How do you turn a part upside down and maintain yur past alignments? Is there a way to stitch things together or do you have to run 2 separate programs? Can't find anything in the hel file on this.
I know that it can work. Unfortunately, the only time I've ever used it, I was merely running what my programmer, at the time, had created. So, I'm not actually 100% sure how it worked.
ETA: Then check out the link within that help file titled: "To Change a Part's Position and Orientation"
How do you turn a part upside down and maintain yur past alignments? Is there a way to stitch things together or do you have to run 2 separate programs? Can't find anything in the hel file on this.
I would suggest that anytime you "move" a part on your machine, you will need to create a new alignment of some sort. I do have programs where I need the operator to flip a part because there is a flatness callout on each side around a bore. In this case, since the part is flat on both sides, all I need to really do is align to the center of the bore with 3 hits and measure a plane on top and I have my flatness.
Remember that gravity has an effect when flipping parts over.
It is possible that measuring the exact same features from the exact same datums will have different results when you flip a part over.
It all depends on the geometry and the weight.
We use this to our advantage at times.
How do you turn a part upside down and maintain yur past alignments? Is there a way to stitch things together or do you have to run 2 separate programs? Can't find anything in the hel file on this.
Search for 'leap frogging' that may give more useful help.
Even if you flip the part you still would need access to the features from previous alignment, so might as well switch to Manual mode, create a New Manual align and a New dCC align.
If the component is moved you will need a new alignment, as simple as that. No human or machine can flip a part and put it in EXACTLY the same place, so your error margin will be whatever the difference is between where the alignment was on the previous die (where the CMM thinks it should be going) and where the part actually is. Depends what your tolerances are as to whether you run the risk.
If, however, you do not need to link both sides of the component together, you can flip it, run a new alignment and carry on. As long as the results from the second side aren't calling from an alignment from the other side and are independent.
Personally I make one programme per side. So the operator turns it over, runs the next programme and it gets treated like a different component. It also means if the setters have issues with one side of the component they can just run the relevant programme rather than wait for the whole thing to finish. Whenever the component has to be touched, or if someone looks at it funny, I do a new alignment, no chances.
Even if you flip the part you still would need access to the features from previous alignment, so might as well switch to Manual mode, create a New Manual align and a New dCC align.
1. in machine alignment, set up part.
2. manual + dcc align accordingly
3. inspect desired features.
4. flip part
5. reactivate machine alignment
6. manual + dcc align THE SAME WAY AS IN STEP 2. must be same features.
7. equate dcc alignments from step 6 and step 2
8. monkey dance.
There are two things I normally have done that fix this situation for me:
1: In the past, I have set up my parts the same way every time, on a peg plate, with fixturing, always putting the pegs in the same holes. This allows me to perform a DCC alignment in the beginning, and, if I have to flip a part, I always set the machine to do another DCC alignment, as a part of the program. This way, the operator doesn't have to actually DO anything, except flip the part and make sure it looks like the picture. I also try to set my fixturing up so that if a measured surface has to relate to another surface, then it will be accessible from the first fixture location, or subsequent fixture locations. I haven't ran across a part yet that I can't order my program in a such way as to allow me to get every dimension and correlation, by simply turning the part, and realigning. Even if I have to use a V-Block or vise, if I fixture my v-block or vise the same way, and place stops in a way that they always are back in the same spot, it still works.
or
2: since a large number of the parts I am doing for my current employer are circular parts, this company uses a machine lathe chuck, mounted to the surface plate, with a recalled global alignment done to the tops of an interchangeable pin set. That way, the center of the alignment is ALWAYS at the same exact location, no matter if the part is flipped or not.
