OK, so sorry for the drawn out explaination, but I’ve worked my machine over quite a bit. I have a German Reprap Neo, that I fit with an E3d Lite6 hot end and an aftermarket thermistor. After Doing my last PID autotune, my hot end runs about 3-5 C below setpoint, and I can never get above 235 now. I can reset the Values to the old ones (P=200 I=0.0 D=200) but that will cause the heat to fluctuate highly. I don’t think German Reprap supports this machine anymore, and really, what I did might be considered total warranty void anyway.
It runs a tweaked version of Marlin on a Printrboard Rev D.
Does this sound like I need to update firmware? I’m a total rookie at that.
Thanks for any help in advance!
I would suggest the thermistor lookup table is not entirely accurate.
This could be because the table is not for the replacement thermistor and your auto-tune results support this conclusion.
I would check that your thermistor is supported by the firmware and, if so, ensure the correct thermistor table is selected.
OK, that makes sense. I’ll see what I can do. Like I said though, very much a newbie with Firmware. The version of Repertier Host I use actually has a feature to adjust firmware settings in it called IEEE config. I’ll take some screenshots when I get home. I’m also semi-aware that if I look it up on my laptop, it should show up as “H” drive, am I right?
You may be able to do some manual tuning to fix the autotune results. The autotune in Marlin isn’t very good, it’s not the correct method to use for this kind of process (fast active heating and slow passive cooling). After auto tuning, if you get oscillation, drop P by maybe 10-20% increments until the oscillation stops, and then go a little farther to be safe. (Raising D can also reduce oscillation but that’s very dependent on your hardware being well-coupled and low sensing noise.)
Once the oscillation is gone, raise I in small increments until the static offset goes away. If it’s too high you’ll get a lot of preheating overshoot and eventually more oscillation.
Manual tuning is more an art than a science… There are more rigorous ways like Ziegler-Nichols if you’re brave.
How small of an increment should I adjust I for? are we talking like .01 at a time? Just trying to get a ballpark. Makes sense, really. No oscillation happening, but the offset is probably where my beef is.
In principle we could dig through the Marlin code to calculate the exact amount you need to raise the I value to eliminate the undershoot, but that sounds like a lot of trouble. I’d probably start with .05 increments or something like that.
Oh, another thing that might work better, I think Marlin’s I_max value is editable in config.h? Raising that by 20% should fix this too. Don’t have time right now to dig through sources to confirm though, sorry.
Sounds great. I’ll check it out when I get home tonight. Thanks!
OK, updating this a bit. My firmware is actually a specific brewed flavor of Repetier, issued by the MFR. I was warned by the OEM that messing with any firmware settings would void by warranty (although I may have already waved that bye bye anyway). I was, however able to settle things in by using @Ryan_Carlyle 's method of playing with the value.
Thanks guys, much appreciated.