Strange results using our XTR Head.

[[Ri...]]

I was conducting a test on location with a Stylus that has a single tip in the -Y axis.
I created a Sphere on the Reference Sphere as my part.
Used a simple 5 points for a Base Alignment.
I created an Alignment from a scanned Sphere using X,Y and Z as Zero location from the same axis.
Then using a copy of the same sphere (Set to the Alignment noted above), I added a pattern to run the scan 3 times.
Copying the Sphere once again, I reversed the tool direction from -Y to +Y axis and ran the same pattern.
The results caught my right away.
The first iteration of the pattern was an average of 0.00040948' (Inch) off location in the X axis.
After rotating the XTR 180°:
The initial X axis values for the first pattern (through all three iterations) is .00000585" (Inch).

I have attached a ZIP file with three spread sheets showing the results.

I am hoping a fellow member who has an XTR head could test this scenario on their CMM.
This will tell us:
1. I have an issue with my CMM and or XTR sensor.
2. There is a bug when using patterns?

Measurement plan 2.zip

[[Ri...]]

Hmm. Very interesting.

I similarly have done some testing on the XTR for repeatability while rotating. The two biggest contributors to the increased error in rotation is unbalance and heat generated while rotating.

10µm is a lot of error though and you shouldn't see it on the initial rotation unless there is either an issue with the plate/sensor or it is unbalanced. If the stylus system is unbalanced, I have seen huge deviations in rotation.


To test your plate/sensor, you can swap over to a MasterProbe style stylus system, and repeat the test. This will help you eliminate the plate/sensor at least.

[[Ri...]]

Also, did you verify that the stylus was qualified properly before performing the test? While your deviation is large, it doesn't appear to move.

[[Ri...]]

Please sign in to view this quote.

I have Referenced the Master Sphere and qualified the probe 4 times.
Nothing in the qualification data would have made me think there was any kind of issue.

[[Ri...]]

Getting a couple odd issues since this has been noticed.
1. Manually inserting a stylus system sometimes generates a weight warning and requires removing and re-inserting the stylus system. I have not seen the weight warning while automatic probe change is in use.
2. This error, 88 function not possible with this stylus plate (ID chip). I have seen this when using old style XT (non-rotating adaptors) with the XTR sensor, but now I am getting them (twice) while using an XTR adaptor in the XTR sensor.

[[Ri...]]

(These posts are mainly for notes now).
This Weight warning seems to be on both of our Cantura's using XTR Sensors only when using XTR Adaptors (weight is 250g compared to XT Adaptors which are 158g).
Info: XT Adaptors have 3 locating pins. XTR Adaptors have 18 locating pins.
This warning only happens when manually loading a Probe and it has an XTR Adaptor, the first stylus will qualify, once it rotates, the warning pops up.

The work around, assign the probe to a holder on the probe rack and have Calypso pick up the probe prior to use.

XTR Sensor Weight Warning.PNG

[[Ch...]]

Here is what I think. The taration mechanism in the probe head is past its limit. It's trying to weigh the probe, but doesn't have enough stroke length left. Think of it like a screw that has bottomed out.

Here is how to fix it: take your heaviest probe and insert into XTR head. You want it as near the weight limit as possible. When you pick this probe up you will hear the taration mechanism really spool up. Now, there is much more travel available for that screw that weighs the probe. You should be good after this.

[[Wo...]]

Rich - I'm seeing both errors that you do.

1. The weight issue - I see this only on the brand new Contura. But it practically only happens when I setup a new stylus system.
I have never seen it on the two Accuras that also have XTR. I will try Chad's suggestions when I have some time.
2. 88 error - it mostly happens when the contact points are dirty on either the stylus system plate or the head. The night shift inspector who is more technically inclined has bought some electric contact cleaner and from the time cleans the contact points and the issue disappears.

As for different results - is it possible that the probe has low rigidity? I would test the result with probe as stiff as possible ( my best ones have a rigidity of around 60 ).
Also do you calibrate the probes with the sphere at 180 degrees and calibrate them in single run?

I have to admit that I have seen some weird calibration results on XTR and it was always on "Y" axis probes for some reason. But it was a while ago and can't remember the details.

[[Ri...]]

Please sign in to view this quote.

Thank you.
Is the weight limit 500g?

[[Ch...]]

Yep, 500g. (picture stolen from Richard's post)

[[Ri...]]

