Calibration issues

[Ididnttouchthat]

[robertsamples]

Some text explanation would be helpful here, we don't know how these photos are related and what you did previously in terms of calibration, or what version of cosmos you are using.

You should run the LOAD_CELL_TARE command with nothing on the bed which is likely your issue with probing.

you should turn the fans off prior to running IS

[Ididnttouchthat]

So once it initially booted it prompted for a full calibration started that, that when I got the error lis2dw It cooled down to 49c the hot end fan turned off but it never ran the input shapping, power cycled dismissed calibration ran bedmed mesh manually that's when I got the load cell probe error quick Google search adjusted the tare/baud rate based on Google went positive first still failed that's when I made it negative then it was actually able to run the bedmesh manual input shapper command finished no problem set z offset have not ran a print yet tho I kinda figured picture would be self explanatory that's why I didn'twrite any thing and this is my first time ever posting on GitHub but I figured I want this project to actually work so I'd throw my 2 cents in also using 0.0.4 the current one

[robertsamples]

ok I would not adjust the baud rate as that will prevent the boards from communicating properly. But if you were able to get bed mesh calibrated and z offset you should run save_config and then reboot the printer. Sometimes things can be a bit strange on the first boot or two after install. After that you can try to run IS again and it should work.

[jbarron16]

I’m having a similar issue on the 0.0.6 update.

I originally started on 0.0.3 and was able to run the full calibration macro without any issues. After updating, the macro started failing during the Input Shaper section with this error:

WARNING: Active fans detected: extruder
Fan vibrations may affect input shaper calibration accuracy.
For best results, stop fans or wait for them to stop automatically.

Invalid lis2dw id (got ff vs 44).
Invalid lis2dw id (got ff vs 44).
This is generally indicative of connection problems
(e.g. faulty wiring) or a faulty lis2dw chip.

I also tried running the input shaper G-code commands manually and got the same Invalid lis2dw id error.

The weird part is that if I remove the M400 call, I’m able to run both X and Y input shaper successfully. I’m not sure why M400 would cause that behavior, but it seems directly related in my case.

[dari0cm]

@jbarron16
Same issue here, but I found a workaround

I had the exact same error on Cosmos 0.0.6:

Invalid lis2dw id (got ff vs 44)

It happened every time during the "Calibrate All" routine, right after the bed mesh finished, input shaper never ran. I tried power cycling multiple times, no luck.

However, running SHAPER_CALIBRATE manually from the Mainsail console worked fine:

SHAPER_CALIBRATE AXIS=X
SHAPER_CALIBRATE AXIS=Y
SAVE_CONFIG

No reboot needed, the LIS2DW initialised and the calibration ran without issues. So the chip itself is working, something in the automated calibration sequence is leaving the LIS2DW in a bad state before it gets to the input shaper step.

My guess is either the heating steps or the load cell calibration interfere with the LIS2DW's power state or I²C bus. If you skip the automated routine and just run the three manual calibrations separately (load cell, bed mesh, then shaper), everything completes as expected.

[jamesturton]

@dari0cm that's a really good insight, thank you! I guess we can try adding a small delay in the calibration routine and hopefully that will help people out!

[dari0cm]

@jamesturton Is there any reason input shaper is calibrated after extruder PID? I belive it would be better to do input shaper first, that way the nozzle is still cold and extruder fan not running.
I belive currently it's Load cell > PID > Z-offset/mesh > Input shaper. Maybe Load cell > Z-offset/mesh > Input shaper > PID would be best?

Just curious.

[jamesturton]

The macros are getting a pretty big overhaul here: #155
Not sure it that PR changes the order though. Maybe @suchmememanyskill or @Atomique13 has thoughts?

[suchmememanyskill]

It doesn't, but might be worth changing, yeah. Maybe load cell > shaper > bed > pid?