you shutdown the controller you will need to jumper the PWR_ON pins on the motherboard until you restore KUKA bios setting in BIOS.

Quote from Jones1403

Hello,

At the university I'm responsible for a KR15 robot with a KRC2 controller. Since last friday, the controller won't boot. The cabinet fan turns on, the 24V rail is present, also at the input of the power supply of the PC, yet the PC refuses to start.

Where and how to start debugging? In the documentation of the KRC2 I found something about some status LED's on the upper right unit in the controller cabinet. yet it's unclear what the current status is trying to tell. LED 1 (Red) is off, LED2 (green) is blinking at 1Hz.

Also the status leds on the motor drives are blinking (red LED burns continuously, green LED linking at 0.5Hz approx.)

Thanks

Hello, does the PC fan turns on also? Is there any picture on the KUKA pendant(KCP)?

Quote from Fafqulec

Hello,

Thx for reply.

Our PLC use the same version, but little newer...This is critical?

Yes of course. You must use profinet gsdml files according to the installed KSS version on the robots and profinet drivers on the robots. It is all written in the KUKA profinet manuals.

You can't use profinet 5.0 for the KSS 8.3 robot for example.

Look at the tables which herman posted and use correct gsdml files for correct KSS versions.

Different versions of profinet gsdml files are available in WorkVisual install directory.

Quote from Matus

Yes, I can ping server from my virtual machine and I also can ping virtual machine from server. I use Hyper-V because this license was delivered for Hyper-V. But I use my PC as a server and the virtual machine run on the same computer. I hope this is not a problem. I got a task to run it at first on my computer. Later when there will be a server available I will spread it for that.

Usually if you can ping the server, it should work. Maybe you can try also to disable the windows firewall just for testing purposes or add program firewall exception for lmgrd.exe and KUKAROB.exe

I also use my PC as server and runing OfficeLight on the same PC and it works.

In the past I had similar error on KRC4 Compact. At the end I need to exchange the motherboard, didn't help to reinstall fresh image.

In the picture KUKA answer regarding the same BSOD screen.

Quote from panic mode

in my case it is a node locked license so LIC file is all i need. and it cannot connect to any of servers.

sigh... and of course it is expired.. again...!

in fact it was running on an expired license for a month...

have to wait for email from KUKA. thus is just ridiculous... same crap every single year

Yes i can feel you, they could just sent permanent .lic file

Quote from Matus

When I click Perform Status Enquiry I get:
--------
Status
--------
Flexible License Manager status on Mon 2/17/2025 22:12

[Detecting lmgrd processes...]
License server status: 27000@192.168.1.16
License file(s) on 192.168.1.16: C:\KUKA\FLEXLM\xxx.lic:

192.168.1.16: license server UP (MASTER) v11.9

Vendor daemon status (on 192.168.1.16):

KUKAROB: UP v11.9
Feature usage info:

Users of KUKAROB_VRC_2: (Total of 1 license issued; Total of 0 licenses in use)

Users of KUKAROB_HMI_8: (Total of 1 license issued; Total of 0 licenses in use)

I didn't install KRCVrcManager in the office lite because I cannot access there. When I start Virtual Machine, nothing happens. I need to launch SmartHMI.exe and StartKrc.exe. It is launching for some time and then appears the screen:
" Unable to establish connection to the KRC.
Robot control not possible"

Now I don't have option to choose Specify the Licence Server System, but yesterday I had that option and when I entered the IP there I obtained: "License for KrcVrc not found or invalid", as I mentioned in the first message. I don't understand what changed.

Display More

This Status Enquiry looks good, the same as mine, then I guess officelight can't see the FlexLicence server from the virtual machine.

Can you ping the FlexLicence server from virtual host or any other IP address? In my VMWare player settings I have NAT option enabled for the network adapter.

Sometimes I also got this message Unable to establish connection to the KRC.
Robot control not possible" I think it was most of the time if I used the wrong version of VMWare player as I can remember.

Quote from panic mode

License is either file or server based. Not both. So if you have license file, why follow instructions for license server ..?

Is it floating license delivered as file and you are setting up local server?

I also needed to obtain .lic from KUKA with HOST ID data sent to KUKA for the floating licence, which I put on the server where lmtools.exe is installed.

What does status window show when you perfom Status Enquiry in lmtools.exe?

And did you also installed KRCVrcManager in the office lite?

What happens when you enter HOST IP in the office light when you choose specify the licence Server system option?