Well, I ran the same test (added two non-pattern scans) on our other Contura, both were bought at the same time, their first calibration isn't due until September 2024.
They share similar issues running the same test program (attached if anyone wants to run the program and show your results).
The program runs in the Y axis, reporting X and Z axis location using a Ø1.5 x 30L stylus (Same name probe, but not the same build).
The first X axis coordinate for the Pattern (Sphere3) and non-pattern (Sphere3) have about .0004+ (Inch) deviation from nominal while all the other X axis values are less than .00010 (Inch).

Strange. 🧑‍💻

Measurement plan 2_Contura3_1.xlsMeasurement plan 2.zip

[[Ni...]]

Please sign in to view this quote.

Have you tried adding filters to your program?
Also i don't understand why you use Sphere1 +Y axis for your base alignment(X, Y, Z) and then create a secondary alignment(Alignment1) and use the same Sphere1 +Y axis for the same axis (X, Y, Z) and then use alignment1 for the remaining sphere features you got?
Also how about doing a helix strategy? or adding a point at the very tip of the sphere?

Another thing would be doing circles instead of a sphere. Would that change anything?

I feel like sphere has different minimums it must be met so they are calculated properly. Having a point at the very tip of it is one i assume, but you have 3 circles.

[[Ri...]]

Please sign in to view this quote.

I have used this strategy for about 20 years. It has never yielded results like I am seeing now.
On an XT Sensor, the probe would have 5 styli of the same size.
On an XTR Sensor there would be one stylus in the -Z axis and one in the +Y axis, rotating around the Z axis for the remaining three stylus locations.
Each stylus will measure the sphere and report a Spherical True Position.
The Reference Sphere is moved around the CMM in 10 places.
X.............X
........X
X.............X
5 runs at Granite surface level.
5 Runs raised 10 inches above the granite.
Every two to three weeks you run a single test in a single location, by the end of one year you should have ten total reports.
This tells me if the CMM may have any kind of issue volumetrically (such as I am experiencing now).
The tests are quick and simple.
All the data is loaded into Excel and charted.
I have used this data to validate changing yearly calibrations from 1 year upwards of 3 years, generally 2-year intervals for calibration.
All of our normal qualifications are performed in various locations as described above. This data too is entered into the same Excel file.

[[Ni...]]

Please sign in to view this quote.

So you have been running this exact same test and program for the past 20 years to catch if your machine is deviating?
And it is showing that it is?

So wouldn't the logical thing be that your machine is deviating? Call in zeiss support to come have take a look. Especially if this is a program you have ran in the past.

I thought this was a brand new program you were trying to set up.
But if its something that you have been using for years without any issues, then i wouldn't spend much time looking at the program. It is doing exactly what it was program to do. It is telling you there is something wrong with the mechanicals of your machine.

[[Ri...]]

Please sign in to view this quote.

Its either a CMM/Sensor issue or a bug, I don't know which yet.
The same issue presented itself the same way on our identical Contura (Both purchased/Installed at the same time).

We will have Zeiss come out; I like to present as much information as possible.

[[Ni...]]

Please sign in to view this quote.

Post update on what they find. Interested to know how it gets fixed.

[[Ri...]]

This seems to happen a lot with brand new $1,055.00 XTR Adaptor Plates.
Seems to only happen when loaded manually.

Capture.PNG

[[Wo...]]

Please sign in to view this quote.

Did you do the update that's mentioned on the box? Device Activator or something like this?
Maybe it would help.

[[Ri...]]

(Notes)
We recently had an RDS XXT Sensor installed on the initial CMM that has a probable issue as defined in the first post.
The RDS Sensor shows a similar issue, in the same axis, but not as much deviation as the XTR Sensor.

The Zeiss techs found no issues during there install of the RDS Sensor, that is, following there installation guide lines.
They did not look at the issue I presented and demonstrated using the RDS Sensor.

And, same as the XTR, Qualification of multiple probes (Articulations for the RDS) reveal no information that could lead to detecting this X axis deviation.

Curious if anyone with an XTR Sensor has used my program to run a report?

[[Wo...]]

Please sign in to view this quote.

Hi Rick,

Have you managed to resolve this issue? I have one brand new adapter plate that's throwing this error no matter how it's loaded, manually or picked up from rack.

Thanks

[[Je...]]

Please sign in to view this quote.

Hallo Wojciech Malig,

same here, have a brand new Micura with XTR and 5 new plates, at 2 of them the same issue (also no matter how it´s loaded). Update with DeviceActivator didn´t help last Friday, I have now an open ticket at Zeiss. Will answer here, if I know something new about it.

Greets
Jens

[[Ri...]]

Please sign in to view this quote.

No I have not, it comes and go's. I haven't even detected a pattern.