Its me again, with another newbie question. Im still a novice to PC DMIS, having only taken the Level 1 CAD class to this point. Lets say I have a part lined up with the CAD model. However there is significant warp to the part so that when I go to measure another feature elsewhere from my original alignment, its way off the CAD model giving me misses and crashes. Is there a way to line features back up? That is to say, try and tell PC DMIS that the CAD feature is actually over here on this part?
If your parts are consistant you can give a larger prehit and retract so your probe has a larger search area. You can use find hole on circles. You can change your target dimensions on your features, without changing your nominal values.
Some features have a realative (RMEAS) option that you can have this feature adjust it's target based on a similar feature. The newer versions allow the use of multiple features. Another option is to create a larger prehit and/or retract.
Something else I thought of. You can try measuring a feature close to your warped area, create a secondary alignment, translate and datum off that feature, and then measure the other features. Just remember to report all the features back to the primary datums I think this will give you the correct nominals and your measured values should show how bad/good the part is. If your parts aere so far off and warped that the cmm can't even find the feature, you might as well use a tape measure instead of a CMM.
LOL, I wish I could use a tape measure. Thanks for all the advice, it was helpful. I deal with a lot of parts that get bolted to something or vibra or sonic welded to something. Unbelievably major warp isnt a concern in parts like this, so I deal with a lot of warped parts.
Something else I thought of. You can try measuring a feature close to your warped area, create a secondary alignment, translate and datum off that feature, and then measure the other features. Just remember to report all the features back to the primary datums I think this will give you the correct nominals and your measured values should show how bad/good the part is. If your parts aere so far off and warped that the cmm can't even find the feature, you might as well use a tape measure instead of a CMM.
+1
We had a need to profile a dovetail along length of a 'blade' (medical retractor, not aerospace) and it was very warped. (at risk of cave monkey snickering, imagine a banana with stem facing Y- and length of blade running in Y axis out to 'tip' of Y+, curve rising in Z+) It was profiled at certain gage points and only thing that would allow program to run through was resetting Z origin at those gage points. If memory serves, part was warped .09+ over about 6". Pretty bad looking report, but it showed to engineers the failure in the process. As grandpaal mentioned, report all items back to the FCF callout. (that's when the 'bend-monkeys' laffed so hard, they fell from their trees) Oh yeah, the prehit/retract had to be tight as well.
