G30- Height correction question

kuhnikuehnast

If I probe 2 points to level the bed driven with 2 independent z- axis, is the H- parameter included in the correction process?

I run 2 independent Z-Axis motors. Yesterday I installed 2 "probing" points (outside the bed, which is floating on silicon dampers) and aligned the bed perfectly. The one probing point is 0.24 mm higher than the other.
If I now run the following code:

G30 P0 X-9.957 Y239.998 H0 Z-99999 ; probe point P0
G30 P1 X332.043 Y239.998 H0.241 Z-99999 S-1 ;S2 - Adjusting; S-1 Report

the Duet always tells me that there is a difference of ~0.241 and doesn't seem to integrate the H-value in the correction???. I am afraid now to run a G30 with S2, so the bed gets "bent" again and I'll have to readjust everything...

Hope you can help me!

greetings kuhni

dc42

Yes, H parameters are taken into account in all G30 Px bed probing operations.

kuhnikuehnast

This post is deleted!

kuhnikuehnast

@dc42 said in G30- Height correction question:

Yes, H parameters are taken into account in all G30 Px bed probing operations.

that's really strange!
I now tried these 2 macros:

G30 P0 X-9.957 Y239.998 H0 Z-99999 ; probe point P0
G30 P1 X332.043 Y239.998 H0 Z-99999 S2 ; H=0

G30 P0 X-9.957 Y239.998 H0 Z-99999 ; probe point P0
G30 P1 X332.043 Y239.998 H0.241 Z-99999 S2 ; H=0.241

I ran both macros directly one after the other. These are the results I got:

Leadscrew adjustments made: -0.013 0.075, points used 2, deviation before 0.044 after 0.000

Leadscrew adjustments made: -0.008 0.003, points used 2, deviation before 0.004 after 0.000

So it seems, that the H-Factor isn't calculated??? Both results are nearly identical but at the 2nd macro there should be at least a difference about ~0.241 mm???

dc42

Hmm, that doesn't look right. Can you try a large H parameter, perhaps 1mm if that won't damage your printer, and test again?

Which firmware version are you using?

kuhnikuehnast

@dc42 said in G30- Height correction question:

Hmm, that doesn't look right. Can you try a large H parameter, perhaps 1mm if that won't damage your printer, and test again?

Which firmware version are you using?

It really seems not to pay any attention to the H-factor at all... I now tried different settings with H from 0 up to 5 and i always got ~ the same result...

This is the code I used (in a macro as well as in bed.g):
[CODE]
G1 Z10 ; lift Z=10 to prevent crashing the probing point

G1 X15.50 Y228.60 F15000 ; go to defined probing point independent of floating bed
G30 X15.50 Y228.60 ; define this point as Z=0

G30 P0 X-9.957 Y239.998 H0 Z-99999 ; probe point P0
G30 P1 X332.043 Y239.998 H0 Z-99999 S2 ; go to defined probing point 2 and do: S2 - Adjusting; S-1 Report

G1 X45.457 Y8.602 F15000 ; go to probing point on floating bed
G30 X45.457 Y8.602 ; define this point as Z=0
[/CODE]

dc42

What type of Z probe are you using?

kuhnikuehnast

@dc42 said in G30- Height correction question:

What type of Z probe are you using?

It is an original Bl-Touch. Homez / homeall works great also with the defined Offset...

kuhnikuehnast

So is this a firmware fault or is there something wrong I made? Tried to level the 0.241 mm by aluminium foil, but this didn't work that fine...
greetings kuhni

dc42

I just checked the code and also re-tested the manual bed levelling adjustment wizard, which uses the same code as the leadscrew adjustment. It's working for me, using firmware 2.02RC3.

Which firmware version are you using?

kuhnikuehnast

@dc42 said in G30- Height correction question:

I just checked the code and also re-tested the manual bed levelling adjustment wizard, which uses the same code as the leadscrew adjustment. It's working for me, using firmware 2.02RC3.

Which firmware version are you using?

I am now also on 2.02 RC3.
It works for me as well, but still the H factor isn't calculated. I already wrote it in your release post of the 2.02RC3 (if you want to, we could also discuss it there?)

