ugs probe is not in the expected initial state

[bvernham]

Version

2.1.9

Hardware / Firmware

GRBL 1.1

What happened

I keep getting this error with UGS "probe is not in the expected initial state" which is very confusing becuase I have never had and issue with probing before. I have probed with bCNC and with EASEL with zero issues. With UGS I am getting (ALARM:4) Probe fail. Probe is not in the expected initial state before starting probe cycle when G38.2 and G38.3 is not triggered and G38.4 and G38.5 is triggered. The machine is zeroed with everything hooked up (clip to the bit) and to the touch plate. The machine probes down to the touch plate. Makes contact, backs off, then set the alarm. Probe cycle looks OK until the makes contact (like it should) then it fails? Never had thie problem with any there g code sender.

How to reproduce

See above. Tried multiple times, same issue.

Operating System

Windows 10

Anything else

Every time I probe. I wouldn't make this an issue if I had no probing experience with other applications. But with Easel you can use machine inspector and the machine check out as expected. The limit switches only set when you push them (as expected) and the probe is registered only when the touch plate is manually touching the bit.

[breiler]

Can you post the log messages from the console that shows the error.

Double check your probe settings, specifically the "Retract amount", you may have to increase that to something like 5 mm.

[bvernham]

G10 L20 P1 Z0

G21 G91 G49
G38.2 Z-90 F100

ok
ok
ok
[PRB:0.000,0.000,-82.823:1]

G91 G21 G0 Z2.0

G21 G91 G49
G38.2 Z-90 F10

ok
ok
ok
ok
ok
**** The communicator has been paused ****
An error was detected while sending 'G38.2 Z-90 F10': (ALARM:4) Probe fail. Probe is not in the expected initial state before starting probe cycle when G38.2 and G38.3 is not triggered and G38.4 and G38.5 is triggered. Streaming has been paused.
ok
Error while processing response : An unexpected command was completed by the controller.

I tried a bigger #....

[bvernham]

Now I am getting this:

G10 L20 P1 Z0

G21 G91 G49
G38.2 Z-90 F100

ok
ok
ok
[PRB:0.000,0.000,-82.844:1]
ok
ok

G91 G21 G0 Z3.0

G21 G91 G49
G38.2 Z-90 F10

ok
ok
ok
**** The communicator has been paused ****
An error was detected while sending 'G21 G91 G49': (ALARM:2) Soft limit alarm. G-code motion target exceeds machine travel. Machine position retained. Alarm may be safely unlocked. Streaming has been paused.
[MSG:Reset to continue]

[breiler]

Yes, you are probing down 90mm, which will go beneath you machines soft limits. You need to decrease the probe depth.

[bvernham]

My limit was 90 but I guess that the probe is being done relative and not absolute so the machine is seeing 90 - 90 = -180.

I had the machine set at 110 which is real but now I have it set at -200 and the probe is working fine. Does not see like a real safe way ot doing this to force set a large soft limit.

Can't the measure proble know where it is at and only need to probe down a small limit.

Also, why is the sender doing this: G10 L20 P1 Z84.423

Should it just return to a safe Z?

Thanks

[breiler]

I could probably improve the second probe cycle to just probe down the retract amount + a couple percent (instead of the full probe depth) to lower the risk of making the second probe exceed the soft limits.

The command G10 L20 P1 Z84.423 sets the new work zero Z offset to the probed position (in machine space).

[bvernham]

Shouldn't it be G10 L20 P1 Z1.61 (the thickness of the probe plate)?

[breiler]

If it is not working as you are expecting, can you instead describe what is happening? Preferably with the console log and your current probe settings.

[bvernham]

I am not sure why the spindle is going all the way back to the top of travel when the work to be done is at Z=0? Should the spindle just go to a safe Z?

[bvernham]

I put a macro in like this and it seems to do the trick:

G21 G91 G49;G38.2 Z-10 F50; G10 L20 P1 Z1.61;G4 P0.1;G0 Z5

[breiler]

I am not sure why the spindle is going all the way back to the top of travel when the work to be done is at Z=0? Should the spindle just go to a safe Z?

And what is safe Z? If the starting point was considered safe to start probing from, it makes sense to move back there as this will work for all probing scenarios (Z, XY, XYZ and Hole center). Otherwise we would need a setting for a "safe" X, Y and Z in all of these probing scenarios.

The algorithm for probing in UGS is like this:

• Probe towards Z with fast rate
• Probe touch - stop and retract
• Probe towards Z with slow rate
• Probe touch - record measured Z distance
• Retract Z to starting point
• Reset Z zero based on measured Z distance

probing.mp4

As I mentioned before, if you are using a large probe distance it might trigger a soft limit (if that is activated on the machine) alarm on the second slow probe. I will need to fix that, so that it probes the second time with just the retract distance.

I put a macro in like this and it seems to do the trick

Great, than use that! It does things a little differently though:

• Probe towards Z
• Probe touch - stop
• Reset Z zero minus probe plate thickness
• Move to safe height Z5 (based on the new Z zero)

[bvernham]

Ahh...so the trick is to start the probe from what is considered "safe" point, and then that is where it will return?

So home, jog to safe position, probe?

Let me give that a shot becuase that I think is the step I missed.