Plan to switch from normal steppers to 1HCL+ magnetic encoders

[martin7404]

Just in case downgraded back to 3.5.4 and the result is the same

[martin7404]

I am stupid these are 0.9 mottors
tried to edit M569.1 with S400 parameter and the calibration went well , but after that the motor entered in such big resonance in stoped position , that crashed the whole board

1/20/2025, 8:37:55 PM	Connection established
1/20/2025, 8:37:25 PM	Connection interrupted, attempting to reconnect...
Network error: Request timed out
1/20/2025, 8:37:17 PM	Warning: Driver 50.0 warning: position tolerance exceeded
1/20/2025, 8:37:16 PM	m17
1/20/2025, 8:36:28 PM	Warning: Driver 50.0 warning: position tolerance exceeded
1/20/2025, 8:36:27 PM	Connection established
1/20/2025, 8:36:12 PM	Connection interrupted, attempting to reconnect...
Network error: Request timed out
1/20/2025, 8:36:03 PM	m569.6 p50.0 v2
Driver 50.0 calibration succeeded, measured backlash is -0.084 step
Original encoder reading errors: min -12.3, max 14.8, rms 5.2
Corrections made: min -9.0, max 10.0, rms 5.1
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
Warning: Driver 50.0 warning: position tolerance exceeded
1/20/2025, 8:35:37 PM	m569.6 p50.0
Error: M569.6: Driver 50.0 calibration failed (no reason available)
1/20/2025, 8:35:14 PM	m569.1 p50.0
Encoder type: rotaryAS5047
Magnetic encoder motor steps/rev 400, agc 80, mag 4715
PID parameters P=100.0 I=0.000 D=0.000 V=1000.0 A=0.0, torque constant 1.00Nm/A
Warning/error threshold 1.00/2.00
1/20/2025, 8:35:03 PM	m17
1/20/2025, 8:34:47 PM	Upload of config.g successful after 0s
1/20/2025, 6:39:33 PM	Connection established

[jay_s_uk]

@martin7404 grab an M122 and M122 B50 so we can see whats the cause of the reboot

[martin7404]

@jay_s_uk Reboot is fixed I uped the error in m596.1
It now just gives a warning when I send m17
if I reset the tuning 569.6 p50.0 v4 it goes away
so the problem seems to be 0.9 steppers
it performs calibration and after that goes in resonance

[martin7404]

@jay_s_uk here is new calibration

and then the m122 p50.0

m122 b50
Diagnostics for board 50:
Duet EXP1HCL rev 1.0a or earlier firmware version 3.6.0-beta.3 (2025-01-15 18:54:31)
Bootloader ID: SAME5x bootloader version 2.4 (2021-12-10)
All averaging filters OK
Never used RAM 51588, free system stack 184 words
Tasks: EncCal(1,nWait 6,0.3%,183) Move(3,nWait 7,0.0%,167) CLSend(3,nWait 6,0.0%,149) TMC(4,nWait 6,66.5%,317) HEAT(2,nWait 6,0.1%,105) CanAsync(5,nWait 4,0.0%,66) CanRecv(3,nWait 1,0.0%,65) CanClock(5,nWait 1,0.0%,64) MAIN(1,running,32.0%,253) IDLE(0,ready,0.0%,29) AIN(2,nWait 2,1.2%,255), total 100.0%
Owned mutexes:
Last reset 00:05:44 ago, cause: software
Last software reset data not available
Moves scheduled 0, hiccups 0 (0.00/0.00ms), segs 0, step errors 0 (types 0x0), maxLate 0 maxPrep 0, ebfmin 0.00 max 0.00
Phase step loop runtime (us): min=26, max=65, frequency (Hz): min=4807, max=18292
Peak sync jitter -7/7, peak Rx sync delay 190, resyncs 0/0, next timer interrupt due in 2 ticks, enabled, next step interrupt due in 4036578614 ticks, disabled
VIN voltage: min 24.2, current 24.3, max 24.3
V12 voltage: min 12.0, current 12.0, max 12.1
MCU temperature: min 28.4C, current 29.1C, maDriver 0: pos 690, 320.0 steps/mm, position tolerance exceeded, failed to maintain position, SG min n/a, mspos 4, reads 25157, writes 8 timeouts 0
Last sensors broadcast 0x00000000 found 0 89 ticks ago, 0 ordering errs, loop time 0
CAN messages queued 2421, send timeouts 0, received 3688, lost 0, ignored 0, errs 0, boc 0, free buffers 38, min 38, error reg 0
dup 0, oos 0/0/0/0, bm 0, wbm 0, rxMotionDelay 0
Closed loop driver 0 mode: closed loop, pre-error threshold: 10.00, error threshold: 20.00, encoder type rotaryAS5047, position -1966
Encoder reverse polarity: no, full rotations -1, last angle 14418, minCorrection=-9.0, maxCorrection=9.2, agc 79, mag 4747, no error
Tuning mode: 0, tuning error: 0, collecting data: no
Accelerometer: none
I2C bus errors 0, naks 0, contentions 0, other errors 0

[martin7404]

I think I found the culprit.
Originally due to heavy gantry my motors LDO-42STH60-MAC were runing at 1700 mA max current cutoff (2 Amps max)
I reduced the current to 1200 mA and now it is without vibrations in closed loop mode

 1/20/2025, 10:05:28 PM	m569.6 p50.0 v3
Driver 50.0 calibration check succeeded, measured backlash is -0.067 step
Residual encoder reading errors: min -5.0, max 11.3, rms 1.4
1/20/2025, 10:04:47 PM	m569.6 p50.0 v2
Driver 50.0 calibration succeeded, measured backlash is -0.074 step
Original encoder reading errors: min -13.8, max 14.1, rms 5.7
Corrections made: min -9.6, max 9.2, rms 5.6
1/20/2025, 10:04:11 PM	m17
1/20/2025, 10:04:05 PM	M906 X1200 Y1200 I0

I still do not know if these numbers are good

[dc42]

@martin7404 that backlash number is good and the other readings looks OK to me. It's more difficult to get good readings using 0.9deg motors than it is using 1.8 deg motors which is what we recommend.

If you want to increase the motor current beyond 1200mA without getting instability, try reducing the PID P parameter in the M569.1 command from its default value.

[martin7404]

@dc42 Does that mean 0.9 steppers will have problems running in closed loop ?

[dc42]

@martin7404 they should run OK, however the speed won't be as high as you can get using 1.8deg motors and the tuning will be more critical. With a closed loop motor it's the encoder that determines the resolution, not the motor.

[martin7404]

@dc42 Thank you . I get this but these are on the machine. Any guidelines about PID tuning. I managed to make it today following the wiki guide and my graphs are more "noisy" is this to be expected. It is a big machine with 0.6 and 0.8revo nozzles. But I limit the acceleration a lot , hopin to improve this by going closed loop