Does anybody know if there happens a kind of communication between Host and MCU at the end of any movement, which can provoke such behavior, if it fails anyhow?
I’m thinking about changing the communication from the USB Port to a GPIO connection. Might this be more reliable and stable?
I looked at one of your earlier logs. It indicates that the host got delayed some how and the command to enable the Z motor became severely delayed. This resulted (correctly) in the “timer too close” error. This indicates a problem with the host computer (software, power, sd card, etc.).
Some suggestions:
- It would help if you could return the printer to a more stable setup. Update to the latest Klipper code and make sure you are using a pristine version of Klipper (no gcode_shell_command.py modifications). Run a
G28command and then attach the full Klipper log file here. - If it is failing, try going back to an earlier version of Klipper and see if that also fails. This can typically be done by ssh’ing into the host and running something like:
cd ~/klipper; git checkout v0.12.0; sudo service klipper restart. It may be necessary to compile (make menuconfig ; make) and reflash the micro-controllers after changing klipper versions. If it fails, please attach the full klipper log file here. You can also try going back to the previous version (cd ~/klipper; git checkout v0.11.0; sudo service klipper restart). - Can you summarize the host hardware that you are using (hardware, OS revision, and similar).
-Kevin
1 Like
Hi Kevin,
yes, i agree to say the problem is on the Host side, since every piece behind the MCU and the MCU itself was already changed or moved across the printer without changing a bit.
let me start with your last topic:
_,met$$$$$gg.
,g$$$$$$$$$$$$$$$P.
,g$$P" """Y$$.". OS: Debian GNU/Linux 11 (bullseye) aarch64
,$$P' `$$$. Host: Raspberry Pi 3 Model B Plus Rev 1.3
',$$P ,ggs. `$$b: Kernel: 6.1.21-v8+
`d$$' ,$P"' . $$$ Uptime: 18 hours, 14 mins
$$P d$' , $$P Packages: 957 (dpkg)
$$: $$. - ,d$$' Shell: bash 5.1.4
$$; Y$b._ _,d$P' Terminal: /dev/pts/0
Y$$. `.`"Y$$$$P"' CPU: BCM2835 (4) @ 1.400GHz
`$$b "-.__ Memory: 147MiB / 859MiB
`Y$$
`Y$$.
`$$b.
`Y$$b.
`"Y$b._
`"""
/=======================================================\
| ~~~~~~~~~~~~~~~~~ [ KIAUH ] ~~~~~~~~~~~~~~~~~ |
| Klipper Installation And Update Helper |
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ |
\=======================================================/
/=======================================================\
| ~~~~~~~~~~~~~~ [ Update Menu ] ~~~~~~~~~~~~~~ |
|-------------------------------------------------------|
| a) [Update all] | | |
| | Installed: | Latest: |
| Klipper & API: |---------------|--------------|
| 1) [Klipper] | v0.12.0-102 | v0.12.0-102 |
| 2) [Moonraker] | v0.8.0-294 | v0.8.0-294 |
| | | |
| Klipper Webinterface: |---------------|--------------|
| 3) [Mainsail] | v2.9.1 | v2.9.1 |
| 4) [Fluidd] | | v1.27.1 |
| | | |
| Touchscreen GUI: |---------------|--------------|
| 5) [KlipperScreen] | | |
| | | |
| Other: |---------------|--------------|
| 6) [PrettyGCode] | | |
| 7) [Telegram Bot] | | |
| 8) [Obico for Klipper]| | |
| 9) [OctoEverywhere] | | |
| 10) [Mobileraker] | | |
| 11) [Crowsnest] | v4.1.3-1 | v4.1.3-1 |
| |------------------------------|
| 12) [System] | System up to date! |
|-------------------------------------------------------|
| B) « Back |
\=======================================================/
Klipper
Host(aarch64, 64bit)
Version: v0.12.0-102-g9f41f53c
This is the SanDisk Ultra 64GB SD card I have used for the new setup with MainsailOS 1.3.2. The Klipper installation is clean. If you wish me to use a specific OS please point me to the image.
