Whats up gang. I have a question, well perhaps I know the answer to this one, but I just wanted to throw it out there and see if anyone had a better idea,
Example, we have a part that is manufactured by an outside vendor. The Vendor operates Calypso software, If I wanted to verify that the program is correct, the sole purpose is to see if the vendors programmer is a good programmer and follows the print etc..
What would be the best way to verify that? If it was pc-dmis, of course I could have just ask them to export me a prg file, but what If I dont have that software? is Visiting the Vendor is the Only option? Purchasing the Software perhaps, but can we really purchase 2-3 software's??
Do you do any work with your local Zeiss distributor? Does your vendor? If you think they could be impartial, pay them a consulting fee to review the program.
Have customer run serialized sample of parts
Send results and serialized parts
Run on your program, if the 2 sets of results correlate
Good to go
Discrepancies = conversation-change and repeat test
I thought about this kind of approach, the issue with that is that you can have the identical or pretty close results, yet it would not mean that the program is correct
While you do not have the full code visible you can set it to display expanded information and that should give enough to verify that the correct features are being used in the alignments and measurements. That is unless they have changed this since I last used it some 8 years ago.
I would just have them run a few parts and label the parts, then when you receive the labeled parts and reports run on your program and verify. If your results are similar then you should be all set. This will not ensure the program is correct but it will show that both of you are doing it similar.
Do you not have confidence in your programs? If the numbers are similar and you know how you programmed yours than you should be confident in the results.
results can be very similar, yet the program still not to print.
Quick example, you can have a TP callout TPØ.005ABC, you may get the same results for achieving the TP, but the datums could be used different from the feature frame. There for just because the results matched this time, it doesn't mean they will match the next. or am I over thinking this one?