Tracing down a bad conductor

3DPrintingWorld

I have a Duet 3 and a TLC 1 on a custom printer. I just started seeing hotend temperature spikes. I'm assuming its a broken wire but I'm having a hard time tracing down the wire. You would think that its one of the moving wires but since its the hotend temperature spiking does that mean that the bad wire is between the thermistor and the toolboard? Or would I also see the temperature spiking if there was a short in the power or Can wires?

jens55

@3DPrintingWorld, 'spiking' would imply 'going high'. A short would not do that.
If this is just a terminology issue and you are referring to a temperature DROP then I would look at two aspects. How is the heater connected to the Duet? Are you using any connectors or is the only connector at the PCB. I have had many issues with connections overheating causing poor connection which then causes heating which then causes even more issues. Look at all connections to make sure the contacts are not discoloured.
I have also run into 'gentle' temperature dropping which can be caused by a defective thermistor. Rapid changes (ie instantaneous) could indicate a bad connection someplace since the hot end can not change instantaneously.
A defective thermistor might also cause sharp drops.
I have seen a number of issues and most of them relate to either the thermistor or bad connections on either the thermistor or the heater.

dc42

@3DPrintingWorld if the sensor is a thermistor then positive spikes in the temperature reading suggest either an intermittent short circuit between the thermistor and something else, or perhaps electrostatic discharge between the hot end metalwork and the thermistor wire. Make sure the hot end metalwork is connected to Duet ground either directly or through a resistor. Version 1.2 and 1.3 tool boards have plated mounting holes connected to ground through resistors, so you can use one of those to ground the hot end metalwork.

3DPrintingWorld

@dc42 Here is what the spikes look like. Moving the thermistor cable around does not seem to trigger it. I can only get it to trigger while running g-code. Would you get a short on a stationary wire, I understand the wires in the cable track.

Its been running great with the toolboard for over a year, would a grounding issue show up after this long? I also dismounted the board and I don't see how the mounting holes are plated, unless I'm missing something. Its a 1.2., should there not be some kind of metal around the holes?

3DPrintingWorld

@jens55 I would think if the short was in the thermistor connector, when I move the connector around I would also get the temperature spikes but I only seem to get them when running g-code.

Phaedrux

Are you able to connect the thermistor directly to the Duet and test if it has the same issue?

3DPrintingWorld

@Phaedrux Going back to my original question, I unplugged the two CAN wires and I get the same temperature spike until I plug it back in. As a result of this test, I think I can assume that this is not a thermistor issue but a short in the CAN bus wires. I'll replace these wires and see what happens.

If the controller is not communicating with the toolboard, does it not know? Why is there no error/alarm?

dc42

@3DPrintingWorld if the CAN connection is lost then after a short time the temperature will read 2000C. In RRF 3.5.0 and later an event will be generated to report that the board has stopped sending status. The default handler for that event will report it to DWC and any other commands channels e.g. USB.

So if the temperature spikes are 2000C then either a temperature reading error or a CAN disconnection has occurred. To find out which, send M308 S# where # is the sensor number, and see what is reported for "last error".

3DPrintingWorld

@dc42 I have not updated my firmware as its been working great, I was on 3.4 which must be why I did not get a report when I unplugged the connector. The temperature is showing 2000C when it spikes.

I took the wires out of the track and let them dangle so I could try to reproduce the short by manually manipulating the wires. Then my plan was to replace them but the spikes completely stopped. I put the wires back in the track and still no temperature spikes. Its so strange, if/when it continues I'll try replacing the wires. Its strange that it stopped all the sudden and that I could not cause the short by manually manipulating the wires.