Here comes your first topic:
I switched everything off for a minute, cleared the browser cache and opened the Mainsail interface, set my config to 32 microsteps, not using stealth chop and interpolation, then called G28.
I got the error message after z-movement as expected.
Then restarted the firmware and downloaded the klippy.log as attached here.
240131_klippy.log (1.5 MB)
for your second topic i need some time … will come up with the results later.
It looks like you recently compiled and flashed your main mcu with an incorrect clock frequency. The code is reporting that the main mcu was compiled for 180Mhz, but the host is detecting that it is actually running at 270Mhz. This can lead to all sorts of issues (and makes it very hard to debug other possible issues).
What board are you using for the main MCU and what options did you select when running “make menuconfig”?
-Kevin
EDIT: Oh - I see that you are using Octopus v1.1 with stm32f446 processor. Be sure to select “low level options” and select a 12Mhz chip as specified in the directions at the top of config/generic-bigtreetech-octopus-v1.1.cfg . Please attach a new log after compiling and flashing the mcu.
4 Likes
Just for reference:
HEAD ist jetzt bei 0d67d9c4 docs: Note the release of v0.12.0
MCU Protocol error
Your Klipper version is: v0.12.0-0-g0d67d9c4
MCU(s) which should be updated:
mcu: Current version v0.12.0-102-g9f41f53c
rpi: Current version v0.12.0-102-g9f41f53c
=> made and reflashed mcu & rpi
needed to power cycle to reconnect, called G28, got error
240131_klippy_v0_12.log (1.8 MB)
Vorherige Position von HEAD war 0d67d9c4 docs: Note the release of v0.12.0
HEAD ist jetzt bei e02b7256 docs: Note the release of v0.11.0
Internal error during connect: can't convert negative number to unsigned
=> made and reflashed mcu & rpi
power cycled again to reconnect, but didn’t.
mcu 'rpi': Unable to connect
240131_klippy_v0_11.log (2.4 MB)
flashed RPi again and sudo service klipper restart to no avail.
git checkout v0.10.0
Vorherige Position von HEAD war e02b7256 docs: Note the release of v0.11.0
HEAD ist jetzt bei 2a964a46 docs: Note the release of v0.10.0
making not possible
$ make menuconfig
make: python2: Datei oder Verzeichnis nicht gefunden
make: *** [Makefile:109: /home/me/klipper/.config] Fehler 127
240131_klippy_v0_10.log (2.4 MB)
git checkout master
=> made and reflashed mcu & rpi
Version: v0.12.0-102-g9f41f53c
needed to power cycle to reconnect, called G28, got error
240131_klippy_master.log (2.6 MB)
I hope this is sufficient. If not, tell me what i should try…
B.t.w.: I’m really impressed how you can see something in the ‘number-jungle’ of the log-files, since I cant. Do you have tools to analyze them?
Wait a minute, please!
Kevin,
YOU ARE MY HERO!
machine is homing again! Thank you very much!
240131_klippy_clock.log (2.7 MB)
1 Like
Phew … that was a cruel pitfall!
If i may make a suggestion: Is it possible to set this as default when choosing the stm32f446 in make menuconfig.
I think it must have been in earlier version, since i never used the “low level options” before.
@LifeOfBrian
… and since to your encouragement much snappier now. I’m going to fulfill my promisis now.
1 Like
Thatswhy I was already asking whether the 270 MHz are correct or not. Unfortunately I was too lazy googling the answer and I’m running the PRO with I think 500 MHz.
This was a very long story that again showed the complexity of that whole business.
As we often do have such topics maybe @Sineos can somehow take this into the basic troubleshooting steps or highlight it at another kb article?!
@limefrog happy homing and printing! 
My Z homing works as well and even Z_Tilt works with the triple Z upgrade. Just have to get the Z motors much quieter and fire up the new bed once the PE cable is installed.
I didn’t understood the point of your hint, since i wasn’t aware that it is possible to compile the firmware for different crystal speeds.
you are my hero also!
1 Like
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.