rikracer I hear ya! I just dont trust demon to do math for me...AND...the majority of my customers submit mathematically incorrect datum structures so if I tried to use GEO TOL, it would spit out gibberish numbers at me. I usually have to email customer and tell them "hey guys I know you tried to make a nice blueprint with your crayons over there but heres how we propose it should be measured so it fits your design intent that we think you dreaming about" and they're happy to follow along
it also depends on the industry you are in. I manually align a fixture (plane-cir-cir or plane-cir-lin, etc) then do an iterative alignment on all the datums/locators, and done. Saved to an external alignment, which is recalled in the check program. AND, even though the 'designers' use (M) on datums and position, NONE of them allow datum shift (so what's the point, right?).
Point of order, if you use an ITERATIVE alignment in a program that then checks the part, that alignment gets recalculated every time you open the program, EVEN IF YOU DO NOT RUN that portion of the program, it gets recalculated every time.
I only have it slightly better than you, I can, and do, force my design "engineers" to comply, even though they didn't have to before I got here. I don't care if it is a 20 year old print, They either make the corrections to the GDAT or they get no real info. It is amazing to me how companies will go on and on for decades doing things incorrectly, and then tell me "that's the way we have always done it" or "they use to pass all the time" I am too old and cranky to put up with that S, so, being the only CMM programmer, leaves me in a decent position to get things altered and corrected
Notice to all reading this thread.... he CHECKS the features in the alignment they are called out to. This is more important than people think it is. You really REALLY don't want to check a profile using ABC if it is called out in DEF as the results can be garbage.
The biggest advantage is when you need to recalculate an alignment far down in the program, say A10, for example. If A10 recalls the Startup alignment, it will be easy for it to recalculate. If A10 recalls A9, which recalls A8, etc. etc. then it takes the software a long time to recalculate every prior alignment in the program, and also makes it more likely to calculate something wrong. Recalling the startup alignment in every alignment is a very good habit to get into regardless of whether or not you use Legacy dimensioning.
When I got hired in here (the 1st time) the engineering manager asked me what I would do if someone questioned the results of the CMM. My response was that I'm not perfect though I strive to be and if someone questioned the results I would find a way to hard check the part. He later told that several other people that were interviewed did not have the knowledge to measure what he used as an example and were not hired because of that. I learned part measurement working on a surface plate.
That also applies to re-running features in the middle of the daisy-chain. We have a few programs that are like that, and I just run the entire program from the beginning.