2 examples:

G30 P0 X20 Y20 H0 Z-99999	  ; probe point P0
G30 P1 X20 Y20 H0 Z-99999 S-1	  ; go to defined probing point 2 and do: S2 - Adjusting; S-1 Report

Results:
G32 bed probe heights: -0.005 -0.008, mean -0.006, deviation from mean 0.001

Here are both H-factors 0 for both points. This should result in no difference.

Now the 2nd test:

G30 P0 X20 Y20 H0 Z-99999	  ; probe point P0
G30 P1 X20 Y20 H1 Z-99999 S-1	  ; go to defined probing point 2 and do: S2 - Adjusting; S-1 Report

Results:
G32 bed probe heights: -0.003 0.002, mean -0.000, deviation from mean 0.002

Here is P1 with H1 instead of H0. So the report should be at least at a difference of ~1 mm. But The results are still almost the same?
Is there maybe an interference with the G31 command as there is a Z- Trigger Z height defined?

dc42

What I meant was that the H corrections work for me. This is my usual bed.g file:

M561		; clear any bed transform
G0 X20 Y0 Z28 F10000

G30 P0 X-80 Y-100 Z-99999
G30 P1 X80 Y-80 Z-99999
G30 P2 X0 Y75 Z-99999
G30 P3 X-80 Y200 Z-99999 
G30 P4 X80 Y200 S3 Z-99999

G0 X0 Y50 Z28

Running this gives me:

G32
Manual corrections required: 0.00 turn up (0.00mm) 0.03 turn up (-0.03mm) 0.05 turn up (-0.05mm)

If I change bed.g to this:

; bed.g
M561		; clear any bed transform
G0 X20 Y0 Z28 F10000

G30 P0 X-80 Y-100 Z-99999
G30 P1 X80 Y-80 Z-99999
G30 P2 X0 Y75 Z-99999
G30 P3 X-80 Y200 Z-99999  H1
G30 P4 X80 Y200 S3 Z-99999 H1

G0 X0 Y50 Z28

then it gives me:

G32
Manual corrections required: 0.00 turn up (0.00mm) 0.63 turn up (-0.63mm) 1.22 turn up (-1.22mm)

So the H values are definitely having an effect.

kuhnikuehnast

@dc42
I think that there may be a problem because I am using 2 independent leadscrews...?

The important values of my config.g are:

; Leadscrew setting
M584 X0 Y1 Z4:2 E3				   			; Define drive mapping- Z= driver number 2 and 4.
M671 X-77:297 Y20:20 S3			           			; Set position of the leadscrews

; BLtouchZ-Probe
M307 H3 A-1 C-1 D-1                                			; Disable the 3th heater on PWM channel for BLTouch
M558 P9 H5 F180 T15000                             			; Set Z probe type/mode 9. H=Dive Height. F=Speed the bed moves
G31 P25 X+26.9 Y5.935 Z1.065                      			; Set Z probe trigger value, offset and trigger height

Now did the same test again:

M561
G30 P0 X20 Y20 Z-99999
G30 P1 X200 Y20 S-1 Z-99999

and

M561
G30 P0 X20 Y20 Z-99999
G30 P1 X200 Y20 H20 Z-99999 S-1

Results were:

G32 bed probe heights: 0.147 0.155, mean 0.151, deviation from mean 0.004
G32 bed probe heights: 0.137 0.145, mean 0.141, deviation from mean 0.004

So it seems, that as soon as you define independent leadscrews, the H-factor isn't taken into account any more?

greetings kuhni

kuhnikuehnast

something else, I just found out:
I can do a trick:

M561					; clear any bed transformation
G31 P25 Z0				; set trigger height to 0
G30 P0 X20 Y20 Z-99999			; probe 1. point
G31 P25 Z10 		     		; set trigger height to 10 
G30 P1 X200 Y20 S-1 Z-99999 		; probe 2. point and tell result
G31 P25 X+26.9 Y5.935 Z1.065 		; correct trigger height to standart

Results:

1. Both Z = 0:
G32 bed probe heights: 1.205 1.227, mean 1.216, deviation from mean 0.011
2. 1. Z = 0; 2. Z = 10:
G32 bed probe heights: 1.205 -8.767, mean -3.781, deviation from mean 4.986

So this seems to work...?

dc42

I think I have spotted the problem. It looks to me that H parameters on G30 commands may not work if your system uses a deployprobe.g or retractprobe.g file. If I confirm this, I will fix it in the 2.02 release.