Heater Confusion in 1.19

dc42

There is one other report of something similar, so I suspect a firmware bug. The temperature monitoring code was refactored extensively in 1.19 so it's by no means impossible.

If you run M305 P1 and then M305 P2 with no other parameters, does it print the correct temperature sensing channels?

MoS-tekknix

Hi dc42,

yep. The Channels are correct in these Outputs:

[[language]]
12:37:41M305 P2
Heater 2 uses Thermistor sensor channel 2, T:100000.0 B:4267.0 C:0.00e+00 R:4700.0 L:0 H:0
12:37:19M305 P1
Heater 1 uses Thermistor sensor channel 1, T:100000.0 B:4267.0 C:0.00e+00 R:4700.0 L:0 H:0

But i cannot understand why it is working in your Setup (and i assume at other Printers too) and not in mine?

dc42

Firmware bugs can be like that, changes in the config.g file cause the memory layout to change, which causes the bug to appear or disappear or its effects to change.

Please try auto tuning heater 2 e.g. M303 H2 S200 and let me know which heater number it says it is auto tuning.

MoS-tekknix

Hi,

i know have dealed a long time with Bugs in Marlin …

12:54:14M303 H2 P0.8 S240
Auto tuning heater 2 using target temperature 240.0C and PWM 0.80 - do not leave printer unattended
12:54:20Auto tune phase 1, heater on
12:56:34Auto tune phase 2, heater off
12:56:41Auto tune phase 3, peak temperature was 243.5
Heater 2 switched off
12:59:35Auto tune heater 2 completed in 320 sec
Use M307 H2 to see the result, or M500 to save the result in config-override.g

13:10:27M307 H2
Heater 2 model: gain 433.2, time constant 189.9, dead time 3.4, max PWM 0.80, mode: PID
Computed PID parameters for setpoint change: P23.3, I0.123, D54.8
Computed PID parameters for load change: P23.3, I2.216, D54.8

Can you do me a favoour and drop me your Dual-Setup' config.g File so that i can try to compare where maybe differences are.

Thanks so far
MoS-tekknix

dc42

Sure, it's at https://www.dropbox.com/s/uak4t4iqdxc4bnv/config.g?dl=0. Please share your config.g file too and I'll see if I can reproduce the problem.

MoS-tekknix

Thanks for that.

Here is mine:https://drive.google.com/open?id=0B39F2gM1VUvOaTc5MlF5emhNWUE

dc42

I've just tried your config.g on a Duet WiFi and the temperature readings on H1 and H2 are different for me. I had to disable debugging because you had it enabled (M111 S1 in config.g).

Is there any chance that your E0 and E1 thermistor inputs are shorted together? You could check with a multimeter.

PS - if you are using a common ground wire for the 2 thermistors to save on wiring, perhaps you have commoned the 'live' side of the thermistors instead of the VSSA side.

MoS-tekknix

Hi,

Fault located. One of the Thermistor-Wires touch the Heater-Block and make perfectly grounding everything!

Seems to be a Layer8-Error after sleeving the cables after successfull dry-run.

Will solder some new wires and test if afterwards. But i'm pretty sure that this was the issue and therefore you are'nt able to reproduce it in a working environment.

Sorry for that Dumb Question.

Cheers
MoS-tekknix

dc42

One thermistor wire touching the heater block wouldn't cause that problem, but one wire from each thermistor touching the heater block could. So check the other thermistor too.

Doesn't E3D supply cartridge heaters with the Chimera yet?

MoS-tekknix

Only to clarify things.

Indeed there where one wire of each thermistor touching the respective blocks. That causes the Problem.

After close inspection i've seen it.

The Heaters of Chimera are simple Heaterblocks and you need to insert Thermistor and Heater yourself.

Anyway i've swapped out the delivered 12v Heaters against 24V and new Semitic-Thermistors. It has worked flawlessly but i assume after sleeving all the things together that i've slide out the inbuild sleves of the Themistors and that the contact was there.

Sorry again for this trouble which was my own fault. Too many wires ….

But again a big thanks for this nice build Hardware and your Fork of the original RepRap-FW!

I'm back in business again.

Cheers
MoS-tekknix

dc42

Thanks, I'm glad the mystery was solved! Your report prompted me to review the heater setup code, whereupon I discovered bug involving named heaters. So it wasn't all wasted time!