ToolBoard: 2 wires fan, heater fault and how to extrude ?

[dc42]

Using RRF 3.2beta4 you can set the H and L parameters automatically even on tool boards. See https://duet3d.dozuki.com/Wiki/Calibrating_thermistor_and_PT1000_readings#Section_Semi_automatic_calibration.

[dc42]

The fault with heater 2 is expected. There is no heater connected, so when the heater output is turned on, the firmware doesn't see a rising temperature, so it flags a heating fault. This is to avoid a dangerous situation, for example if the heater is working but the thermistor has fallen out of the heater block.

The reason that heater 1 is reporting a fault is probably that your M307 parameters are inappropriate for your hot end. In particular, the A597 parameter is probably too high. What type of hot end is it, and what is the power of the heater cartridge?

[dc42]

@Pseud3mys, I have investigated your M122 report and discovered that the heat task stack can overflow when a tool board reports a heater fault. I have fixed this in release 3.2beta4.1 of the tool board firmware, and put an updated Duet3Firmware-TOOL1LC.bin file in the release at https://github.com/Duet3D/RepRapFirmware/releases/tag/3.2beta4.1. Please update your tool boards to this version. M115 B# (where # is the CAN address of the tool board) will report version 3.2beta4 if the old firmware is running, and 3.2beta4.1 if the new firmware is running.

[Pseud3mys]

@dc42 hi, With beta 4.1 boards don't reset, and with the semi auto config for H and L thermistor, 1 of the 2 return the correct temperature, and the heating work with this one. But, the second is still wrong (by 15°C, at ~17°C), what is the problem ? The two have exactly the same config except for H and L setting.

[Pseud3mys]

I check the 2 resistances values, both are at 4.34 Kohm (~11°C). but one thermistor return ~10°C degrees more than the real temperature (~21°C)..
And it is enough to raise a fault with the heater, so it is disturbing.

M308 S1 P"121.temp0" Y"thermistor" T500000 B3600 C1.196220e-7 H-12 L0
M308 S2 P"122.temp0" Y"thermistor" T500000 B3600 C1.196220e-7 H-10 L-12

All the value are equal exept for H and L but they don't change anything significantly (I tried to put the same value).
So why one of the thermistor return a wrong value ?

[Veti]

measure the resistance of the thermistors at room temperature and see of they measure the same value.

[Pseud3mys]

@Pseud3mys said in ToolBoard: 2 wires fan, heater fault and how to extrude ?:

I check the 2 resistances values, both are at 4.34 Kohm (~11°C).

Yes I already did it. It is the same value (4.34 kohm) at 11°C

[Veti]

can you measure the resistance of R21 on the boards

another thing to try is switch to the second thermistor port

121.temp1

[Pseud3mys]

Okay thank you I will try that this evening.

[Pseud3mys]

okay so the R21 resistance is about 2K18 for both two board.
I saw on an other post that is the correct value.

[Veti]

then i dont know where the difference comes from.
you could get some pt1000 ones. maybe they provide better results.

[Pseud3mys]

@Veti said in ToolBoard: 2 wires fan, heater fault and how to extrude ?:

then i dont know where the difference comes from.
you could get some pt1000 ones. maybe they provide better results.

but the thermistor isn't the problem, the two have the same resistance.

[Pseud3mys]

@Veti I there a way to have the value that read the borad directly ?

[Veti]

i dont know where it could come from.

i would play a bit with the H and L values.

what do you get when you set the same on both?

[Pseud3mys]

@Veti I have a difference too, but I played a bit with the H and L value and now it is okay, the hotend doesn't return fault.

[dc42]

@Pseud3mys said in ToolBoard: 2 wires fan, heater fault and how to extrude ?:

I check the 2 resistances values, both are at 4.34 Kohm (~11°C). but one thermistor return ~10°C degrees more than the real temperature (~21°C)..
And it is enough to raise a fault with the heater, so it is disturbing.

M308 S1 P"121.temp0" Y"thermistor" T500000 B3600 C1.196220e-7 H-12 L0
M308 S2 P"122.temp0" Y"thermistor" T500000 B3600 C1.196220e-7 H-10 L-12

All the value are equal exept for H and L but they don't change anything significantly (I tried to put the same value).
So why one of the thermistor return a wrong value ?

You don't need to set the H and L values in config.g because after you run the semi-automatic calibration procedure, the required H and L values are store in non-volatile memory.

