E3D V6 Temperature drop mid print

Recently I have been trying to print with petg instead of the usual pla. I changed out my thermistor to a screw in version and updated marlin accordingly. I ran several PID auto tunes with the part fan running full speed and got about Kp:11 Ki:.4 Kd:66. I ran the new settings as a test and doubled my Ki after the hot end would sit just below the target. Finally I began a print only to be met with the horrendous clicking from the extruder. After twenty seconds into the print the temp was down to 210 from 235 and slowly dropping. The power supply is a 430 atx supply from an old computer that I am sure can put out enough power yet the hot end can't seem to maintain temperature. I'm sure the supply has enough juice as the hot end could idle at 250 easily. I am not sure what settings I need to change in order to fix this. Even when I slow down my printer to below 50% (originally 30mm/s) the temperature slowly drops. If I wait long enough the hot end will come all the way back to room temperature. Jiggling the little yellow squares on the ramps board has no effect and at this point I am out of ideas.

The biggest problem here is the Kd, it's way too high. Drop your Kd to 20, double your Kp, and see what happens. I've found autotune of hotends to be pretty useless without manual tuning afterwards.

I've printed PETG with the regular thermistor.
Is the new probe far from the original probe or does it get fresh air?
Can you print something with PLA and see if that works?

Quote
nebbian
The biggest problem here is the Kd, it's way too high. Drop your Kd to 20, double your Kp, and see what happens. I've found autotune of hotends to be pretty useless without manual tuning afterwards.

I changed the Kp to 20 and the Kd to 20 yet the problem still persists. How did you know which values to change and how much?

Quote
o_lampe
I've printed PETG with the regular thermistor.
Is the new probe far from the original probe or does it get fresh air?
Can you print something with PLA and see if that works?

The new probe is glued into a M3 nut which is then screwed into the heater block. I am sure that the thermistor is reading properly in it's current position. I also tried printing PLA and got the same results. The heater is able to reach the right temperature yet it cannot keep it there once the print starts. If I stop the print though, the temperature will begin to rise back to the set value. Could this be an issue with the firmware?

For any extruder temperature oddity, it's worth checking the crimp connections between the solid core wires from the cartridge and the multi-strand used for the bulk of the connection. I've had multiple heaters with intermittent connections that were easily fixed by re-crimping.

Hmm that's odd. Is the max PWM value in firmware set to something other than full power?

Edit: EXT0_PID_MAX in repetier, should be 255.
PID_MAX (or BANG_MAX, one is defined by the other) in Marlin, should be 255.

Edited 1 time(s). Last edit at 02/07/2017 10:21PM by nebbian.

Quote
nebbian
Hmm that's odd. Is the max PWM value in firmware set to something other than full power?

Edit: EXT0_PID_MAX in repetier, should be 255.
PID_MAX (or BANG_MAX, one is defined by the other) in Marlin, should be 255.

Both of the PID and bang bang values are sitting at 255. If it was a typo like 25 the heater would never reach full temp.

Very odd. You haven't changed the heater cartridge have you?

Quote
JamesK
For any extruder temperature oddity, it's worth checking the crimp connections between the solid core wires from the cartridge and the multi-strand used for the bulk of the connection. I've had multiple heaters with intermittent connections that were easily fixed by re-crimping.

Same issue can happen with the thermistor wires. The solder can get soft at those temps and that looks like dropped temperature for the controller.

OP:- If I wait long enough the hot end will come all the way back to room temperature
Does the hot end actually drop to room temp or just the reported temp? Assumed answer is that the temperature drops or you wouldn't have extruder issues.
Seems very much like something is loose and the movement of the carriage is making it worse.
Since its the temperature that is dropping I would suspect its the heater cartridge that is the issue.
There is also a max temp limit set in firmware. I am not experienced in its operation and we may be assuming you have not had an error message but if the temperature is a bit unstable you may have hit this and turned the heater off.
Hope you solve it.

Quote
mrmuffin
The new probe is glued into a M3 nut which is then screwed into the heater block. I am sure that the thermistor is reading properly in it's current position. I also tried printing PLA and got the same results. The heater is able to reach the right temperature yet it cannot keep it there once the print starts. If I stop the print though, the temperature will begin to rise back to the set value. Could this be an issue with the firmware?

Based on that, your not using a genuine E3D V6 hotend. The genuine ones are either held in place by a washer and a small screw or the new style has a cartridge thermistor. They never did a thermistor that was screwed into the side the block.

Anyhow, that point aside, If you heat the hotend, and then wiggle your thermistor wires around to simulate your head moving do you still get the temperature drop? Does your temperature drop happen in the same place on every print? Could it be related to powersupply brown out issues over time? Are cooling fans for the print coming on when your temperature drops?

There is an aftermarket thermistor which screws into the original e3d block where the screw for the thermistor usually goes. The maker of it and e3d had a public row about e3d paying a third party to issue take downs of cloned products passing off as genuine and agreed they were okay with this as it's an add on not a clone. Doesn't work with the new cartridge thermistors/pt100's. It's not a bad idea if it's decently made but I haven't used one to know.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

I'm sure it's not a thermistor problem. I have had thermistor problems previously and I know the symptoms. I know it's not simply a temperature read error because as the temperature drops the filament begins to jam. I did some testing and found that the maximum my hot end can handle is 30mm/s pla at around 220C. Even when I increased the temperature mid print, it could not push above 220. I think I might need to replace the heater cartridge, this one is a couple of years old and might not be able to keep up anymore.

Are you sure the problem is in the heater? Maybe the MOSFET controlling the heater is overheating?

TL;DR...
I don't know if this is correct, or related to your issue, but in case.... I had several prints fail because the firmware (Repetier) decided, close to an hour after the print started, that the heater & thermistor were decoupled (I guess that meant that the heater was on, but the thermistor was showing the temperature decreasing) and stopped the print. The graphs confirmed the temperature drop. I believe (without knowing anything about MOSFETs or what happens if they overheat) that the problem was that the MOSFET controlling the heater was overheating and therefore stopped conducting. I also had problems with stepper drivers overheating, and, since I solved that, the decoupling hasn't recurred.