Notifiche
Cancella tutti

Accelerometer doesn't work  

  RSS
k1mu
 k1mu
(@k1mu)
Eminent Member
Accelerometer doesn't work

Copying a long message sent to Prusa Support that basically got dropped. (Suggested that I upgrade to the same firmware that I'm already running. Then no further response.)

Any ideas why the printer is throwing "Measurement Failed"?

---------
I recently upgraded my MK4 to a MK4S and added the accelerometer to the order.

It's now upgraded and I'm trying to calibrate using the accelerometer and that's not working.

I run the calibration, first trying the X axis. The printer ramps through the frequency range, vibrating as expected. After it gets to the end of the frequency range, I see "Measurement failed" on the display. No indication of why it's failing.

If I try the calibration by manually sending gcode ("M959 X F35.0 G70.0 N30 H1") I seem to get a valid response. See the capture below.

What's the next step here? I can do this manually (set the step smaller and probe around the resonance), but I'd like to use the menu to do the measurements.

Thanks,

Send: M959 X F35.0 G70.0 N30 H1
Recv: Sample freq: 1299.38
Recv: frequency[Hz] excitation[m/s^2] X[m/s^2] Y[m/s^2] Z[m/s^2] X_gain Y_gain Z_gain
Recv: 35.00 2.51 2.30851 0.21871 0.05778 0.92019 0.08718 0.02303
Recv: 36.00 2.53 2.35284 0.63358 0.19260 0.93137 0.25080 0.07624
Recv: 37.00 2.53 2.12672 1.03154 0.26661 0.83947 0.40718 0.10524
Recv: 38.00 2.53 1.80560 0.42627 0.08098 0.71377 0.16851 0.03201
Recv: 39.00 2.51 1.86599 0.16977 0.05962 0.74211 0.06752 0.02371
Recv: 40.00 2.53 2.19526 0.28267 0.14133 0.86885 0.11188 0.05593
Recv: 41.00 2.53 2.89112 0.56597 0.18463 1.14269 0.22369 0.07297
Recv: 42.00 2.52 3.11459 0.49677 0.24602 1.23377 0.19679 0.09746
Recv: 43.00 2.51 3.32254 0.54764 0.34713 1.32413 0.21825 0.13834
Recv: 44.00 2.53 3.21983 0.53270 0.38681 1.27178 0.21041 0.15278
Recv: 45.00 2.55 3.06031 0.69797 0.44573 1.20096 0.27391 0.17492
Recv: 46.00 2.51 2.59704 0.85535 0.43476 1.03629 0.34131 0.17348
Recv: 47.00 2.51 2.34732 1.06981 0.48302 0.93622 0.42669 0.19265
Recv: 48.00 2.50 2.16499 1.33345 0.49917 0.86553 0.53309 0.19956
Recv: 49.00 2.55 1.89406 1.49735 0.40171 0.74352 0.58779 0.15769
Recv: 50.00 2.53 1.80278 1.67659 0.37242 0.71282 0.66292 0.14725
Recv: 51.00 2.50 1.66055 1.83868 0.32629 0.66345 0.73462 0.13036
Recv: 52.00 2.54 1.61708 2.06612 0.31590 0.63782 0.81494 0.12460
Recv: 53.00 2.56 1.56457 2.06542 0.31136 0.61010 0.80541 0.12141
Recv: 54.00 2.52 1.48187 1.72332 0.33811 0.58846 0.68434 0.13426
Recv: 55.00 2.54 1.25963 1.11877 0.39322 0.49636 0.44085 0.15495
Recv: 56.00 2.55 1.05100 0.43404 0.44530 0.41160 0.16998 0.17439
Recv: 57.00 2.57 0.99420 0.25643 0.54740 0.38755 0.09996 0.21339
Recv: 58.00 2.57 1.05863 0.34340 0.51826 0.41142 0.13346 0.20141
Recv: 59.00 2.58 1.03903 0.37533 0.44749 0.40324 0.14566 0.17367
Recv: 60.00 2.58 1.00680 0.45429 0.32193 0.39084 0.17636 0.12497
Recv: 61.00 2.57 1.00637 0.52541 0.17431 0.39147 0.20438 0.06780
Recv: 62.00 2.56 0.91092 0.37915 0.08742 0.35571 0.14805 0.03414
Recv: 63.00 2.55 0.80354 0.26484 0.03462 0.31558 0.10401 0.01360
Recv: 64.00 2.53 0.69897 0.17599 0.02981 0.27664 0.06966 0.01180
Recv: 65.00 2.50 0.59571 0.13645 0.02633 0.23810 0.05454 0.01052
Recv: 66.00 2.58 0.55336 0.14747 0.01708 0.21452 0.05717 0.00662
Recv: 67.00 2.55 0.43774 0.14377 0.02716 0.17183 0.05643 0.01066
Recv: 68.00 2.51 0.28162 0.10116 0.01038 0.11220 0.04030 0.00413
Recv: 69.00 2.58 0.20695 0.10454 0.02763 0.08008 0.04045 0.01069
Recv: 70.00 2.54 0.13897 0.08353 0.03637 0.05474 0.03290 0.01433
Recv: Maximum resonant gain: 1.32 at frequency: 43.00
Recv: ZV shaper selected Recv: Frequency: 43.00 damping ratio: 0.51203 [...]
Recv: ok

