[FluidNC Issue#1506] Macro variable for probe trigger point

#1 – MitchBradley 于 2025-05-27

> Currently, FluidNC macros can only see the position after deceleration,

Are you sure about that?

#2 – massiveradiation 于 2025-05-27

Hey Mitch,
I’ll send you some concrete info on it tomorrow.
Cheers,
Tyler

On Tue, May 27, 2025 at 12:00 AM Mitch Bradley @.*>
wrote:

> MitchBradley left a comment (bdring/FluidNC#1506)
> <https://github.com/bdring/FluidNC/issues/1506#issuecomment-2911027472>
>
> Currently, FluidNC macros can only see the position after deceleration,
>
> Are you sure about that?
>
> —
> Reply to this email directly, view it on GitHub
> <https://github.com/bdring/FluidNC/issues/1506#issuecomment-2911027472>,
> or unsubscribe
> <https://github.com/notifications/unsubscribe-auth/AGUKYUVPK5FOXZJA4SOT4233APPOFAVCNFSM6AAAAAB56ZO2MGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMJRGAZDONBXGI>
> .
> You are receiving this because you authored the thread.Message ID:
> @.*>
>

#3 – bdring 于 2025-05-27

Is it different than the [PRB: …] value?

Note: FluidNC also has a hard stop option for probes

#4 – massiveradiation 于 2025-05-27

Hey Bart,
I was trying to create a safety measure for my TLO macro: probe once at
faster rate, record that value, then reprobe much slower, record that value
and then compare the two. Set a tolerance and if out of that tolerance
range, then alarm out.

The terminal print outs of probez1 and probez2, would always be about
0.01mm or so difference, my toolsetter is decent. But, when comparing the
values later in the tolerance check code (see below), the value was
consistently about -0.48mm, which is not correct. Maybe I’m missing
something.

Of course, if I set to something higher than 0.48mm, it works.
Thoughts?

I also got a response from Mitch, asking, “are your sure?” Ha! I’m not
100%, but I don’t know what other explanation for this might be. I even
created little delays before the prints and tried that to see if there was
some kind of data buffering. Should I also send this reply to Mitch or does
it automatically update on github. This is my first feature request so I’m
trying to do things right. Thx

The tlo code:

#probez> = 300 ; Probe feedrate for first (coarse) probe (mm/min)
#finez> = 50 ; Probe feedrate for second (fine) probe (mm/min)
# = 0.05 ; Max allowed diff between probes (mm, positive)

; ==================== PROBE TOOLSETTER ===================

(print,Probing toolsetter…)
G53 G0 X#x> Y#y> F#
G53 G0 Z#rapidz> F#rapidz>

; —– First probe (coarse, fast, downward) —–
G38.2 Z[#depth>] F#probe_z>
G90
#z1> = #<absz>
(print,probez1: #z1>)

; —– Retract upward from setter (relative move) —–
G91
G0 Z#retract> F#rapid_z>
G90

; —– Second probe (fine, downward from above setter) —–
G38.2 Z[#depth>] F#fine_z>
G90
#z2> = #<absz>
(print,probez2: #z2>)

; —– (Tolerance check is disabled – FluidNC cannot access true probe hit values) —–
; #diff> = [#z1> – #z_2>]
; (print,Probe diff: #diff> Tolerance: #tol>)
; o301 if [[#diff> GT #tol>] OR [# LT [0 –
#]]]
; (print,ERROR: Double-probe difference # exceeds tolerance
# mm! Aborting.)
; $Alarm/Send=10
; M0
; o301 endif

#probez> = 300 ; Probe feedrate for first (coarse) probe (mm/min)
#finez> = 50 ; Probe feedrate for second (fine) probe (mm/min)
# = 0.05 ; Max allowed diff between probes (mm, positive)

; ==================== PROBE TOOLSETTER ===================
(print,Probing toolsetter…)
; —– First probe (coarse, fast, downward) —–
G38.2 Z[#depth>] F#probe_z>
G90
#z1> = #5063 ; True probe trigger point
(print,probez1 (trigger): #z1>)

; —– Retract upward from setter (relative move) —–
G91
G0 Z#retract> F#rapid_z>
G90

; —– Second probe (fine, downward from above setter) —–
G38.2 Z[#depth>] F#fine_z>
G90
#z2> = #5063 ; True probe trigger point (second/fine probe)
(print,probez2 (trigger): #z2>)

; —– Tolerance check using true probe trigger points —–
#diff> = [#z1> – #z_2>]
(print,Probe diff: #diff> Tolerance: #tol>)
o301 if [[#diff> GT #tol>] OR [#diff> LT [0 – #tol>]]]
(print,ERROR: Double-probe difference #diff> exceeds tolerance #tol> mm! Aborting.)
$Alarm/Send=10
M0
o301 endif

On the fluidnc probe wiki I found this line of code: #ztouch>=#5063 ; save the z touch WCO location

Which is a bit misleading because what’s actually being stored is the Z machine coordinate upon a G38, yes?

#10 – MitchBradley 于 2025-05-27

    // last probe
    axis = id - 5061;
    if (is_axis(axis)) {
        float probeposition[MAXN_AXIS];
        motorstepstompos(probeposition, probe_steps);
        result = toinches(axis, probeposition[axis]);
        return true;
    }

Looks like mpos to me.

#11 – massiveradiation 于 2025-05-27

Maybe an update on the wiki for clarity…

Instead of this: #ztouch>=#5063 ; save the z touch WCO location

Maybe this: #ztouch>=#5063 ; save the z touch mpos (machine position)

What does WCO even stand for? Work Coordinate “”?

Thanks

#12 – bdring 于 2025-05-27

Yes, Work Coordinate Offset.

I’ll check the wiki

#13 – massiveradiation 于 2025-05-27

OK Gents,
The problem seems to have been me using #<probez> instead of #5063! After a lot of testing with #5063, the difference between my toolsetter probes, the first one at F200 and the 2nd at F50, is never more than 0.01mm and is often almost 0! So, please close this feature request and thanks for walking me through it.
Cheers,
Tyler