Probing gcode G38 VS G30

[Shoki]

Hi,

I have some questions about probing.

I have an E3D Toolchanger with a Duet3, running RRF3.5rc2. It is equiped with 3 tools : 2 printing and one probe (this one).

I'm trying to setup probing procedures to probe printed parts and other parts, for measurement and calibration.

**My question are simple :

• Which command should be used for general probing in XYZ ?
• Why are the results from G38 different from the results from G30 ?**

In the gcode dictionnary, i have found the G38 command which seems perfectly suitable for this.
However, i'm noticing that there is no precision at all with it.

I made some simple tests, starting with just the Z switch which i know is very precise.

The simple script I used is this below. Everything is probed at the same slow speed of 6mm/s in Z.

var number = 10
var xpos = 0
var ypos = -7
 
;Z switch
M558 K0 P8 C"io1.in"  H5 F360 ; Set Z probe
 
G1 X{var.xpos} Y{var.ypos}
 
;Probing using G30 with Z switch
while iterations<var.number-1
    G30 K0 P{iterations} X{var.xpos} Y{var.ypos} K0 Z-9999
    G1 Z5 F360
G30 K0 P{var.number-1} X{var.xpos} Y{var.ypos} Z-9999 S-1
G1 Z5 F360
G1 Z10 F360
 
;Probing using G38 with Z switch
M558 K0 F360 ; Set probe speed
while iterations<var.number
    G38.2 K0 X{var.xpos} Y{var.ypos} Z-5
    echo "Right:"^move.axes[2].machinePosition
    G1 Z5 F360
G1 Z10 360

The results are the following :

For the G30 command :
-0.686,-0.686,-0.688,-0.686,-0.686,-0.686,-0.686,-0.686,-0.686,-0.686
So it's very good, a single value is different and only 0.002.

For the G38 command :
1.158, 0.014, 0.014, 0.014, 0.014, 0.014, 0.614, 1.080, 0.014, 0.024
So, the value is not the same and they range a full millimeter !

I tried to do the same thing with my CNC probe. I managed to do it using the G38.2 command and also had a wide range of measurements.
For the G30 command, the probed crashed into the bed because it was using the machine coordinates and not the active tool coordinates... I will fix it and try again.

So G38 seems very unprecise compared to G30 ? Why is that ? I can use G30 when probing in Z but it does not work for other direction.

Thanks in advance for the help.

[Shoki]

Found the answer to my own question, should have searched the forum first.

@dc42 said in EndStop Accuracy Test:

@Nazar in RRF 3.5.x the machine position is read during moves instead of only at the end of a move, and at most every 250ms. So if you read the machine position immediately after the G1 H4 or G38.2 move has completed, you may gat a value that is up to 250ms out of date.

I suggest you insert command G4 P260 after the G1 or G38.2 command, before you read the machine position.

This solved the repetability problem with the G38.2 command.

I also found this post :

@dc42 said in EndStop Accuracy Test:

@Nazar I think G1 H4 will do exactly what you need.

Will the G1 H4 command work for my application of XYZ probing too ?
In the gcode dictionary fo the documentation, i found this on G1 H4 :

G1 Xnnn Ynnn Znnn H4 Sense endstops while moving, update the current position at which the endstop switch triggers (supported in RRF 3.2-b4 or newer).

But which endstop does it sense while moving ? All of them ? I did not find a parameter to tell him which probe it needs to monitor ?

[oliof]

@Shoki said in Probing gcode G38 VS G30:

But which endstop does it sense while moving ? All of them ? I did not find a parameter to tell him which probe it needs to monitor ?

usually you do this per axis, so G1 Xnnn H4 for X, G1 Ynnn H4 and so on