Yet another probe tip problem

[kneislyd]

Spent an hour or so looking in the forum to see if this problem was had by anyone else. It was not. Or I couldn't find it.

Here's the setup: PCDMIS 3.7 MR3. Scirrocco with PH10m/TP200B/ with 4x50 and 2x30 tips each on it's own magnetic module. Both are setup and qualified at A0B0. Tip changes are made manually.

The problem: During a measurement using the 4x50 tip, part alignment is made and Z is set to the top of the part. Later in the program I change to the 2x30 tip. When I hit on the top surface of the part it measures .904 when it should read 0.000.

I made a simple program using both tips measuring my qual sphere. The first tip measures the qual sphere and the alignment is made to the center of the results. The second tip is loaded and the same qual sphere is measured. The measurement results made by the second tip showed the sphere xyz to be

-.336, .336, .904 (should be 0,0,0 within a few microns)

The diameter results from the second tip is correct.

Any ideas?

[Matthew D. Hoedeman]

Spent an hour or so looking in the forum to see if this problem was had by anyone else. It was not. Or I couldn't find it.

Here's the setup: PCDMIS 3.7 MR3. Scirrocco with PH10m/TP200B/ with 4x50 and 2x30 tips each on it's own magnetic module. Both are setup and qualified at A0B0. Tip changes are made manually.

The problem: During a measurement using the 4x50 tip, part alignment is made and Z is set to the top of the part. Later in the program I change to the 2x30 tip. When I hit on the top surface of the part it measures .904 when it should read 0.000.

I made a simple program using both tips measuring my qual sphere. The first tip measures the qual sphere and the alignment is made to the center of the results. The second tip is loaded and the same qual sphere is measured. The measurement results made by the second tip showed the sphere xyz to be

-.336, .336, .904 (should be 0,0,0 within a few microns)

The diameter results from the second tip is correct.

Any ideas?

Well, steps needed are as follows:

1) Calibrate first tip, telling CMM tool is in new location
2) Calibrate second tip, telling CMM the tool has not moved.

You must calibrate BOTH tips with the tool in the same location and you can NOT move it after calibrating the first tip and you MUST tell the CMM the tool is in the same place. It also would NOT be ab bad idea to calibrate the first tip a second time before calibrating the second tip.

OK, next possible issue:

Do you, at any time, un-screw anything when chaning from tip to tip? If so, any previous calibration is no good. Anytime anything is un-screwed ro screwed in, all previous calibratinos are no good.

Another issue: Is there any possibility that you are too close to a magnetic source with just ONE of those tips, either during calibration or measurements?

[kneislyd]

Thanks, the problem was not following the procedure of qualifying the first tip, specifying that the qual sphere has moved, then qualifying the second tip specifying that the qual sphere has not moved.

Thanks,
Dave

[Matthew D. Hoedeman]

Thanks, the problem was not following the procedure of qualifying the first tip, specifying that the qual sphere has moved, then qualifying the second tip specifying that the qual sphere has not moved.

Thanks,
Dave

Everytime you re-home the machine, the machine absolute zero points change, they do not repeat, they are close, **** close, but not exact, so, everytime you re-home you must treat everything as if it had been moved on the table. If you ahve already calibrated tips on the tool and the tool has NOT been moved, you can tell it the tool is in the same place, but not otherwise.

Also, Pcdmis CAN now calibrate a single tip (say, 0,0) without any other tips. In days of old, you used to have to have a minimum of 4 tips, this would allow Pcdmis to construct a sphere of all the tips (more or less) and get the relationship between all of them. They changed that at some point (not sure when) and now you can calibrate a single tip. Not to say that THIS is what caused the problem, just some info.

HOWEVER, lets say you place your tool on the table and make a NEW probe build-up and lets make one that is a total mess. Put in a tip that has been fish-hooked. Now, ONLY calibrate the 0,0 tip. Do this in MANUAL mode (since it is a fish-hook) and take the 5 touches. Now, LOOK at the results of the calibration. The vector of the tip is perfect as well as the length. Since you only did 1 tip, Pcdmis has NO reference values to calcualte the correct actaul length of the tip, so it is 'perfect'. Now, make another build up, but make this one perfect, no fish-hook, no bends, correct length. Now, since you have already 'calibrated' a tip, Pcdmis SHOULD know where the tool is, so calibrate this tip in DCC mode and tell it the tool has NOT moved. It will not calibrate this tip. WHY you might ask? Well, it is because the first tip, the fish-hook' was assumed to be perfectly straight and the perfect length according to the components you used when building it probe. SO, the probe you use to locate the tool has to be pretty **** close to exact. Can't do a 2x12 instead of a 2x10 in your build up because you will be 2mm off for all the rest.

THIS was why I suggest a second calibration of the first tip and telling Pcdmis the tool didn't move.