Reference Sphere Position two rotation angles one sphere

[[Ni...]]

Hi everyone,

I need help understanding something i just saw which makes no sense to me but its possible that it is ok to do.

So they have one single reference sphere that has an inclination of 135°.
Because of that inclination, some styluses don't finish qualifying.
So here is what they do.

They set the Reference sphere Inclination at 135° and Rotation at 45° and click on "Ref Sphere Position" with the Master Probe.

Then they choose the next Stylus System and click on "CNC Probing System Qualification". Then the Stylus System will automatically start qualifying. They skip few probes that cannot be qualified at that angle and let it complete.
They continue to do the same with the rest of the Stylus Systems.

that's normal right? Here is where I was baffled,

Once they are done, they unscrew the Reference Sphere and rotate it to 225° and screw it again. Call the Master Probe again and do the "Ref Sphere Position" again but now at the Rotation of 225° instead of 45°.
Then they call back the probes that didn't finish qualifying and do it manually i believe.
I didn't stay to finish because i wanted to come and ask if that is normal to do??

I was under the impression the stylus is qualified to the position of the reference sphere. So if you move the reference sphere and re do the sphere position again with the same Reference Sphere, then everything that was qualified first becomes invalid.
Or am i wrong and the way they do it is perfectly fine?


Thank You,
Nixon Uruchima

[[Mi...]]

We do exactly what you described. Yes, we try not to but it's unavoidable in a lot cases. The most important thing is that you do both sphere positions at the same time. Some of our techs try to only do 1 because they figure that's the only one that needs it, but then they cause a worse issue with a characteristic that relies on probes from both positions.

Many of our calibration programs look like this, Standard Stars all done on a standard rotation, but our fancy ones have to be done on another.

We have some where not all the probes on a single system can be calibrated at once, in that case we have 2 calibration characteristics for each but they would be named something like "Star Port 5 - Except Back" in one rotation group, and then "Star Port 5 - Back" in the other. Each calibration characteristic would only have the probes selected that can be done at that angle.

[[Ni...]]

Please sign in to view this quote.

I get that and understand that completely. But what has me concerned is that they are redoing the Reference Shere position after rotating on the same Sphere number. It is overwriting itself after rotating.

Wouldn't it make sense to make a copy of the reference sphere. Keep one at 45° and the other at 225°?
Reference the sphere location of sphere number 1, Rotation at 45°, and do all the stylus for that rotation.
Then Reference the sphere location of sphere number 2(Which is just a copy of sphere number 1), Rotation at 225°, and do all the stylus for that rotation?

That way the R, S, X, Y, Z, Tilt, Rotate values of the first time don't get overwritten by the second time?
That is my concern. Those values being overwritten and then giving bad results if you have a characteristic that uses probe from both qualifications

[[Mi...]]

Nothing is invalidated, the only danger we've run into is people deciding not to calibrate the whole CMM.

Having two reference spheres virtually while only having one physically is a headache and very likely to cause problems, somebody will invariably reference them in the wrong positions, and select the wrong one when going to calibrate. That and the software doesn't even want you to do that, you'll have to add a character to the serial number, it won't let you have 2 with the same serial.

If this was an issue, we wouldn't be able to run half of our parts.

[[Ni...]]

Please sign in to view this quote.

I am just trying to get my head around how Calypso handles the stored values. So then even if we reference sphere position and qualify half the rack with Sphere number 1, then rotate it to a different position and reference the position again, it won't mess with the first half of probes that got qualified?

[[Mi...]]

Please sign in to view this quote.

Not unless you try to qualify them again, and they fail because they can't get the sphere coverage.

[[Da...]]

The position of the reference sphere is not important for the use of the machine. The only thing that matters, is that the position is known when calibrating your probes.

Let’s say you would calibrate a probe system number 1. You set up the reference sphere and calibrate. After the calibration process, the effective diameters and bending parameters of the probe system are known. You can work with the probe system.

If you do the same an hour later for another probe system number 2 and you set up the reference sphere in another place, then after calibration, you can use probe system number 2. The problem is, you can’t use them in the same program, because while they measure diameters okay, the positions of measured elements will differ between probes.

To overcome this problem, you need the master probe. When you set up the reference sphere you don’t just measure its position, you calibrate the reference sphere to the master probe. The master probe is the connecting element between calibrations. By using a master probe, you can easily move the reference sphere in different positions and different directions between calibrations and still the probe systems you calibrate will always refer to the master probe. This is why you should never use the master probe for normal measuring. Its only purpose is the calibration process. For that it needs to be rigid, undamaged, and clean.

An easy way to prove this is to compare the reported positions of the reference sphere using multiple probe systems that were calibrated on different days. If you set up the ref sphere shortly before calibration each time, you will see that the differences in position among the different probe systems are extremely minimal.

So, it is absolutely okay to move the reference sphere between calibrations, if you set up the position of the reference sphere by use of the master probe shortly before the calibration. Do this even if the reference sphere did not move but a certain amount of time has passed, otherwise you will not get repeatable results.

[[Cl...]]

This was my solution.

[[Ni...]]

"The problem is, you can’t use them in the same program, because while they measure diameters okay, the positions of measured elements will differ between probes."

I think i am just going to make a copy of the Reference Sphere 1.
That way Ref Sphere 1 is at 45° and Ref Sphere 2 is at 225°

That way when i qualify and rotate the sphere, all of the probes that are qualified on Ref Sphere 1 don't get disturbed when all the probes are qualified on Ref Sphere 2 and i can still use probes that were qualified from both spheres.