RRF 3.4 beta 7 | Heater and Driver Problem!

CR3D

Yesterday we had a big problem with a system on which RRF3.4 beta 7 is installed.

The error only occurred once for us and then again at the installation at the customer.

A Duet3 6HC and a 3HC expansion board are installed. Heaters are connected on the 6HC. X and Y axis steppers on the expansion board.

The setup is already in use several times and is actually not a problem from the ground up!

For error description:

after the restart, heater 1 (nozzle) could not be heated up. Even the M562 command could not solve this problem.

Curiously, another error occurred at the same time. The X and Y axes could no longer be homed! A look at the console then showed Driver 1.0, 1.1 and 1.2 not found!

Multiple resets from the 6hc and 3hc brought no improvement. also no M999 or other possibilities.

Only after a M303 H1 (PID Tuning Heater 1) then solved the problem with the heater 1 heating error.

After that (perhaps coincidentally) the X and Y axes also went again.

@dc42

Are there any connections here?
is this problem known?
Are there any solutions for this?

The main problem is that the printer is already with the customer and I can no longer test it!
As I said, the error occurred exactly once within 2 weeks of continuous operation in the test run and could not be reproduced again.

I would be very grateful for a solution.

Many greetings
Christian

Phaedrux

@cr3d said in RRF 3.4 beta 7 | Heater and Driver Problem!:

A look at the console then showed Driver 1.0, 1.1 and 1.2 not found!

Is there a G4 delay in your config.g to allow time for expansions to startup?

When the problem occurs are you able to send M122 B# where # is the can address of the expansion board?

What is the state of the status LEDs on the expansion when this occurs?

3.4 RC1 is now available to test with.

https://github.com/Duet3D/RepRapFirmware/releases/tag/3.4.0rc1

CR3D

@phaedrux

as I said, the main problem is that the machine is no longer with me since yesterday! the error occurred to the customer immediately. Unfortunately, I can no longer test and check anything

the config as it is worked so far with older firmware on several machines, which is why I assume that it is not a fundamental problem.

But yes I was able to address the expansion board. The LEDs also works well...

jay_s_uk

@cr3d the heater error was known in beta 7 and was fixed in interim releases supplied by DC42.
Is it really a good idea supplying a machine to a customer with BETA firmware? And then afterwards being surprised that not everything works?

CR3D

@jay_s_uk

I like your humor... I agree with you 100%! but if the last stable also had BUGS that made this version unusable for us, we just had no other choice!

Unfortunately, that's the reason why I've been saying for a long time that there shouldn't be such long intervals between the stable versions. An attempt should be made to shorten the distances.

look here:

https://forum.duet3d.com/topic/27082/when-will-the-next-stable-version-finally-be-released/2?_=1644755309361

we have the same setup for two other printers, also with BETA 7, and so far we haven't had any problems. Even with the system we could not reproduce the error again.

jay_s_uk

@cr3d you need the build from here to fix the heater issue https://forum.duet3d.com/topic/26787/unofficial-rrf-3-4-0beta7-4-files/18

CR3D

@jay_s_uk

Thank you i will take a look...

I hope for a stable version in the next days....

jay_s_uk

@cr3d i doubt stable will come that quickly

dc42

@cr3d is the Duet running in standalone mode, or with attached single board computer?

If it happens again, please ask the customer to run M122 and save the report before powering the machine off or otherwise resetting it.

We have one other report of a similar fault in 3.4beta firmware, and I implemented what I hope would be a fix prior to 3.4RC1 - I couldn't be sure if it would fox it because were were not able to reproduce this issue.

CR3D

@dc42

Standalone Mode

ok i can check M122 then... Powering off did not work for about 5 times, also reseting the mainboard and expansion board had no effect.

Only PID Tuning of the nozzle "activated" the board...

Yes as i said...we had this issue one time two weeks ago at our workshop... we were not able to reproduce it and we said ok... perhabs it made to much resets with firmware changes or something else. Two other machines with the very same setup had no issue like this till now...

I hope for a solution to fix this... Thank you very much!

infiniteloop

@cr3d said in RRF 3.4 beta 7 | Heater and Driver Problem!:

Yes as i said...we had this issue one time two weeks ago at our workshop... we were not able to reproduce it and we said ok... perhabs it made to much resets with firmware changes or something else. Two other machines with the very same setup had no issue like this till now...

Just my two pence: Check the wiring, especially the power lines in vicinity of the boards.

If two identical machines have no problems, but a third one shows spurious fails on multiple actors at once (Heater refuses to work, "Driver 1.0, 1.1 and 1.2 not found"), if, on top, the same problem shows up again after shipping the machine, my first bet is on a less-than-optimal crimp. The "Driver 1.0, 1.1 and 1.2 not found" message indicate a potential loss of power to the 3HC, but these lines should have one or more connectors in common with the heaters, too. That's where I would look first.

But, of course, it's only a speculation. Good luck!0

CR3D

@infiniteloop

Thanks for your answer!
I agree with you that it is very strange.

I was just surprised that after an M303 H1... it worked again...

infiniteloop

@cr3d said in RRF 3.4 beta 7 | Heater and Driver Problem!:

I was just surprised that after an M303 H1... it worked again...

Sure. But from the software side, the motion system (e.g. the stepper drivers) is far away from the heaters, I would have to look for two independent faults happening at once - which, on top, is not reproducible. Using Occam's razor, the search for a single error common to both events is more promising: that's why I would look at the power lines. Call me lazy

oliof

@CR3D did you already try upgrading to 3.4RC1 which was released last week?

dc42

@cr3d it's a known issue in beta7 that sometimes a heater will not heat up. Usually it only occurs after doing a print and then letting the heater cool down to below 25C. Fixed in RC1.