persistent Driver 2 error: phase A short to ground

[SoundWorksNW]

Hello everyone,

I am encountering an issue with the Z axis/ driver 2 and am hoping to tap into your wisdom.

Machine: Cartesian 3D printer using 4 Z motors, 2 X motors, 1 Y motor
Board: Duet 2 WiFi (2WiFi)
Firmware: RepRapFirmware for Duet 2 WiFi/Ethernet 3.4.2 (2022-09-13)
Duet WiFi Server Version: 1.27
Duet Web Control 3.4.2

The machine (and board) in question has run reliably for about a year and a half. This firmware has been running for about 3 weeks. Yesterday while starting the second print of the day, during the homing sequence the Z axis emitted a loud stuttering sound and triggered the following:
-Driver 2 error: phase A short to ground
-Warning: Driver 2 warning: phase A short to ground, phase B may be disconnected

Troubleshooting steps taken:

1. Disconnected cables from all drivers on board.
2. Repeated following process for each test:
   a) Shut off power to board
   b) Connect test cable and test motor to driver output to be tested
   c) Turn on power to board
   d) Allow un-homed movement
3. Test Driver output by commanding arbitrary movement to connected axis
   a) Z-A result:
   i) Action: No motor movement
   ii) Faults:
   (1) Driver 2 error: phase A short to ground
   (2) Warning: Driver 2 warning: phase A short to ground, phase B may be disconnected
   b) Z-B result:
   i) Action: No motor movement
   ii) Faults:
   (1) Driver 2 error: phase A short to ground
   (2) Warning: Driver 2 warning: phase A short to ground, phase B may be disconnected
   c) Y MOT result:
   i) Action: Normal motor movement
   ii) Faults: None
   d) X MOT result:
   i) Action: Normal motor movement
   ii) Faults: None

So, the fault seems to be inherent to Driver 2 and not related to the cables or motors. I don't see anything on the board that looks obviously wrong to me, though I will not claim to have a well-trained eye.

Could anyone recommend other tests to run?
Does this mean the driver is ruined or could there be something else going on?

Thank you you in advance for any help.

configJan2023g.g

(edited formatting for readability as indentations disappeared)

[dc42]

@SoundWorksNW unfortunately it sounds to me that the driver has developed a short circuit in one of its 8 output transistors. Here are a few things to check:

• Check that nothing conductive underneath the board could be touching the back of the board in the vicinity of that driver or the output connectors
• Check that there is no metal swarf between the pins of that driver (I can't see any in your photo)
• With nothing (not even jumpers) connected to either ZA or ZB, command movement and see if short circuit is still reported. Expect to see "phase may be disconnected" messages.

It's possible to replace the driver chip, but removing the old one requires either a hot air tool or low melting point solder.

[SoundWorksNW]

@dc42

Thank you for the helpful reply. I did confirm nothing touching the back of the board, absence of metal swarf, and that the fault to ground occurs even when nothing is connected to ZA and ZB. It is nice to have the problem diagnosed.

A replacement and spare Duet 2 are enroute, but I would also like to attempt a repair.

I did not see any proximate components which looked to me like transistors (had 3 leads). Are the transistors integrated to the chip and therefore does a repair constitute replacement of the TMC 2660-PA?

[T3P3Tony]

@SoundWorksNW yes the TMC 2660-PA has the FETs built in so thats what needs to be replaced.