3.2beta3 - 1LC loses connection and network errors



  • Duet 3 6HC 3.2beta3 Rasberry Pi 4
    Tool board 1LC 3.2.beta3 and boot loader updated, ID 121
    Z probe is a BLTouch 3.1.

    Start print, H0 and H1 are showing expected temps.

    Pruas Slicer start gcode:
    T0 ; make sure the hotend is selected
    M98 P"/macros/PETGOffset" ; set z probe offset
    M104 S[first_layer_temperature] ; start heating the hotend
    M190 S[first_layer_bed_temperature] ; heat the bed and wait
    G28 ; home all axes
    G32 ; bed calibration - RRF
    M109 S[first_layer_temperature] ; finish heating the hotend
    G1 E5 ; extrude a bit

    H1 temp connected to temp0. Remove thermister connection from temp0 and the resistance is as expected. Tried another thermister, no change.

    Home all runs fine, no errors. Start the print, after the bed heats up some, H1 temp goes to 2000C.

    Then sometimes these errors pop up, but not in the console:
    Failed to switch off remote heater 1: Board 121 does not have heater 1
    Board 121 does not have input handle 1000
    Failed to enable endstops
    Board 121 does not have input handle 3000
    G0/G1: insufficient axis homed
    Failed to enable Z probe

    The input handle 3000 error doesn't always show up.

    The tool board would also error on all it's connections when starting the print.

    Emgergency stop and the Duet 3 reset switch restores the hotend temperature reading.

    I was able to get the print to start twice. It would start and print fine for a while, then H1 would go to 2000C and the extrusion would stop. Never happened at the same layer.

    I'm also getting network errors when first connecting to DWC, while starting a print (file error), during printing, and while selecting Jobs (doesn't get file list). Running Advanced IP Scanner makes the initial DWC connection work for some reason.

    3.2beta2 would randomly get the insufficient axis homed, after X,Y have homed and the Z home wouldn't complete after going to G1 X190 Y200.



  • Same problem when the Pi 4 is powered by it's wall wart. Start print and H1 goes to 2000C.



  • Did an emergency stop after the fail. Now it decides to print. Now to see if it will complete it.

    (edit)
    Nope. Went to 2000C during the skirt. Failed again.



  • Forgot to mention.

    Installed 3.2bate3 on Sunday. Had to do a reset since the H1 temp show n/a. Worked fine for several prints. Monday, did one print just fine. Today, it doesn't.

    I always use Putty to shut down the Pi4.


  • administrators

    Did you update the tool board firmware after the update to 3.2-b3? If not, try running M997 B121



  • @chrishamm Yes the tool board was updated with 3.2.beta3 and the new boot loader.

    Just tried to set the hotend temp on first boot up, via 7" Paneldue 3.20. Set to 250C for petg. Never increased, but it went green and then showed 2000C.



  • Oops. Forgot to update my sd card for the 1LC setup. H1 now showing temp.

    Started print, bed heats up, then the below errors occur and print canceled. H1 now show 2000C.

    Error now says:
    Board 121 does not have input handle 1000
    Failed to switch off remote heater: Board 121 does not have heater 1
    Cancel print
    Failed to enable endstops



  • Looks similar to the issues I saw.



  • Reverted to: Board TOOL1LC firmware 3.2beta (2020-09-14b1)

    H1 has temp, but doesn't increase. Heater measures 13.1 ohms. Will see if I have a replacement. Found a new one, 19.1ohms, no change.


  • administrators

    @Stephen6309 said in 3.2beta3 - 1LC loses connection and network errors:

    Duet 3 6HC 3.2beta3 Rasberry Pi 4
    Tool board 1LC 3.2.beta3 and boot loader updated, ID 121
    Z probe is a BLTouch 3.1.

    Start print, H0 and H1 are showing expected temps.

    Pruas Slicer start gcode:
    T0 ; make sure the hotend is selected
    M98 P"/macros/PETGOffset" ; set z probe offset
    M104 S[first_layer_temperature] ; start heating the hotend
    M190 S[first_layer_bed_temperature] ; heat the bed and wait
    G28 ; home all axes
    G32 ; bed calibration - RRF
    M109 S[first_layer_temperature] ; finish heating the hotend
    G1 E5 ; extrude a bit

    H1 temp connected to temp0. Remove thermister connection from temp0 and the resistance is as expected. Tried another thermister, no change.

    Home all runs fine, no errors. Start the print, after the bed heats up some, H1 temp goes to 2000C.

    Then sometimes these errors pop up, but not in the console:
    Failed to switch off remote heater 1: Board 121 does not have heater 1
    Board 121 does not have input handle 1000
    Failed to enable endstops
    Board 121 does not have input handle 3000
    G0/G1: insufficient axis homed
    Failed to enable Z probe

    The input handle 3000 error doesn't always show up.

    The tool board would also error on all it's connections when starting the print.

    Emgergency stop and the Duet 3 reset switch restores the hotend temperature reading.

    I was able to get the print to start twice. It would start and print fine for a while, then H1 would go to 2000C and the extrusion would stop. Never happened at the same layer.

    1. If possible, please try running 3.2beta3 in standalone mode, so that we can separate the different issues that you appear to have encountered.

    2. The error messages relating to board 121 and the 2000C H1 temperature could indicate that the tool board has reset. Next time this happens, run M122 and M122 B121 and post the results here.

    We expect to release an updated RRF binary for the 6HC later today, to address issues with communication with the Pi.



  • I'll update when installed. I did the M122 before I got back on here.

    When H1 goes to 2000C, the hotend fan does not turn on.
    I can turn on the parts cooling fan, but can't run it off since there is no fan0. The hotend fan is a Noctua NF-A4x10, it is a very quiet fan.

    M122 and M122 B121, before and after attached.
    M122Before.txt m122After.txt


  • administrators

    Thanks, from your M122 reports I deduce that there was a stack overflow on the tool board caused by an attempt to report a temperature sensor error or a heater fault. This will be fixed in the beta that I hope to release this afternoon.



  • @dc42 From your comment, I checked and it turn out to be bad connection between the the female XT30 connector and out0. Replaced the wires, connector and ferrules and it now heats up. The problem did let me find a bug. 🙂

    It was a bad solder connection to the XT30.


Log in to reply