Does the Extruder PID need to be re-tuned for higher temps?

[dpmacri]

I followed the extruder thermistor PID-tuning instructions in the R-MAX assembly guide and that worked really well for the PLA I was using. The guide has you tune to 200C and the PLA worked well around 185C to 190C. I recently started trying some ABS which needs somewhere in the 235C-240C for the particular ABS I'm using. It seems like the hot-end has a harder time keeping the temperature stable (it fluctuates mostly downwards to around 236C before going back up to 238C when I'm trying to maintain 238C).

My question is whether I need to re-run the PID tuning with the higher temperatures. Anyone have any suggestions?

Thanks,
Dean

[lordbinky]

In the firmware settings try lowering the minimum extruder power value..(I forgot the actual setting name) it is something like ext1_minPID. I think I set mine to like 20 and the power no longer oscillates around the temp I want.

[dpmacri]

In the firmware settings try lowering the minimum extruder power value..(I forgot the actual setting name) it is something like ext1_minPID. I think I set mine to like 20 and the power no longer oscillates around the temp I want.

Thanks! I'll give that a try.

[edward]

Maybe. I wouldn't be surprised if you likely have to re-tune the extruder PID parameters when your temperatures vary that much. PID is a great feedback control method, but it usually isn't that robust unless derived from a detailed system model, which the auto-tuning doesn't have/do. That is, it works great within a narrow operating range. If you look through the Configuration.h file you'll notice parameter called PID_CONTROL_RANGE, and right above it is the a description that says the PID only works within TARGET+/-PID_CONTROL_RANGE, otherwise the heater is full-on or full-off. This even more limits the robustness of the PID method, but it also keeps some other bad things from happening.

It doesn't take much to re-tune, so write down the coefficients for your current settings and run the M303 again. If it works well, write those down also and you can just change them in the EEPROM quickly enough when changing materials.

Lordblinky: that minPID paramter is not a minimum power value. I'm pretty sure it's a lower bound on the I-term in the feedback loop, part of the integral-windup prevention. If it helps to decrease it, great! I just wanted to point out the parameter that directly limits the power out (we're talking Repetier, here) is the EXT#_PID_DRIVE_MAX, which ranges from 0-255 for the 8-bit PWM driver.

[lordbinky]

That's what is was. Ext1_PID_DRIVE_MIN . I was going completely off memory from work where things aren't allowed to be sensible, so my brain doesn't instantly transition well.