Universal Robots Forum

C4A1 error when jogging manually on UR10 CB3.0 above 3.11

Summary

Since upgrading both our UR10 CB3.0 to 3.12 and up till 3.14 we have had constant C4A1 errors while jogging the bots manually. They run flawlessly for days when running programs, but when stopped and moving manually or moving to a waypoint with the automove tab, the bots will consistently throw a C4A1 error after some minutes. Same issue on 2 different robots.

This makes it very impractical or impossible to use with sensitive payloads as we risk crashing expensive tooling while calibrating or programming moves while close to static objects and surfaces.

The only workaround has been to downgrade to 3.11 where the issue has disappeared completely. We obviously would prefer to be on 3.12 and above due to the safety alert fixes.

Please contact your distributor/integrator for a better investigation of this issue.

Hi, I am already in touch with our local distributor. I thought it would be useful to post here in case others are seeing the same behavior and might have found solutions. We initially thought that the common behavior seen on both robots could be an electrical noise issue, but this was also observed on other locations with the same robots, and totally disappeared since going back to 3.11.

I will report back here once we have found a solution.

1 Like

Hi! Did you end up finding a solution? I think I’m having the same issue on 3.12.

Hi,

Unfortunately not yet. Only solution was downgrading to 3.11 which completely eliminated the problem. I really hope we figure this out so we can upgrade again to 3.14.x and above. I will post here if we find any solution, but for now our 2 UR10 are on 3.11 and working fine.

Btw, if you also downgrade to 3.11 and it also eliminate the issue please report here as that would be an independent confirmation that this is indeed an issue not locally isolated to us, thanks.

I just downgraded and it fixed it for me as well. Before this I was getting the C4A1 error every few minutes.

Interesting. Is it also a CB3.0 controller? UR10? Are you in touch with UR directly? might be good to raise the issue with them directly, good to know we are not alone!

Can you run programs ok? we only see the issue when jogging manually, running programs is fine.

Hello,

as I could not find any bug report for that specific problem yet, I am about to make a new one, can you just give me some details:

  • the robot SN
  • confirm it is a CB3.0 (at least one of them)
  • Polyscope versions tested (for example 3.12.1)
  • any URcaps installed (list and versions)
  • steps to reproduce (what I understand, it is just jogging a few minutes the robot, if it stands still it does not come?)
  • a sample program with installation file

Thank you

Universal Robots

Hi,

The issue has been reccuring on both our UR10 CB3.0, 2014300371 and 2016303281.

As mentionned any version above 3.11 will show the error while jogging, 3.12.x, 3.13.x, 3.14.x. As soon as we downgraded to 3.11 it disappeared on both.

We don’t have any URcaps installed, both are basic installations.

The error happen when going in the Move tab and jogging manually on the teach pendant, jogging axis-by-axis or cartesian will show the error after few minutes. Sometimes nothing happens for 20min and then the error happens after 20sec, seems to be completely unpredictable besides the jogging manually. Both can run 24/7 programs without problem in automatic.

Attached a log and a recording (before we downgraded and it disappeared (some errors while in automatic were due to some earlier earthing issue with our electrical installation, since then resolved but the error C4A1 still happened before downgrading)

recording20201022_11_10_14.zip (1.7 MB) ur_2016303281_2020-10-22_11-10-56.zip (3.5 MB)

I will need a bit of time for a sample program and installation but basically without loading a program and in the default installation the error would happen when jogging, with or without a program loaded.

We have discovered that this bug can be triggered when sending e.g. “textmsg(“msg”)” script inside a program without adding any sync() node.I could reproduce this bug on both CB3 and e-Series.
I was able to fix this by adding 0.002 s wait on e-Series and 0.008 s on CB3 between consecutive script executions.
Perhaps there is a place in the customer’s program where some commands are also sent very frequently.

We will be looking into this issue shortly.

Has there been any progress internally on this issue? We are still running 3.11 because of it.