 

Questa discussione è stata modificata 1 month fa da k1mu
Postato : 24/10/2024 11:19 pm
Stefan
(@stefan-18)
Active Member
RE: Accelerometer doesn't work

I did my update to MK4S today and also ordered the accelerometer.
Unfortunately the same error as you describe.

Did you send the gcode by simply creating a new file with ending .gcode and adding M959 into it?
What are the parameters N30 H1, Prusa help does not mention any parameters.

Does this line
Recv: Maximum resonant gain: 1.32 at frequency: 43.00
not say that one has to set 43 manually in the printers menu?

I was wondering that the accelerometer only fits lose into the heater block. How can it work when the accelerometer can move a bit during the progcess?

Postato : 27/10/2024 10:20 pm
k1mu
 k1mu
(@k1mu)
Eminent Member
Topic starter answered:
RE: Accelerometer doesn't work

 

Posted by: @stefan-18

I did my update to MK4S today and also ordered the accelerometer.
Unfortunately the same error as you describe.

Did you send the gcode by simply creating a new file with ending .gcode and adding M959 into it?
What are the parameters N30 H1, Prusa help does not mention any parameters.

Does this line
Recv: Maximum resonant gain: 1.32 at frequency: 43.00
not say that one has to set 43 manually in the printers menu?

I was wondering that the accelerometer only fits lose into the heater block. How can it work when the accelerometer can move a bit during the progcess?

I sent the gcode manually using the terminal in Octoprint.

My accelerometer fits the heater block snugly so it's not that.

For the description of the Accelerometer commands (M959) - see https://github.com/MCDELTAT/MK4_Accelerometer

I've looked at the source code. "Measurement failed" when testing an axis means the spectrum returned has an invalid value. In this case, "valid" means greater than 0.001 and all of my measurements are above that (but some are pretty close at 0.004).

The mystery continues.  I may have to just get into a chat with prusa support.

Postato : 29/10/2024 4:01 pm
Stefan
(@stefan-18)
Active Member
RE:

Yes, strange things.
I printed a new case for the small board which fits better to the nozzle, but still errors.

However, I just finished to do the same as you with proterface.
That was working and I got values which look reasonable. Will send them tomorrow to the printer and am curious if that will make a differenze.

By the way, some of my measurement with pronterface caused an error at the printer, saying the damping ratio is out of range or something.
May be because I do not use the original bearings but some sinter metal ones.

Questo post è stato modificato 1 month fa da Stefan
Postato : 29/10/2024 9:55 pm
Stefan
(@stefan-18)
Active Member
RE: Accelerometer doesn't work

OK, did not work 🙁

The new values caused the printing result to get worse.
Either the accelerometer does not work well or I made some mistake.

Postato : 30/10/2024 4:42 pm
Condividi: