Heated chamber PID and 3.4 stable

Tinchus

Hello. I have upgraded to 3.4 stable version. One of the new features is the ability to to have 1 heater with 2 or more output ports. Since y heated chambers had 2 heaters (physical heateris), now I can turn them on/off configuring a single heater in the firmware, great!

But I have this problem: the old PID values are not working anymore. SO I ran an autotune routing and I got:

R0.097 K0.077:0.000 D93.61 E1.35 S0.90 B0

I saved these values but still IM getting an error saying the temperature is rising too slowly.
the command:

M307 H0
Heater 0: heating rate 0.097, cooling rate 0.077, dead time 93.61, max PWM 0.90, mode PID
Predicted max temperature rise 118°C
PID parameters: heating P19.7 I0.054 D1288.1, steady P19.7 I0.093 D1288.1

SO I changed to M307 H0 R0.05 K0.077:0.000 D93.61 E1.35 S0.90 B0

And I still get the error saying now:

Error: Heater 0 fault: temperature rising too slowly: expected -0.00°C/sec measured -0.02°C/sec

if I issue a 307 command I get:
Heater 0: heating rate 0.050, cooling rate 0.077, dead time 93.61, max PWM 0.90, mode PID
Predicted max temperature rise 72°C
PID parameters: heating P38.1 I0.104 D2499.0, steady P38.1 I0.181 D2499.0

Questions: why autotune cant give something that works and what can I do to hel autotune?
What is the meaning of the preducted temperature rise?

The board is a duet3 running in SBC mode, 3.4 version
Thanks in advance

martin7404

@tinchus Use bang bang
m307 B1

dc42

@tinchus said in Heated chamber PID and 3.4 stable:

Error: Heater 0 fault: temperature rising too slowly: expected -0.00°C/sec measured -0.02°C/sec

It appears that turning the heater on caused the chamber temperature to drop. Is there much noise in the temperature reading?

Tinchus

@martin7404 It is a heated chamber, not a heated bed. It would be much better to use PID to control its temperature than just bang bang

Tinchus

@dc42 If with "noise" you mean if I can see temperature fluctuations, then the answer is no. I have watched the heating process , temperature is always going up, sometimes there is a drop of 0.5 degrees (it should be, there is heated air moving inside so some fluctioations should be there)

Tinchus

I tried to use bang bang: M307 H0 R0.09 C284.700:284.700 D130 S0.9 V0 B1

This still gives me an error like this: Error: Heater 0 fault: temperature rising too slowly: expected -0.09°C/sec measured -0.21°C/sec

martin7404

@tinchus mine is working great

martin7404

@tinchus here it is mine

M308 S2 A"Chamber" P"temp2" Y"thermistor" T100000 B4725 C7.06e-8
M950 H2 C"out2" T2                                 ;
M307 H2 A11 C99000 D2000 B1   ; 
M141 H2
M143 H2 S70

Tinchus

@martin7404 But you are using bang bang mode, not PID

Tinchus

@dc42 I have noticed something. PID start to work when the temperature is like 15 degrees down of target. That is when I can see the SSR start to be on and off, like trying to regulate temperature. So PID is trying to regulate temp way much before target temperaure is not even close to be reached, and that is where I get the error.

martin7404

@tinchus yes it is a chamber anyway.
Mine holds temp within 0.2 deg C

Tinchus

@martin7404 this topic is not for a debate about this I guess. But you chamber limit is 70, is way low compared to me 180 degrees chamber. At those temperatures, using bang bang is somehow complicated in order to mantein stable temperatures across the chamber for many reasons. Reaching 70 degrees is easy for my chamber with the actual "failed PID", the problem Im having is for reaching above 85 degrees, that is where I start getting errors.

This arise with version 3.4, with previous 3.3, I had the PID working with no problems, and those values also dont work with version 3.4

Phaedrux

Does the PID tune fail at the same point each time?
Does it succeed when tuning to a lower temperature?

Tinchus

@phaedrux The tunning finishes and reports me the value, I saved them using M500 and I checked those values are in the override file. Printer has been rebooted too and those values are being used. But when I try to reach the same temperature the calibration was suposed to tune, I got the errors described above.

Tinchus

More information I forgot: definition of the heater is this:

M308 S0 P"temp0" Y"thermistor" T100000 B4138 A"chamber" ; sensor 0 en pin temp0
M950 H0 C"out7+out8" T0 ; pins out 7 and out8 for chamber heaters
M307 H0 R0.015 K0.083:0.083 D150.25 E1.35 S0.90 B0 ; enable PID mode for the chamber heater and set PWM limit
M141 P0 H0 ; Define H0 as a chamber heater

Phaedrux

What's the reason you're using only a 0.9 PWM value?

Tinchus

@Phaedrux voltage tension at home is over normal values. I should have 220V, but I have 235 V always!. The electric resistor I have is meant to be used on 220V, if I use it at 235 it burns. So by reducing the pwm to 0.9 I can use it without the need to get a new one able to work at 235V.
I already tried both ways: I tried to tune the PID with the PWM value at 0.9 and also at 1.0 without success. Each time tunning finishes, I get what are suposed to be the values for PID. Then those values... just dont work.

I couldnt find an explanation of what te new PID values mean or how to tune them manually, is that information available?

Tinchus

I have ran the tunning again. Tunning ends and I get this (in the case of tunning being made for a temp chamber of 85 degrees):

M307 H0 R0.097 K0.077:0.000 D93.61 E1.35 S0.90 B0

I have just noticed that second pair of values on the KK0.077:0.000

My config includes 2 heaters separated, so just in case 1 burns I can easily detect wich one, replace it, etc.

Version 3.4 allows me to put tgether 2 heater under the same control, so my config uses:

M308 S0 P"temp0" Y"thermistor" T100000 B4138 A"chamber" ; configuro sensor 0 en pin temp0
M950 H0 C"out7+out8" T0 ; create bed heater output on out0 and map it to sensor 0, pins out 7 and out8
M307 H0 R0.097 K0.077:0.000 D93.61 E1.35 S0.90 B0 ; enable PID mode for the chamber heater and set PWM limit
M141 P0 H0 ; Define H0 as a chamber heater

As I said, when I try to heat up the chamber, it goes to something like 60/70 degrees and I get a fault. Can be those cero values on the second pair of the K be a problem?

Thanks in advance

baird1fa

@tinchus I have the same issue changing from 3.3 to 3.4, my chamber heaters M307 values don't seem to work anymore and I too get the heater0 rising too slowly error.

I find that odd that I'm getting a heater0 error since my chamber is heater3. But that aside I've tried to change the vales to get the heater to work and I'm having a tough time. My chamber is much slower heating than yours and it takes a very long time to even reach 50C. I've been running the autotunner now for several hours.

Phaedrux

@baird1fa said in Heated chamber PID and 3.4 stable:

I find that odd that I'm getting a heater0 error since my chamber is heater3.

I think that is a bug in 3.4 that has been fixed in 3.4.1 RC1.

When a heater fault event was generated on the main board, the heater number and CAN address parameters were swapped. This resulted in the heater number always being reported as 0.
When a filament error event was generated on the main board, the extruder number and CAN address parameters were swapped. This resulted in the extruder number always being reported as 0.

Would you be able to test 3.4.1 RC1?
https://github.com/Duet3D/RepRapFirmware/releases/tag/3.4.1rc1