Auto calibrate 1 probe?

[pernilla]

Hi.

I've got some concerns about calibrating probes.
I calibrate all the tips and angles once every week, using a master probe and an auto-calibrate program. It takes about 1.5 hour.
But every now and then an operator fails to read instructions and break a probe. So I have to run the whole calibration program after replacing it (or an operator does it if it happens at night).

But is this necessary? What if I make a few auto-calibrate programs like, one for each probe? They would start by locating the sphere with the master probe and then calibrate the one probe that has been replaced... (I want to make it simple for the operators). Would the probes still relate to one another?

We could save a lot of time by doing this but I want to make sure I don't compromise the accuracy.
And I know there are a lot of topics about calibrating probes but I couldn't find anything that answered this question.

[vpt.se]

I guess it depends on if the calibration tool (sphere) has been moved or not. If it has been moved, I believe you will have to recalibrate all tips? If it hasn't been moved, just replace the tip and calibrate that particular probefile.

[pernilla]

I have to move it after every calibration.

[AndersI]

If the sphere is located with the (one and only) master probe, and the master probe is always reset to theo (or never calibrated with NO), then it should be enough to do as you suggest. As long as the master probe doesn't get bent or damaged in any other way…

[pernilla]

I always locate the sphere with the master probe, and it's always reset to theo. And we don't use the master probe for anything else.

So if I try this, I should be able to perform some test to reassure that the relation between the probes has not been compromised, right?

[William Johnson]

I do what Andersi does but after I finish the calibration I run a program that will verify that all probes are still relate to each other. Just use the primary probe (A0B0) to measure a sphere and set XYZ on it. I do this 2x
then I rotate the probe to a few of the primary (Most Used) angles. Tolerance the XYZ and Dia. of each probe rotation for all probes in the rack. I set the tolerance to +/-.005mm
If something is wrong it will show up. I have 2 racks and 12 probe builds and it only takes about 15 minutes to run. Much shorter than a full calibration.

Also you do not need to create a program. If you have a auto calibration program already Just call up the primary probe and calibrate A0B0 with NO did not move. Then call up the probe that needs calibrated and do a calibration on it. Then run the test.

[pernilla]

Thank you so much. I'm going to try this.