You could try removing the H and L values and run the semi-automatic calibration again on the board that is showing the error. But if that doesn't work, use your manually-adjusted H value.

[Pseud3mys]

@dc42 I change the H value manually and now it work. Thank you for your help !

[flajzi]

@dc42 I have same problem. After few hours of printing the extruder heater goes offline, temperature goes to 2000 °C (I'm using PT1000)

Updated to latest 3.2-beta4.1 and problem remains. It keeps printing so I guess it's still keeping 220 °C, but to CAN bus it reports offline status.
M122 on Toolboard says:

Diagnostics for board 1:
Duet TOOL1LC firmware version 3.2-beta4.1 (2020-12-03)
Bootloader ID: not available
Never used RAM 4120, free system stack 36 words
HEAT 46 CanAsync 89 CanRecv 83 TMC 54 MAIN 216 AIN 64
Last reset 11:35:12 ago, cause: software
Last software reset at 2020-11-22 15:02, reason: HardFault, available RAM 4168, slot 0
Software reset code 0x0060 ICSR 0x00000003 SP 0x200011c8 Task 
Stack: 20001018 20001304 0000000f 00000001 20000928 000172cd 00018250 01000000 00077dfa 00077ef4 20001218 200032d0 00000020 00017a61 20004660 00000000 20001300 00008895 00077dfa 20004670 00077ef4 42651000 a5a5a5a5 a5a5a5a5 a5a5a5a5 a5a5a5a5 a5a5a5a5
Driver 0: position 11794334, 409.0 steps/mm, standstill, SG min/max 0/112, read errors 0, write errors 1, ifcnt 35, reads 15910, writes 16, timeouts 0, DMA errors 0
Moves scheduled 562468, completed 562468, in progress 0, hiccups 0
No step interrupt scheduled
VIN: 24.2V
MCU temperature: min 39.2C, current 41.7C, max 51.2C
Ticks since heat task active 43, ADC conversions started 41546663, completed 41546662, timed out 0
Last sensors broadcast 0x00000002 found 1 47 ticks ago, loop time 0
CAN messages queued 501025, send timeouts 0, received 938189, lost 0, free buffers 36

Any thoughts? Can I contribute with more debug infos to solve this?

[dc42]

@flajzi, it looks as though the heat task stack may still be too small. Please try the tool board firmware at https://www.dropbox.com/sh/jv86vnaqjs1j4qd/AAA00AGN-pHCX4ok4L81ukv-a?dl=0. Post a fresh M122 B1 report when either the problem occurs again or the print finishes.

[flajzi]

@dc42 Hi again. Problem with 2000 °C is back.
It's keeping same temperature, but it reports nonsense. This is M122 output:

Diagnostics for board 1:
Duet TOOL1LC firmware version 3.2-beta4.1+1 (2020-12-12)
Bootloader ID: not available
Never used RAM 3960, free system stack 36 words
HEAT 86 CanAsync 89 CanRecv 83 TMC 54 MAIN 209 AIN 64
Last reset 04:39:13 ago, cause: software
Last software reset at 2020-11-22 15:02, reason: HardFault, available RAM 4168, slot 0
Software reset code 0x0060 ICSR 0x00000003 SP 0x200011c8 Task 
Stack: 20001018 20001304 0000000f 00000001 20000928 000172cd 00018250 01000000 00077dfa 00077ef4 20001218 200032d0 00000020 00017a61 20004660 00000000 20001300 00008895 00077dfa 20004670 00077ef4 42651000 a5a5a5a5 a5a5a5a5 a5a5a5a5 a5a5a5a5 a5a5a5a5
Driver 0: position 7738615, 409.0 steps/mm, standstill, SG min/max 0/112, read errors 0, write errors 1, ifcnt 42, reads 53596, writes 14, timeouts 0, DMA errors 0
Moves scheduled 219075, completed 219075, in progress 0, hiccups 0
No step interrupt scheduled
VIN: 24.2V
MCU temperature: min 38.4C, current 54.5C, max 57.9C
Ticks since heat task active 151, ADC conversions started 16686702, completed 16686701, timed out 0
Last sensors broadcast 0x00000002 found 1 155 ticks ago, loop time 0
CAN messages queued 201492, send timeouts 0, received 370230, lost 0, free buffers 36