Alignment Update Question

[PCDMISRON]

Good morning All..I am attempting to move the alignment of my part 0.0304" in the Z+ direction. When I do so, should not each feature programmed in this alignment move relative to the new alignment, i.e., the features should move off of the CAD model.

Thanks,
Ron

[Matthew D. Hoedeman]

Good morning All..I am attempting to move the alignment of my part 0.0304" in the Z+ direction. When I do so, should not each feature programmed in this alignment move relative to the new alignment, i.e., the features should move off of the CAD model.

Thanks,
Ron

When you change ANYTHING in a 3-2-1 alignemnt, it should ASK YOU if you want to UPDATE the features following the alignment to the new alignment. ALMOST all the time, you will want to say NO or it WILL move those points OFF the cad model. If you are not being prompted, then you need to change one of the F5 warnings to default to "NO" and to POP-UP the window asking you.

[PCDMISRON]

Okay....I got it. the CAD equals Part must be applied prior to the transformations.

Ron

[Matthew D. Hoedeman]

Okay....I got it. the CAD equals Part must be applied prior to the transformations.

Ron

Well, yes and yes......

If you do a 3-2-1 alignment, you MUST do CAD=PART if you are using cad data to program (I do it 100% of the time, with or without cad)

Now, this is the most common time I would change the 3-2-1 alignment:

I get a holding fixture in. I align by the numbers ON THE BASE
I do a DCC alignment, using the same touches as the manual alignment I did.
I will then (ALWAYS!) check the fixture locators to make sure that they check within fixture tolerance (generally less than 0.05mm / 0.002"). If so, then my alignment is done.

IF THEY ARE NOT in tolerance, I will adjust the offset values in the alignment (right in the EDIT window) ONLY for the alignment made from the DCC features. The locator check features are already in the program and are AFTER the alignment. They are TO CAD DATA and they need to stay that way, so when (IF) I tweak the offset values, I TELL IT NO DO NOT UPDATE.

I will then run the DCC alignment and check points to see if they are all in tolerance now. I usually get it with one adjustment, but if the locators are on wierd angles, it may take more than once. Each time I adjust the offsets, I TELL IT NO.

Once I finally get all the locators in tolerance and I have made my last tweak YOU THEN MUST DO CAD=PART AGAIN or any new point you pick off the cad will reflect the amount you shifted (or changed) the offsets in the alignment. AND, in V3.7 MR3 I have noticed that you MUST DO THIS TWICE to get it to 'be right'.

[PCDMISRON]

Hey Matthew, thanks for the input on the alignment problem. I will also make CAD=Part a part of my programming routine 100% as well. I may not always need to transform my model and/or translate alignments, but I do not se where it would be harmful if implemented correctly.

Ron

[Matthew D. Hoedeman]

Hey Matthew, thanks for the input on the alignment problem. I will also make CAD=Part a part of my programming routine 100% as well. I may not always need to transform my model and/or translate alignments, but I do not se where it would be harmful if implemented correctly.

Ron

Be warned about this!
1) CAD=PART can not be un-done (but then, I have never wanted to un-do it) (but, you CAN re-do it to a 'corrected alignment)

and

2) Some here will tell you that CAD=PART is the biggest mistake you can do.


However, I have been using Pcdmis since, well, FOREVER it seems and I have never had an issue with it. Maybe back in the day, the instructors actually KNEW how to teach CAD=PART or maybe we listened better back then, but never, not once, have I had an issue with it. (now I'll get slammed!)

As a side note:

Many people have had issues with nominals changing in the features in their programs, slight, subtle shifts and outright changes. This is also something I have never encountered. (I'll get slammed again for this one!)

I will let the student infer any connection......