hexagon logo
Register Sign in
  • Replies 20 replies
  • Subscribers 222 subscribers
  • Views 334 views
  • Users 0 members are here
Options
Related

Calibration Error- Probe Reading Wrong

MatthewHaines
Straight to it. All measurements in metric. Versions: PCDMIS 2013 & 2013 MR1.

We calibrated our probes, using the usual method in probe utilities. But now they are all reading wrong. Up to 25 microns too small to be precise.

We calibrated a probe last week, then we ran a job today which had a 11.000 mm bore, it measured it as 10.980. Yet a 11.030 plug gauge slips in fine. We then ran it around a UKAS verified 8.000 setting gauge and it was reading 7.975. We recalibrated the probe and ran both again, and it still measured small on both the 11.000 and the 8.000. I edited the probe file manually (adjusted the diameter size) until it read precisely 8.000 on the setting gauge, then I ran it round the 11.000 bore again. This time it came out at 11.040 (which explains why the plug gauge went in fine). We checked the calibration spheres and they are measuring fine with a micrometer (which in itself is reading fine on UKAS verified slips). We then did this entire process on our other CMM and exactly the same thing has happened. Something is going wrong between the probe measuring the sphere and the software adjusting the probe data to bring it in line with the sphere size.

So why is PCDMIS not properly adjusting itself when it measures the calibration sphere? Why does it think everything is small?

We have deleted the probes files and rebuilt them to remove any possible accumulative error, but we are getting the same issue. We have verified everything and the probes are definitely reading wrong, this isn't a "maybe the job is wrong issue". The probes aren't damaged either, and we have tried swapping them over. This error is occurring on all 18 of our probes across 2 CMM's.
  • These are the results from this morning:

    T1A0B0 THEO X 0.000 Y 0.000 Z 167.500 D 1.000
    T1A0B0 MEAS X 0.000 Y 0.000 Z 167.500 D 0.921 StdDev 0.004

    Now, we can measure the styli ball (gently) with a mic and get 0.934. If I manually set the diameter at 0.934 all my results suddenly come back to spot on.

    So something is causing it to not read the diameter properly.
  • The 0.004 StdDev looks high to me we run a couple of tp20 on a few low spec jobs and normal is around 0.000/0.001. Have you done all the normal checks for the whole probe build, adapter joint tightness, clean the kinetic mag seats, styli tightness, ruby checked for tightness/damage.

    What you do not say is what length of styli you are using and ball size, how many hits, levels and speed of calibration as they all have to be considered if you require tight tolerance checks.
  • Are you sure you're using the exact same speed for the calibration as you are for the measurements?
  • Max inscribed? Least squares?

    Sent from my SM-G920V using Tapatalk
  • The probe files are stored locally and individual to each machine? I.e. not shared and not on a network.
  • What you do not say is what length of styli you are using and ball size, how many hits, levels and speed of calibration as they all have to be considered if you require tight tolerance checks.

    This is happening on ALL of our styli, so it doesn't matter what sizes they are. We use 9 different types of styli, all of which are showing the same error, so this isn't a styli problem.
  • The probe files are stored locally and individual to each machine? I.e. not shared and not on a network.

    Yes they are stored locally.
  • Are you sure you're using the exact same speed for the calibration as you are for the measurements?

    Yes, same speeds, same number of hits, same probes, same CMM, same planet, same universe.
  • A few points here...


    1) The tip diameter which you get after calibration should be smaller than the tip diameter as measured with a micrometer. This is because you need to move slightly past contact point to trigger a hit, and there will be some bending of the stylus also. This is why I refer to it as the effective diameter.

    2) It does sound like possibly the calibration isn't holding for some reason. After calibration (and viewing the results) make sure to say Okay on the probe utilities dialog and not Cancel. After that, F9 the probe again, click on tip A0B0 and then Edit (as you have done when modifying the tip diameters) and just double check that the Diameter matches the diameter reported in the Results file.

    3) What tip diameters and lengths are we talking about here (I can see it's a1mm tip in the results you've posted, in the first instance I'd stick to a shortish fairly large tip whilst we figure out what's going on here)

    4) Calibration settings - How many hits and what touch speed?


    Having a look at those results again they don't look right.

    Here are mine from a 1by50mm tip on a Std Force Module using 13 hits on 3 levels @ 2mm/sec (Note for a longer tip you usually get a smaller diameter and higher Std Dev) 

    T1A0B0         THEO X   0.0000 Y   0.0000 Z 191.5200 D 1.0000
T1A0B0         MEAS X   0.6036 Y   0.1514 Z 191.6765 D 0.9842 StdDev 0.0020
  • So you have two CMM in two locations that you say are now reading incorrect by about the same amount your setups are the same the calibration method/styli are good, so what has changes that could affect both independent machine readings.

    Is it possible that the errors you are seeing have always been there but were not noticed until now.

    Although you say that it affects all the styli so that's not the problem it could be if all the styli are on long extensions and yes I have seen that in a sub con shop who were complaining that CMM's could not measure better than 0.05 on diameters.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement