Error: Failed to switch off remote heater 0: (while printing)

Phaedrux

Please post your config.g, as well as the results of sending M122 and M122 B# where # is the CAN address for your expansion boards. Check the results to confirm the expansion boards are also running FW 3.3. If not, send M997 B# where # is the CAN address for your expansion boards to update their firmwares.

MathV

@phaedrux , sorry for the very long delay to answer due to my holiday trip.
I updated the firmware to the 3.3 release and checked that the 6HC and both 1LC are running on it (M122, M122 B#).
I also now run on standalone mode to keep it as simple as possible.

Please find below my config and the M122, M122 B20 and M122 B21 responses below.
config.txt
M122.txt
M122 B21.txt
M122 B20.txt

I continue the investigation and keep you updated.
Thanks a lot for your help.

MathV

Observation:
After couple of tries, I noticed a pattern:
If I want to print a gcode with the heating bed ON (let's say 60°C), then as soon as the temperature of the bed is around 55-58°C, the error appears.

Error: Failed to switch off remote heater 0:
Error: Failed to switch off remote heater 1:
Cancelled printing file 0:/gcodes/IDX420_Left_2x_Impeller_4h43min_93g.gcode, print time was 0h 0m
Emergency Stop! Reset the controller to continue.

It looks like the error occurs as soon as the regulation (as soon as the command is not 100% any more) of the bed heat starts.
I do not explain the link between the bed and the heater 0 and 1 (respectively the hotend left and right on my setup).
I am controling the bed in 220V with an SSR zero cross (line frequency 60Hz) with a control frequency of 6Hz (although still 10Hz on the config I sent).
I am controlling the SSR with the 2-wire fan output out7 from the 6HC.

Test:

I tried to print with the bed controled in bangbang, and the error appears more randomly (sometimes at the beginning of the print, sometimes after couple of hours).
I am trying to suppress the bed heating. The print starts properly, but I will let it print over night to confirm that it prints without any issue).

dc42

@mathv it sounds as though turning the bed heater on/off s generating some sort of transient that is causing the board to think it has received an emergency stop command. Do you have a PanelDue connected? If so, is it or the cable connecting it to the Duet close to the bed heater wiring?

MathV

Thanks @dc42.
I do not have a PanelDue, but the 220V bed cable is running close to the CAN bus cables. I will isolate the bed cables from my other cables, test it again and come back with some results.

MathV

@dc42 I tried to connect power the bed with cable running outside the machine, but still the same issue.

I found again more elements this morning:

If I start a print with heating bed ON (with a PID control for the bed temp) when the bed is already at 55°C (for a 60°C target), then the error doesn't occur.
I started the gcode below --> no error.

; no error
T0
M190 S60

I started the gcode below --> no error

; no error
T0
M190 S60
M104 S213

I started the gcode below --> no error

; no error
T0
M190 S60
M109 S213

I started the gcode below --> error

; error
T0
M190 S60
M104 S213
M109 S213

I started the gocde below --> error

; error
T0
M190 S60
G90 G1 X0 Y0 Z200
M400

The error also occur with the last gcode ran as a macro (so without the start.g)

I emptied my tpre0.g and tpost0.g without any impact.

Any idea?

MathV

I double checked my setup, and the SSR wasn't a zero cross as I thought I purchased. Replacing it by a zero cross solved the issue.
Thanks for your support.

dc42

@mathv I am glad you solved it. A non zero crossing SSR can turn on at any point during the mains cycle, so it will generate much more interference than a zero crossing SSR when you try to apply PWM to it.

kb58

@mathv I've used SCRs in the past to control the speed of brushed 115VAC motors. All you need is an AM radio nearby to appreciate the HUGE amount of interference such switching produces. That was a nice catch on your part to double check your SSR; many users have no idea the consequences of using a non-zero-crossing switch can have.

dc42

@kb58 said in Error: Failed to switch off remote heater 0: (while printing):

@mathv I've used SCRs in the past to control the speed of brushed 115VAC motors. All you need is an AM radio nearby to appreciate the HUGE amount of interference such switching produces. That was a nice catch on your part to double check your SSR; many users have no idea the consequences of using a non-zero-crossing switch can have.

Washing machines here in UK used to use phased control of an SCR to control the speed of the drum motor. 10+ years ago they changed to using 3-phase variable speed drives (I found this out because I had to change the VFD board in our washing machine). I was told that this was because SCR-controlled brushed motors can no longer meet EMI regulations, even with suppression components.