Strange calibration issue

[darius]

Hi
I like to ask about help with probe calibration.
Machine Global S with Leitz LSPX1H_T
Problem concerns D4 X 75 probe and PrbRdv field:
aaa.jpg
Generally my standard procedure is to calibrate all probe from the rack approx. 1 time per week.
I observe that probe D4 X 75 has huge different between calibration from 6 January and 9 January (PrbRdv field)
(All calibration after 9 January showing same problem on this probe.)
Off course I can understand, that something could happened with ball or stem, but I was check that and it looks like physically everything is ok with probe.
Surprising is also, that when I checking all probe on ring gauge with nominal dimension Ø30.997 D4X75 showing results around 0.01 to 0.013mm smaller on diameter. Every other probe showing deviation on diameter ±0.001
At the moment only solution to get proper results is manually changing probe data for each tip… like in below example result after calibration was 0.011, I have change that to 0.0055 to get proper measuring results from ring gauge.
bbb.jpg
Like I mention before, probe is checked for mechanical damage, ball is sitting fast on stem.
I was also reset all tips, calibrated again nothing help, then probe file was removed and created again, calibrated and still wrong.
I decide to measure ball on this probe like a sphere with new fresh Ø2 and Ø3 probe with 100 hits and result is perfect diameter and roundness:
ccc.JPG

After calibration assignment ASSIGN/V1=2*(PROBEDATA("DIAMETER")/2+PROBEDATA("PRBRDV")) showing diameter 4.022mm
Maybe you can advise me some other solution? How to verify what is wrong?

[William Johnson]

Could it be that your LSP module is defective?

[darius]

Honestly, i didn't think about that. I will try to verify this tomorrow.
I just wonder what could be wrong with this module?

[DungT]

Have you done everything correctly from the get go? Lower matrix calibration? Trax calibration? Controller type (LSXP1H settings are different depending on controller type)?

[BKulpa]

Did you give the probe file a completely different name when you recreated it? Corrupt probe files are saved in the controller.

[darius]

I have replaced LSPX module and calibrated again, but results are still the same. I have created also new name for this probe file and it not help.
I don't know nothing about lower matrix I have checked trax calibration.
I was checked calibration results and everything was all right until 6 January, then next calibration from 9 January (and all after, showing this results)
Only advice from Hexagon is to replace probe, they think that probe is tired and damaged.
It is interesting why result shows big different between FAST and MEAS on PrbRdv field? On other probe this different is insignificant.

[mb0258]

check dirt on the probe holder, retight the probe,
probe D4 X 75, is it one piece or probe + carbon extension ? change extension.

[darius]

Cleaning was first, retight was also done. This is carbon stem probe in one piece, no extension...

[mb0258]

I'm curious how you will solve it, please post also solution/root cause.

[DungT]

Each CMM Controller/LSP module combinations have their own characteristics which must be mapped by lower level matrix, this process alone would take 10+ minutes and should be done when setup a new probe, crashed, removed and inserted, or if large SDEV values occur when calibrating. We can only guess if you don’t take proper steps to setup your LSP module from the beginning

[pcdmisstudent]

Hey im stealing your thread, because i really dont wanna create a new calibration thread since there are so many already.
Many people say that the most important thing in the cal. result is the STD and maybe sometimes the diameter.
But what about theo - meas. I know theo is built from the probe buildup, and measure is what it acctually is.

So for my master tip for example, in A0B0 during calibration it gets theo = meas.
I swap to A90B0, or A90B180 for example (with the answer no - sphere has not moved) and the theo X = 0, measured about 1.5mm.
This is a really big difference no? It varies from tip to tip, but its usually about 0.3mm - 1 or even 2mm ( from theo X/Y-value)

Does this really matter tho or is it just the tips that are a lil bent etc (which the machine later compensates for)?

[Douglas]

T1A0B0 is your master tip. When you index the wrist to any other angle that tip is no longer your master tip. Only use that one wrist position as master... always every time A0B0 must be your master, so you are doing that part correct it sounds like.

A bent master tip can certainly cause that, 1.5mm is a fair bit especially if the tip is short. For a longer tip it still seems like a lot to me but it is more likely a longer tip is out that much... what is the build of your master? shorter is better but if you are measuring with it too that might not be possible, I use a 4x10mm tip as dedicated master , never touches a part only the sphere

[Douglas]

Also what Dung means is not that it is faulty, but that it is mounted not quite perfectly square to the machines axis. If you just had it installed I'd tend to believe that is the cause, Dung nailed it. I know for a fact my wrist got moved by the tech during the last calibration... the first thing I do when calibration techs leave is to double check all assembly/mounting of everything they touch and how tight everything is.

[pcdmisstudent]

Douglas Our master tip is a 2x20mm tip, and unfortunately its also used for measuring
edit: I now checked thru the old calibration results from before the new head installation and even a year ago (im fairly new at this place). And it seems like we've always had a big varience in theo - meas xy. But it've increased over time. So maybe its the tip thats the problem then?

[darius]

Question about low level matrix:
Should I just calibrate my problematic probe using low level matrix function? Is there any special procedure? When I read help file about that its looking little difficult. DungT can you provide how to do it safe step by step?

[michaelwildschutz]

I believe lower level matrix should only be done using a 5x20mm probe

[DungT]

Question about low level matrix:
Should I just calibrate my problematic probe using low level matrix function? Is there any special procedure? When I read help file about that its looking little difficult. DungT can you provide how to do it safe step by step?

The data will be saved in the controller after low level matrix operation is completed. This will affect everything you measure including existing probe data; therefore you have to recalibrate all probes.

You need to pair up Leitz LSPX1H/controller before low level metric calibration.
Hit F10, click on ‘Probe Options’ tab
(Note: Leitz module names have been changed few years ago . This’s for Leitz LSPX1H and I don’t know about the _T. )

FDC Controller.
Max force: .096 N
Low force: .019 N
Upper force: .072N
Trigger Force: .033N
Offset force: .06N
Return speed: .4mm/sec

B3c-LC or LSP Controller.
Max force: .072 N
Low force: .0096 N
Upper force: .048N
Trigger Force: .030N
Offset force: .036N
Return speed: .4mm/sec

Use a short rigid stylus that you have, it does not have to be 5mm and made that your Master tip. Open Probe Utilities dialog box, check ‘Trax’ then hit ‘Measure’ – select ‘low level matrix’ and follow the instructions. When the operation is completed, lower matrix data analog box will be appeared “If this’s acceptable, press ok….” There are 3 sets of number above that, the difference should not greater than .15 between the first number of the first row (X) and the second number of the second row (Y).

‘Use TRAX calibration’ must be turned on for all tips including master tip (A0B0).
Open Probe Utilities dialog, select master tip and say YES to “Has the qualification tool been moved, or has the Machine zero point changed?” what it’s really asking is has the distance between machine zero point (home position) and calibration sphere changed? IMO, that distance can change between morning and afternoon especially on large machine due to temperature and humidity change, even though nothing have been moved, so I always calibrate master tip and yes it has been moved regardless. Answer NO on other tips

[darius]

I received today new D4x75 probe from Renishaw. Replace with old one, calibrate and results are perfect.
Only explanation: something must be mechanic wrong with the old. I started to watch under the microscope (again) and this time I notice, that is very sensitive clearance between carbon and steel connection.
Before I was focused on ball, but I almost sure now, that it is the reason of my problem. Probe, before register touch, need to move few microns more to get proper force and same time diameter of probe will be wrong.Udklip.JPG