Hey folks, Once again I'm overhauling a J-Head but this time I'm doing something

3D Printing
, , ,
1

Hey folks,

Once again I’m overhauling a J-Head but this time I’m doing something a little bit different. I didn’t have any of the standard j-head electronics handy (thermistor, heater) so I scavenged them from a QU-BD hot-end I had in the parts bin.

Overall this has gone OK, but I’m having trouble getting it to hold temp. It seems to ocillate about +/-5 degrees around the target temp. I’ve tried doing some PID autotune, and it’s gotten better, but +/-5c is about the best I can do.

I’m wondering if anyone has any advice on using these ceramic heaters or troubleshooting this sort of problem? It seems to get hot super fast, outrun the electronics which then over-compensate, etc. I almost wonder if more thermal mass or something would help?

Anyway, any advice is appreciated, thanks!

2

The PID autotune (M303) should get you pretty close to the settings you need. But you need to let it run through the autotune loop like 5-10x. Are you sure you have PID turned on in the firmware? You could also look up what each value does (P, I, and D) and hand-tune them.

3

Why would you need to run the autotune loop for so many iterations to get good values? Is the data stored cumulatively?

4

@Anthony_White it’s simply what Marlin suggests doing. I haven’t looked into why.

5

Hey @ThantiK . I did run the 303 a few times (two or three) but I didn’t understand why running it more than once mattered. That said, I’ll give it a try and see what happens :slight_smile:

6

It’s not running the command over and over, just to be clear. It’s letting M303 continue to measure for a while.

7

M303 runs, returning values every so often, until it declares itself complete and gives you the “best” parameters

8

Might be worth checking that both the heater and the thermistor are making excellent thermal contact with the heater block. I remember a time when I just couldn’t get PID to stabilise in my Huxley and it turned out the thermistor was loose.

9

I’ve been doing what @Anthony_White describes, letting M303 run until it says it’s done :slight_smile:

10

That may be the issue @Richard_Gain , the thermistor from the QU-BD parts is a bit “fatter” than the one that came with the J-Head and doesn’t quite fit in the hole; I thought I could get away with taping it tightly to the surface (since that’s how it was arranged on the QU-BD) but now I might drill-out the original hole in the J-Head to accommodate the new thermistor (I was avoiding making irreversible changes, but now that I know it at least kinda works, I’m feeling more brave :wink: