Bed levelling (G32 S3) skips the final move and probe

gtj0

When this gets run viaG32 S3, the G30 P2 move and probe is skipped. Then when the G28 XY is run, it moves the carriage correctly but gives G91 Homing Failed and leaves X unhomed and in a weird state. If I then home X manually, the movements are so slow I thought the printer was stuck in Busy mode.

; from config.g
; Axis Limits
M208 X0   Y0   Z0   E0  S1                    ; Set axis minima
M208 X460 Y450 Z465 E-1 S0                    ; Set axis maxima

; Endstops

M574 X1 S1 P"io0.in"
M574 Y1 S1 P"io1.in"

; Z-Probe
M574 Z1 S2
M558 C"io8.in" P5 I1 R1 F300 H5.0 T6000 A4 S0.05
G31 P500 X0 Y0 Z0
M557 X0:460 Y0:450 S25                  ; Define mesh grid

;Lead Screw
M671 X-23:238:498 Y13:485:13 S5

; bed.g
; called to perform automatic bed compensation via G32
M561
G29 S2
G28
G1 Z2
G1 X2 Y2 
G30 P0 X2 Y13 Z-99999 ; probe near a leadscrew
G30 P1 X238 Y450 Z-99999 ; probe near a leadscrew
G30 P2 X460 Y13 Z-99999 S3 ; probe near a leadscrew and calibrate 3 motors
G28 XY

kraegar

@gtj0 If your leadscrews adjust in the wrong direction the most likely thing is that you have your stepper wires in the wrong ports. Pay careful attention that:

your leadscrew definition is the the correct order for your driver connections
and
your leadscrew location definitions match that order.

For mine my leadscrews are arranged like this:

1
          2
0

and are plugged into drivers 2, 3, 4 so my config is:
M584 X0 Y1 Z2:3:4 E5 ; Map X to drive 0 Y to drive 1, Z to drives 2, 3, 4, and E to drive 5
M671 X-10:-10:283 Y22.5:227.5:125 S7.5

Also note the drivers on the duet3 are in this physical order
0 1 2
5 4 3

Hope this helps!

fcwilt

Hi,

I have found that explicit move speed values are usually advisable.

Also you don't that M561 - G29 S2 does the same thing and is preferred.

As to the other issues I need some clarification please:

Why are you doing G32 S3?
Why are you doing that G1 X2 Y2 near the middle of bed.g?
Why are you doing that G28 XY at the end of bed.g? You have already done a plain G28 above homing X,Y & Z.

Thanks.

Frederick

T3P3Tony

What happens if you use M400 to ensure the moves are completed?

Also is this the latest firmware and DSF? I had a similar issue with an older version of DSF

gtj0

@fcwilt said in Bed levelling (G32 S3) skips the final move and probe:

Hi,

I have found that explicit move speed values are usually advisable.

Yep but this bed.g file worked perfectly on my Duet2 even with RRF3 so something's different.

Also you don't that M561 - G29 S2 does the same thing and is preferred.

True, but doing both is safe.

As to the other issues I need some clarification please:

Why are you doing G32 S3?

That was actually a copy and paste error. I don't actually do the S3.

Why are you doing that G1 X2 Y2 near the middle of bed.g?

It's not the middle it's just after the homing and I was testing moving the tool off of the endstops.

Why are you doing that G28 XY at the end of bed.g? You have already done a plain G28 above homing X,Y & Z.

No reason other than moving the tool back to 0,0.

Thanks.

Frederick

gtj0

@T3P3Tony said in Bed levelling (G32 S3) skips the final move and probe:

What happens if you use M400 to ensure the moves are completed?

Also is this the latest firmware and DSF? I had a similar issue with an older version of DSF

Yep, M400 worked. I'm using DSF 1.1.0.0. Maybe an old bug crept back in?

fcwilt

@gtj0

Hi,

Thanks for the explanations.

Frederick

gtj0

OK so now the moves are completing but the correction never converges. The leadscrews are being adjusted but they keep moving in the wrong direction. I'm wondering if the adjustment magnitudes are correct but maybe the signs are being applied backwards.

Leadscrew adjustments made: -0.052 -0.252 -0.448, points used 3, deviation before 0.285 after 0.000
Leadscrew adjustments made: -0.030 -0.427 -0.819, points used 3, deviation before 0.506 after 0.000
Leadscrew adjustments made: 0.008 -0.773 -1.562, points used 3, deviation before 0.948 after 0.000
manually moved leadscrew 1 to bring the bed further away from the nozzle.
Leadscrew adjustments made: 0.051 -0.813 -2.292, points used 3, deviation before 1.317 after 0.000
Leadscrew adjustments made: 0.149 -2.160 -4.496, points used 3, deviation before 2.705 after 0.000
The bed is quite tilted now.

T3P3Tony

@gtj0 i think @kraegar had this working

kraegar

@gtj0 If your leadscrews adjust in the wrong direction the most likely thing is that you have your stepper wires in the wrong ports. Pay careful attention that:

your leadscrew definition is the the correct order for your driver connections
and
your leadscrew location definitions match that order.

For mine my leadscrews are arranged like this:

1
          2
0

and are plugged into drivers 2, 3, 4 so my config is:
M584 X0 Y1 Z2:3:4 E5 ; Map X to drive 0 Y to drive 1, Z to drives 2, 3, 4, and E to drive 5
M671 X-10:-10:283 Y22.5:227.5:125 S7.5

Also note the drivers on the duet3 are in this physical order
0 1 2
5 4 3

Hope this helps!

gtj0

Duh. The first and third motors were reverse. I'm going back to bed.
Thanks @kraegar !

gtj0

Are you guys upvoting the fact that I'm an idiot???

fcwilt

@gtj0

Nope - we are just glad to hear you got it working.

We've all made foolish mistakes from time to time.

Frederick

kraegar

I knew what it was right away, because I get this exact question on our printers about twice a week or so. It's an easy mistake to make!

gtj0

Actually, when I was mounting the Duet3 I made a mental note that the drive connectors were ordered clockwise from the top left. I just couldn't find it when I was hooking up the connectors.

T3P3Tony

@gtj0 nope, just adding to your karma