Tuning the heater temperature control bed duet 3 (3.4 beta7)

massimilianoconti

hardware duet 3 v. 1.01

bed 310x310
silicon pad 380W 24volt controlled whit mos module

Thermistor epcos B57560G1104F 100K

temperature test 130°C

firmware 3.4 beta 7+1

I notice this problem: by tuning the heating plate in PID mode, the measurement loop does not end. in the end a result is given with anomalies for too long a time of tuning. I tried in all ways but it does not solve, I think there is an error in the firmware ...

in the previous stable version I never had this problem

Phaedrux

The tuning process can take a long time on large beds that have a very slow cool down. Try letting it finish.

massimilianoconti

@phaedrux I let it go until the end, but from a crypto in yellow with the values obtained. if I perform the tuning of the extruder the result appears ok in blue color. anyway I also put the extra firmware 3.4 beta 7 + 1 but the results do not change so much. I do other tests and paste the values thanks

dc42

@massimilianoconti if the temperature readings are not noisy, you can speed up the tuning process by using a custom Y parameter in the M303 command, for example Y3.

massimilianoconti

@dc42 Hello and good start, I try to add the parameter Y I let you know. Thank you very much. I wanted to suggest to insert in the Duet Web Control a feedback button so as to send to you directly the problems that are encountered to speed up the procedures.

massimilianoconti

in this way putting Y3 in m303 works and quickly. now I see the stability and accuracy of the readings and update the outcome. Thank you

test ok

@dc42 5/1/2022, 18:40:15 Auto tuning heater 0 completed after 3 idle and 5 tuning cycles in 1753 seconds. This heater needs the following M307 command:
M307 H0 R0.456 K0.402:0.000 D8.73 E1.35 S1.00 B0
Send M500 to save this command in config-override.g
5/1/2022, 18:34:16 Auto tune starting phase 3, measuring
5/1/2022, 18:29:03 Auto tune starting phase 2, settling
5/1/2022, 18:11:06 Auto tune starting phase 1, heating up
5/1/2022, 18:11:01 M98 P"0:/macros/Tuning Bed"
Auto tuning heater 0 using target temperature 120.0°C and PWM 1.00 - do not leave printer unattended

massimilianoconti

@dc42 I am using the machine intensively and I see that the temperature of the heating dish is very stable. so playing on the parameter Y tightening the hysteresis you can speed up the tuniong time and solve the problem of too long tuning times. if I notice other anomalies I communicate, for now thank you and ask if you can insert in the duet web control a button to immediately send any problems or requests for improvements ... Thank you

I have tested for a long time it seems to me that there are problems. although I performed the tuning successfully then sometimes when mold stops the bed heater without logic

massimilianoconti

I have tested for a long time it seems to me that there are problems. although I performed the tuning successfully then sometimes when mold stops the bed heater without logic

massimilianoconti

12/1/2022, 00:31:11 Auto tuning heater 0 completed after 3 idle and 5 tuning cycles in 1234 seconds. This heater needs the following M307 command:
M307 H0 R0.513 K0.572:0.000 D7.42 E1.35 S1.00 B0
Send M500 to save this command in config-override.g
12/1/2022, 00:26:54 Auto tune starting phase 3, measuring
12/1/2022, 00:22:51 Auto tune starting phase 2, settling
12/1/2022, 00:10:42 Auto tune starting phase 1, heating up
12/1/2022, 00:10:37 M98 P"0:/macros/Tuning Bed"
Auto tuning heater 0 using target temperature 100.0°C and PWM 1.00 - do not leave printer unattended
12/1/2022, 00:10:28 Upload of Tuning Bed successful after 0s

tuning concluded well but after a while an error occurs see attached photo

Phaedrux

Were the tuning results saved with M500 and M501 added to the end of config.g?

massimilianoconti

Yes. I solved and I put all the indications in the site:

https://duet3d.dozuki.com/Wiki/Tuning_the_heater_temperature_control?revisionid=HEAD

to the voice
Heater failure bed: temperature rising too slowly firmware 3.4 and later

TLAS

@Phaedrux
Sorry to dig up an old post - I'm having similar issues with 3.4.4. I'm tuning a warmer to a target temperature of 40C. The oscillations are quite long to drop 5C - (~15 min cycles at ~20C ambient), and I'm now pushing 3 hours of tuning with no indications of letting up.

(now 13:20)
12/29/2022, 10:55:48 AM Auto tune starting phase 3, measuring
12/29/2022, 10:19:25 AM Auto tune starting phase 2, settling
12/29/2022, 10:17:26 AM Auto tune starting phase 1, heating up
12/29/2022, 10:17:21 AM M303 H3 S40
Auto tuning heater 3 using target temperature 40.0°C and PWM 1.00 - do not leave printer unattended

I'm curious on the noise standpoint - is the noise an absolute value to determine iterations or is it relative (i.e. taking a 5% tolerance of 15 minutes for cooling is very different than something within 5 seconds.

I'm going to be pulling out the limit on the number of iterations - just didn't want to lose the data from all the iterations it's already done if possible. 3 hours is a long time to ask for a relatively slow moving heater.

jay_s_uk

@TLAS the cycles are 5 initial oscillations and then a maximum of 25 recording oscillations. You can limit the amount of temperature drop from the set temperature (default of 5 degrees) using the Y parameter on M303.

TLAS

@jay_s_uk
Any way to force it to stop early and accept the parameters it calculates?

jay_s_uk

@TLAS nope. Wait it out for it to complete for reset and start again.
Longest tuning cycle I heard of was 12 hours and I don't think that completed either