PID Tuning Edge temperature AC Bed Heater

fcwilt

Hi,

I have a similar setup and it works well. Have you tried reducing the duty cycle?

Frederick

dc42

Some overshoot is normal when you run tuning. In what way does tuning fail?

Killernoy

@dc42 it over shoots in the sense that when the edge reaches 60 I could boil an eeg in the middel and there for the temperature does not fall in phase 2.
10:40:41Auto tune cancelled because temperature is not falling
10:39:42Auto tune phase 2, heater off
10:35:05Auto tune phase 1, heater on

half a minute later it starts to cooling down

Killernoy

@fcwilt what would you recommend to start testing with? p0.5 gives me this failure:
Auto tune cancelled because temperature is not increasing

Killernoy

@Killernoy said in PID Tuning Edge temperature AC Bed Heater:

@fcwilt what would you recommend to start testing with? p0.5 gives me this failure:
Auto tune cancelled because temperature is not increasing

P0.75 same Problem

dc42

Try P0.25

Killernoy

@dc42 same with p0.25 and p0.1

Killernoy

I will work my way up from p0.75 till it starts working
If it doesn't I'll post again

Killernoy

p0.9 did the Job
16:04:42M307 H0
Heater 0 model: gain 159.6, time constant 1084.6, dead time 50.7, max PWM 0.90, calibration voltage 24.1, mode PID, inverted no, frequency default
Computed PID parameters for setpoint change: P23.9, I0.102, D849.4
Computed PID parameters for load change: P

16:03:01Auto tune heater 0 completed in 1388 sec
Use M307 H0 to see the result, or M500 to save the result in config-override.g

15:46:21Auto tune phase 3, peak temperature was 64.5

15:45:22Auto tune phase 2, heater off

15:39:58Auto tune phase 1, heater on

15:39:52M303 H0 P0.9 S60
Auto tuning heater 0 using target temperature 60.0°C and PWM 0.90 - do not leave printer unattended

Phaedrux

I'm not a fan of the idea of having such a high wattage heater being so far from it's control sensor. I would much rather have the sensor directly attached to the heater so that it's able to be accurately modeled and tuned, and then adjust my warm up procedure to allow a high enough set point and warm up time for the bed surface to reach the desired temperature.

Having the control sensor so far removed from the heating element just seems reckless?