MCU discconnecting, rpi not respoding

Basic Information:

Printer Model: flsun SR
MCU / Printerboard: BTT skr 1.4 with RPI 3b
klippy.logtoo big, must upload it somewhere klippy

Hello. Because freezing problem with my SR and flsun speeder pad, I decided to move to RPi.
I installed new board, rpi 3b. And issue still persist, so it wasnt speeder pads fault. Checked cables, connectors, using original rpi 3b PSU, replaced PSU in printer, replaced USB cable, installed better cooling to board. And still verry randomly my printer is shutting down mid print without any sign of error. Cant SSH into Pi in this stage, need to unplug it from power, then I can SSH and login to mainsail. Included log. I really dont know what to do anymore.
I just wondering, if heating stepper motor on extruder could cause something like this? I am using lgx lite with nema 14 pancake inside chamber (chamber temp 50C) ad nd stepper motor is getting really hot (even ASA cable holder softened). But it should be good up to 180C according to datasheet for stepper motor. Also I am pushing only 0.4 vref into it. I generated graphs from log, but I dont understand them.

Describe your issue:

Are the PSUs inside the heated chamber?

RPi PSU is outside, printer PSU is under the printer, basically under the bed. But it have fan + it is insulated. Not directly in the chamber. Forgot to mention, it was freezing even at winter where it was relativelly cold in my basement + even when printing long PLA prints without enclosure

And the printer board?

At the top, also insulated. Will post a picture

It appears in the log, that a sudden power loss happened.

Hard to say if the PSU failed or the plug was pulled.


Cables are messy rn, i am still moving things to check everything. On the photo top plate with another fans.

I personally think that the power loss was me unpuging RPi from outlet to reboot it.

How full is the SD card of the RasPi?

I maybe the Pi had an issue writing to it.

Fresh new 32gb sd card with 28 gigs of space left. UHC1 class 10. RPi is brand new, same the BTT board

It is the first time it happened that way?

First time with RPi and BTT board. At least 50 times before with speeder pad and MKS board. Replaced it yesterday

Installed small fan on the nema 14 stepper to hopefully keep it more cool. Started same 2 day long print, didnt inserted filament, already wasted almost 10 kilos… Lets see. If anybody have any sugestion, please write here, i will check.

So 2 day long print without filament finished. Yesterday at 10pm I inserted filament and started printing. Now something strange happened. Cant connect to mainsail, cant SSH into pi and I cant see pi in the wifi setting. For some reason it disconnected. But print continues, so hopefully it will be OK, cant monitor it from my bedroom through. I think I will connect it through LAN cable, both Pi and new manta for new printer have LAN, so it will be probably better

Printer stopped at around 10pm with error like this… But at least it gave me something hopefully useful… Another 0.5kg of ASA wasted


Found online that info could be hidden inside syslog maybe? Uploaded it. I am not expert, but could mobileraker or klipper screen cause that? See something about it in syslog.

klippy.log (32.2 KB)

See Timer too close for potential reasons.

If I understand it correctly (I am not native english speeaker, so forgive my stupid question), this error is related to something inside Pi or USB between pi and my skr 1.4? It has specifically something to do with RPi, not BTT SKR 1.4 board, right?

Typically the error is caused if the host (your RPi) cannot fulfill the timing requirements. The potential causes are listed in the linked article.

Just to give an update. Replaced SD card and USB cable, got rid of kiauh and mobileraker. Till now everything works like it should. Still need to make 2x 2day long prints, hope it will finish. Thanks for help

1 Like

To prevent timing issues, I’ve changed the way Linux schedules the klipper.service in order to increase its priority. This way, other services (from kiauh) will not delay klipper.

# ...
[Service]
# ...
# Increase scheduling priority for the service
# https://www.freedesktop.org/software/systemd/man/systemd.exec.html#Scheduling
Nice=-20
CPUSchedulingPolicy=rr
CPUSchedulingPriority=99

@Sineos, If it makes sense, we could increase the priority by default. Have you seen other people mentioning this problem going away after getting rid of extra services? Or of issues appearing when loading a frontend after some period of inactivity?