.003 error after removing a couple measurements

[RMILLER]

We have some programs that are full first article inspections that we needed to reduce for production. After removing some of the measurements, the program no longer measures correctly. We didn't change anything related to alignments, datums etc.. The nominal values are correct, the reporting commands have correct nominals. Nothing visible in the code explaines the .003 in. difference with the reduced program. This has happened on several programs now. I know PCDMIS has issues with deleting too many features at a time so we just cut out a couple at a time. Any ideas?

PCDMIS 3.7 MR3 B&S ONE

[Matthew D. Hoedeman]

We have some programs that are full first article inspections that we needed to reduce for production. After removing some of the measurements, the program no longer measures correctly. We didn't change anything related to alignments, datums etc.. The nominal values are correct, the reporting commands have correct nominals. Nothing visible in the code explaines the .003 in. difference with the reduced program. This has happened on several programs now. I know PCDMIS has issues with deleting too many features at a time so we just cut out a couple at a time. Any ideas?

PCDMIS 3.7 MR3 B&S ONE

Well, you weren't using a relative measure for any of the points, were you? What happens if you jsut makr those features you want in the FULL check program and run just those instead of removing the points you don't want checked? In fact, this is a GREAT way to help you find any possible problem. If you go through and MARK just the dimensions YOU want reported, Pcdmis will automatically mark any and all features required to get those dimensions (unless you have changed a setting, I don't rememeber what or which or where).

[RMILLER]

Thanks for the reply Matt.
The program didn't have any RMEAS type measurements. I wish we could just mark what we wanted to measure but the program moves around and rotates probes too much to get away with that. We have to carefully cut out 2 of 3 cross sections of measurements and create a new probe path in order to reduce the measurements.
I'm more concerned about the root cause of the .003 error than a workaround. It bothers me that the code doesn't show any reason for it. If I can't see it, I can't fix it.

[rangerboat72]

Good catch Matt. I bet that is exactly what it is. Relative measure has gotten me quite a few times on that one especially when I go back to modify an existing program.

[rangerboat72]

Ooops. I thought that was what it was...

[Matthew D. Hoedeman]

Thanks for the reply Matt.
The program didn't have any RMEAS type measurements. I wish we could just mark what we wanted to measure but the program moves around and rotates probes too much to get away with that. We have to carefully cut out 2 of 3 cross sections of measurements and create a new probe path in order to reduce the measurements.
I'm more concerned about the root cause of the .003 error than a workaround. It bothers me that the code doesn't show any reason for it. If I can't see it, I can't fix it.

Well, the other option is to make a copy of the full check program, mark those dimensions/features you need, then insert moves/tips that will make it run and try it to see if the error still shows up.

Another possibility, are you still taking ALL the hits with the same tip you did in the full check? That might also be a reason for the difference, you might be shanking out or something OR, depending on the setup you have, you might be getting magnetic interference that you weren't before.

[bob mappes]

is it possible that you deleted an alignment and did not do the "update" thing that pops up when you add or change an alignment?

[DaddyBluto]

Good lord this is exactly like my "It Don't Add Up" post. My delta was .0033". I still have not had the time to dig into it any further. If you figure it out let me know.

[Matthew D. Hoedeman]

Good lord this is exactly like my "It Don't Add Up" post. My delta was .0033". I still have not had the time to dig into it any further. If you figure it out let me know.

For your approval, two men, widely seperated in both time and space, encoutner the same phenomena, in the Twilight Zone.

[DaddyBluto]

If you go through and MARK just the dimensions YOU want reported, Pcdmis will automatically mark any and all features required to get those dimensions (unless you have changed a setting, I don't rememeber what or which or where).

This does not completely work. It will mark the feature(s) you are dimensioning and the feature(s) used to make it, but it will not mark the features used to create the alignment(s) you need for the measurements. It seems to me that they should be linked.

[Matthew D. Hoedeman]

Yeah, I guess you are right, big-daddy, but it WILL keep you from eliminating any features used for the dimensions. And, I think, that if you 'mark' the alignment, it will mark the alignment features used in that alignment, won't it?

[DaddyBluto]

Not for me. All alignments stay marked all the time for me.

[DaddyBluto]

For your approval, two men, widely seperated in both time and space, encoutner the same phenomena, in the Twilight Zone.

Yes, but I am going to Ohio in July...

do do do do do do do do do do do do

[Matthew D. Hoedeman]

Yeah, they 'appear' marked, but try this:
Un-mark the entire program
Place the cursor ON the alignemnt and hit F3
At this point (V3.7 MR3) all the features used in that alignment will mark themselves.

[DaddyBluto]

Thanks Matt. I have must to learn from you master. Some day I will be able to ****** the pebble from your hand.