sensorless homing with S4 option behaves same as S3 in 3.4 b7

demonlibra

I strongly recommend that you remove the belt from pulleys before tuning.

I run this program and changing parameters while excecuting:

global counter=0

M569 P0 D3 V20     ; tpwmthrs 40 (117.2 mm/sec)
M569 P1 D3 V20     ; tpwmthrs 40 (117.2 mm/sec)
M906 X1000 Y1000   ; Current mA
M913 X35 Y35       ; Current %
M566 X1 Y1         ; Jerk mm/min
M201 X300 Y300     ; Accelerations mm/sec^2

M220 S20           ; Set speed factor override percentage
M915 X Y S0 H95 F0 R1

;G28
G92 X0 Y0

G0 X20 Y20
while true
  if global.counter=10
    echo "10 pass"
    set global.counter=0
	
  G4 S3
  G1 X145 F6000
  G1 X10 F6000
  M400
  set global.counter=global.counter+1

You must be sure that using StealChop mode for TMC2209. Check speed for switch between StealthChop to SpreadCycle mode.
Decrease motor current to 30-35%. But not too less or you will get errors "Error: short-to-ground reported by driver(s) 0"
Use speed 20-30 mm/sec. Less is better but if you set too low speed stall will not detect.

Theoretical Hnn = <speed> / <length perimeter pulley > x <motor steps per one rotate> x 0.95
For speed 20 mm/sec, motor 1.8, pulley GT2-20
H = 20 / 40 x 200 x 0.95 = 95
You will get false signals if set too low Hnn or not get signals at all if set Hnn to high.

While excecuting change parameters Snn in command M915.

Look to the console. Increase S if you get false messages.
Or decrease S if you touch the pulley but do not get message to console.

Toch the rotating pulley on the motor and find minimum Snn.

https://youtu.be/OI_uVmXZqaI

moth4017

@demonlibra Hi thank you for the reply the issue i have at the moment is that S4 function behaves the same as S3.

moth4017

@phaedrux Hi has DC42 had any thourghts on this?
cheers
Mark

Phaedrux

It's on his list to investigate I believe.

Phaedrux

Can you test S4 homing in 3.4 beta found here?
https://www.dropbox.com/sh/i5vox3xmkd55gaz/AAC19mI0WEC5GmEjLOBRbKs-a?dl=0

moth4017

@phaedrux

Hi Loaded up the beta version 3.4 beta S4 is still acting like S3

Phaedrux

Thanks for testing.

dc42

@moth4017 thanks, this is now on my list to look into.

moth4017

@dc42 Thank you very much

dc42

@moth4017 I have now fixed this in the source code and tested it. The fix will be included in RRF 3.4.0rc1. Thanks for your help in getting this fixed.