Temperature oscillation which can't be fixed with Autotuning

Hi everyone,

I'm new in the forum and hope I found the right place. I have an Anet A8+ 3d printer running the latest Marlin firmware release.
My issue is with the hotend temperature control. If I tell the printer to heat the hotend, the temperature won't be stabilize, but instead goes into a perfect oscillation.
It looks exactly like bad PID tuning, but PID autotune doesn't seem to fix this issue. What I did:

M106; turn on fan
M303 E-0 S220 C10; PID autotone
M301 P21.0 I1.25 D86.0; set the result of M303 as PID values
M500; save to EEPROM

I have attached a temperature graph. It shows first the heating to 220degrees. then the PID autotune and finally a print that I started which failed as the temperature dropped too far.

The weird thing is, that this behavior started randomly in the middle of a 15h print and caused that print to fail. Before, the temperature was much more stable. I checked the wiring, if anything is burnt, but seems fine. The thermistor and heating element seemed a bit loose, but tightening the screw didn't help to resolve the issue.

Does anyone have any advice on how I can debug this issue further? Or what the issue might be. Broken thermistor, burnt out Mosfet, bad PID autotune?

I have ordered a new hotend and thermistor, as well as a silicone sock. I will test those once they arrive, but it will take 2-4weeks and I would like to resolve the issue and print again before that.

Thanks for any help!!

This is a flaw in the later Marlin releases, and a lot of folks are seeing it. I recently installed a PT1000 platinum RTD on my hotend, and had to install an amplifier and a custom thermistor table to get Marlin to co-operate at all.

As an experiment, you might try changing your P value from 21.0 to 10.0 just to see what happens. It won't hurt anything, but might make the oscillation go away, or at least get better.

Thanks for the quick reply.

Are you saying this is a bug with the Marlin PID autotune?

I've reduced the PID value as you suggested, and it reduced the oscillations, but the temperature control is still terrible. it's not stable at all, and at the start of a print doesn't manage to keep the temperature, so that the printer shut's off with a thermal runway warning.

No, it's not a Marlin PID issue, it's a change in the way Marlin makes the actual temperature measurements with the control board analog to digital converters. Bug reports have been made.

Alright, I restored the original printer firmware and now everything works fine again. Thanks for the help!

Can you by any chance point me to GitHub Issues on the topic? I couldn't find any :/ Then I can track the issue myself and if there's a solution coming.

Quote
corny
Alright, I restored the original printer firmware and now everything works fine again. Thanks for the help!

Can you by any chance point me to GitHub Issues on the topic? I couldn't find any :/ Then I can track the issue myself and if there's a solution coming.

[github.com]

Scroll way back and look for anything to do with PID tuning, temparature instability, etc. I believe this really became a problem after 2.0.7.