MCU Shutdown Rescheduled timer in the past (No Hair left)

Printer Model: Corexy
MCU / Printerboard: PI 3b with SKR E3 v2.0 & SKR E3 V3.0

klippy.zip (6.2 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

Hi,
I have been running klipper for about 3 years, initially on an ender 3 Now I have a heavily modded tronxy x5sa pro using an SKR E3 V3 for X, Y and Extruder, and a SKR E3 V2 for Z and bltouch. The SKRs are attached via USB, the V2 one does not a have power wire in the usb cable, the V3 one has the pin set to isolate the power in the USB.
I have been running it for 12 months with dual Z. All worked well, but I wanted a triple Z, so about 1 month ago I implemented triple Z. I have printed around 20 items since upgrading to tripleZ with out any real issues.
However…
I noticed that z_titl_adjust sometimes failed getting further apart and often took a lot of iterations to complete, so I started looking at probe accuracy and ran lots of PROBE_ACCURACY which seemed to show a large variance.
Googling this I decided to try changing some stepped parameters to make the Z more reliable.
I turned off stealthchop, removed interpolate and set microsteps to 32.
This seemed to work and PROBE_ACCURACY showed a smaller std dev. SO I updated all steppers to use this.
When I tried to run a print, the printer homed ok, ran z_tilt_adust, moved to the front of the bed ready to draw the priming line, then errored with Timer in the past.

I had not come across this before, so spent some time on google trying to work out what was happening.

My version of klipper was up to date, but I had not updated the firmware on the MCUs, so I updated the firmware.
I tracked down the gcode that was cauing the failure by running each command in my start_print manually. I determined that
G1 X0.1 Y200.0 Z0.2 F1500.0 E15
was causing the issue.

so I reset the stepper params and tried again.
But it still failed with Timer in the past.

I looked in my klippy.log . I can see thousands of lines like this
queue_step 42065: t=13805940860 p=345395 i=749 c=24 a=26
immediately before the failure.
It seems very strange that all these messages were generated from so few (2) gcode commands.

I did try all sorts of combinations, but could not get anywhere.

Eventually I decided to revert to a known printer.cfg from 2 days earlier.

BUT
it still fails with Timer in the past.

I don’t have much hair left, and am running out of things to check/change.

so to sum up
I had a working printer.
changed my stepper driver params and caused the Timer in the past shutdown.
Reverted back to old printer.cfg
Still fails with Timer in the past

Any clues on what may be happening ? or how I can dtermine what is really happening ?

Thanks

See Rescheduled timer in the past
Have you tried powering down the board completely after resetting the driver parameters?

Hi, thanks for the idea. The printer and mcus have been off over night, but probably not since I reverted to previous printer.cfg.
So I switched all off for 30 minutes, swicthed on and tried again.
Unfortunately it failed again with Timer in the past on exactly the same gcode command.

I was starting to wonder if its faulty pi, or sd card or usb cable, although they have all been working fine for 12 months.
But why does the klipper.log show literally thousand of queue_step statements, all for one gcode command that moved the printhead 200mm in a straight line. I cannot really see why a faulty pi,sdcard or cable would cause this very repeatable error.
Thanks again
Ian

I think I found the issue for both your error and the insane amount of lines:

grafik

You are advancing a tiny little bit too far

1 Like

You are absolutely right. Thnak you !!!
There should be a decimal point in there

I forgot that I upped that value from 0.075 Should have been to 0.1175.

I have just tested it, and yes all OK

Luckily I havnt ordered a new pi or sdcard yet !!

Thanks again

Ian

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.