TMC Overtemp, ShortToGND, and OpenLoad errors

General Discussion
1

Basic Information:

Printer Model: MatterHackers Pulse XE
MCU / Printerboard: Azteeg X5 GT (LPC1769) with TMC2660’s
Host / SBC : Raspberry Pi 3 B+
klippy (4).log (3.9 MB)

Fill out above information and in all cases attach your klippy.log file (use zip to compress it, if too big). Pasting your printer.cfg is not needed
Be sure to check our “Knowledge Base” Category first. Most relevant items, e.g. error messages, are covered there

Describe your issue:

Seemingly randomly, I keep getting these error messages:

TMC 'stepper_z' reports error: READRSP@RDSEL2: 000e0360 ot=1(OvertempError!) otpw=1(OvertempWarning!) s2gb=1(ShortToGND_B!) ola=1(OpenLoad_A!) se=24 sg_result@rdsel2=1

and

TMC 'stepper_x' reports error: READRSP@RDSEL2: 00ffffff stallguard=1 ot=1(OvertempError!) otpw=1(OvertempWarning!) s2ga=1(ShortToGND_A!) s2gb=1(ShortToGND_B!) ola=1(OpenLoad_A!) olb=1(OpenLoad_B!) stst=1 se=31 sg_result@rdsel2=31

and

TMC 'stepper_y' reports error: READRSP@RDSEL2: 00000000 se=0(Reset?)

Now I know the third one is similar to one of the pages in the Knowledge Base category, however mine is se=0, not se=1. Also it seems possibly relevant to the first and second errors.

So far I have gotten each of these errors on all 4 steppers connected to my machine. X, Y, Z, and extruder. They happen while homing, just jogging axes, and soon after starting a print. To be clear, all steppers do work for a certain amount of time before it throws the error.

The first and second errors mention overtemp, so I turned off my machine for an hour or so, came back and sure enough, after a minute of just jogging and homing I get the error again. Felt the motors and heatsinks, and they were slightly warm but nothing bad.

A few hardware things I have tried:

  • Two different power supplies, both 12v and 24v
  • Checked resistances on all stepper wires
  • Pulled and put all TMC drivers back in (idk man, im just trying everything now)
  • Messed around with currents to the drivers
  • Replaced power supply connections (they were a bit shotty and needed to be replaced anyway)

Any help would be appreciated. Thanks!

2

Hello @addamm1 !

Have you tried actively cooling the drivers (heat sink + fan)?

3

This amount of errors and also basically all errors such TMCs can produce would point to a fundamental issue like dead drivers or issues with the board.

I would try stripping down the config to just one driver / stepper and test all available drivers, potentially switching between slots / motors etc.

4

I definitely could, but should I not be able to tell if they are getting warm just by feeling them? They have heatsinks installed already, and they barely get warm. Also, I tried running them at just 0.4 amps each, and it still happpened.

5

Good idea. I’ll try this later. How far can I strip it down? If I have it set up to be cartesian, I have to have all three steppers configured, no?

6

Oh, it seems that I just had to read. Looks like I would use kinematics: none. I’ll try this later.

7

Hmm, doesn’t seem to be as simple as I originally thought. I figured kinematics: none might allow me to supply only one axis, but I guess i’m wrong.

[mcu]
serial: /dev/serial/by-id/usb-Klipper_lpc1769_1040FF0EC70406AF6E7B5D58C12000F5-if00

[printer]
kinematics: none
max_velocity: 300
max_accel: 4200

[stepper_x]
step_pin: P2.1
dir_pin: P0.11
# enable_pin: P0.10
microsteps: 32
full_steps_per_rotation: 200
rotation_distance: 40
endstop_pin: !P1.24
position_endstop: 0
position_max: 235
position_min: -2
homing_speed: 100
homing_retract_dist: 5
homing_retract_speed: 50
second_homing_speed: 15

[tmc2660 stepper_x]
cs_pin: P0.10
spi_bus: ssp0
run_current: 0.4
sense_resistor: 0.11
idle_current_percent: 50

Any ideas on how best to test each axis separate from the others?

8

Probably the easiest way is to have your original config and comment out all [tmcXXX ...] blocks except one, and use the FORCE_MOVE command.

9

So, I went through the smoothieware, and marlin configs provided by Matter Hackers for this printer, and have come up with an unfortunate conclusion. Both just ignore all TMC errors, which are options in their config files. After getting a beautiful 5 hour print out of the printer on marlin, tmc steppers only ever getting barely warm, I queried the steppers and all of the errors were there. Just being ignored by marlin.

Which brings me back to klipper. I assume you guys have not changed your stance of not adding a way to ignore TMC errors?

I am not sure what could be causing the issues, however they seem to be fake errors or non-issues for one reason or another.

10

Forgot to mention, I did this with all steppers and had no issues individually, however as soon as I added in a second one, the errors started popping up.

11

IMO, this would absolutely be the wrong thing to do.

  1. The errors are there, and they should not be if everything was OK.
  2. Since not everything is OK, it is fair to assume that broken things have consequences - be they immediately visible or hidden failures.
  3. Hidden failures are the most problematic, as they are virtually impossible to diagnose.

This could point to some issues with the SPI bus.

  • Check if any other devices are on the same bus.
  • Try setting the bus parameters as spi_software_*_pin.
12

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.