Probe not getting the correct X0 Y0 but zips ok

[StuartH]

Dave
Help I have now got the hardware for the TP and the probe set up and working reliably no funny things are hapenning , I normally run my machine with 0 debounce but for test I have tried with a lager setting but no improvement

The problem is that doing a Z probe leaves the Z0 correct , probe z then move off do a MDI g0 z0 levels the probe centre at the z0 ( part face ).

The dia setting for the probe is 4mm

When doing a X or Y probe with the xy clearance set to 5 mm,
The X and y zeros when doing a MDI X0 or Y0
Leaves the axis 2mm off the part face in both cases

A corner probe gets the Z ok but off in X Y

I bet it's a setting that I have missed but it seems that it's not picking up the probe tip dia

Stuart

[StuartH]

Dave I have been testing

I does not matter what I sett the probe Dia to it still shows a error

It's a constant 1mm error DRO says 10mm in Y AFTER probing but centre of probe is 11mm

Z is ok from the tip that's the point it trips at ,checked with my Jo blocks

Note I do not get any mis movement ,it is doing the correct movements .


Stuart

[StuartH]

Dave a bit more info

I have carried out the tests set out in your paper to heist the hardware with g90 g31 g91 etc
And the system passes all the tests

Using the probe as a maser tool as in your video
If I set any dia into the tool setting the z is wrong by half the setting eg set at 4 and then probe do the tool change and put in g0z5 the tool is 3mm above the part but set the dia to 0 then repeat and the tool is at 5mm

The system does what it should but it's not picking up the dia and applying it

Stuart

[DaveCVI]

HI,
Please refresh me re the basics of your situation:
1) MSM version
2) Mach version
3) motion control device in use?

a) Is the tool diameter set in the tool table for the probe tool #?
b) Have you calibrated the probe to compensate for cosine error of the probe shaft etc?
- probe tip is centered in the spindle?
- probe has been calibrated via MSM calibration procedure?

Dave

[StuartH]

Dave

I will get back to you bit busy at this time.

1. Current
2 latest
3 don't know but it's USB and came fitted to the machine ( but a pound to you know what it's the problem

the others
Yes the dia has been set
The calib will not as the system will not apply the above

Preset sit rep
I have uninstalled MSM and mach3 ( after deactivate ) and reinstalling both

Probe / tool setter are on two pins , probe on one and TS on timing ,
But try probe ver throws up an error probe and TS on two different pins edit forgot to shut down Mach to save the pin change

But further testing will have to wait for a few days

[StuartH]

Dave
Found some time

Ok
I have done the tests for probing as set out in the current manual, I have also done the tests set out in your paper dated April 5 2013, mach3 motion devices and probing support
Results of all test is that the system passes all the tests

But it is still failing to apply the dia from the tool table
The Z is ok I think because it uses the PLO of the probe and ignores the dia.
The Y Z are both off

More tests on the circular cal. Routine used a bearing of 35mm I'd with the probe tip set correctly it says the bore is 31.65mm
And will not correct it

BTW. Master tool mode to the tool setter is perfect ( not using the probe but a normal master tool )

The probe is aligned axialy with the spindle to 0.1mm

Before you ask the probe is home made and the trip point all ways resets to 0.001 mm with a starret Dti 0.002 per div
To clarify set the test indicator to the tip zero it, nudge the tip and the indicator retuned to zero , this is the case in any direction
Stuart

[StuartH]

Unless you can come up with any ideas the probe plan is on the back burner

Whatever it's doing its consistent it's always 2 mm out in X and Y
One thing though Z probe finds the surface and then continues for 2mm half the tip dia but the X and Y do not

If it is the error that the motion board is doing it will have to stay as is due to machine being still under warranty
So if I need to swap out it will have to wait

Stuart

[DaveCVI]

I'm a bit stumped on this one....

The only thing I can think to do for testing on my end is to retest the X,Y motion to double check things. I can get to that next week after the holiday weekend.

In the mean time, please provide the exact version numbers you are using for Mach and MSM - that way I can know that I am testing the same combination (instead of assuming that "current" means the same thing to you and me).

I will also need to know what motion control device your machine is using. If is has firmware revisions etc, please also tell me that info.

The X.Y tip diameter compensation is handled internally to mach. There are Mach registers that MSM sets to tell mach the tip diameter and mach does the compensation for the tip diameter when reporting the trigger position at the end of a G31 move.
I've never had that be incorrect when reading the trigger position from mach, but the symptom sounds like the tip compensation is note getting done by mach. I don;t know how that could happen.. unless...

Some motion control devices are interfaced to mach via the use of a plugin. If a plugin is being used, then all kinds to things could be different as the plugin interface allows a motion device manufacturer to get at low levels of the mach data structures - which makes for the ability to do many things and also to mess up many things. I'm not accusing the (unknown to me) motion device here, just speculating from the little info I have at the moment.

Dave

[StuartH]

Dave

I have taken the board out to find out the info , alas it looks like it's a leafboy job

As to the software data
MSM 2.3.6
Mach 3 3.043.66 the latest they posted so that's why I said current eg up to date

If you think that it's the card don't waste your time , I do understand that the plugin can interfere with things .

The thing that puzzled me was that the error was very repeatable but it would not follow changes to tip settings , but remained the same

My conclusions for What it's worth is that the contact point is recognised repeatedly hence the results from the calibration test I set the hint to 35mm but it measured 31.6xx with the tip dia set to 4mm

As I said above if you think that it's the card/plugin that's responsible please don't waste your time
Regards
Stuart

[DaveCVI]

Uh Oh.. one of those...

I think the controller is the culprit here. Please see this thread:
http://www.calypsoventures.com/forums/v ... p?f=6&t=39

Which notes re the leafboy:
"Company: Various Chinese venders
Devices: USB Breakout Board (leafboy77)
Status: MSM users have reported many problems with these boards when attempting to do probing. The same mach/MSM combinations that work correctly on PP and Smooth Stepper boards, fails with these boards.
Additionally, CVI has received reports of these boards randomly causing unexpected motion."

Looks like we can add broken probe tip compensation to the list of flaws for the leafboy.

If you want I could probably make a little test script that verifies the problem (as there are registers in mach to get both the compensated and uncompensated trigger point coords). I'm thinking it would show that the two sets of registers are returning the same values and that would tell us that the device is not doing the correct calculation. That would give a test case to show the problem to the vendor...
But with the "no responsible vendor" Chinese devices there is no one I can contact to work with to get this behavior corrected.
So I don't see how the test script would be of any use.

I'm afraid that you have a bad motion control device.

Dave