True position on a hole with datums that have 2 virtual conditions

[Jan d.]

Another one for you brainiacs...

Please review the attachment. Sorry for the funky colors. On the right there is a flat plane defined, called datum -H-. There is a 0.1873 hole called datum -J- and one called datum -K-.

As you can see, both have 2 VC's: one of location and one of orientation.

My question: when I evaluate the 2 holes on the left, what do I use for the VC on the datums when looking at the TP of those 2 holes?

My thought:
Datum -J-:
VC location: 0.1872 - 0.001 = 0.1862
VC orientation: 0.1872-0.0002 = 0.1870

I believe that in that case, 0.1870 must be used for the VC on the datum.

Same goes for Datum -K-. VC = 0.1870

Looking at PC-DMIS 4.1, it uses 0.1860 as the VC on both datums. It takes both VC's and it adds them. I do not believe that this is correct. It seems to me that in my case the VC for orientation ought to be taken.

The net result: PC-DMIS gives so much "wiggle" room on the virtual gauge that my TP's are almost always zero. I can be off by as much as 0.003" when evaluating with both datums at RFS, still with the VC, it wiggles so it brings it almost back to zero. Using only the orientation VC, TP becomes much more reasonable.

So the question is a simple one: is it correct to add both VC's when dealing with datums that have dual VC's?

Come on Mark F, you ought to have an opinion on this one!



Jan.

[kbotta]

I believe there is an "unwritten rule" that you use the more restrictive one for your datum simulator Jan (I could be wrong - it's been a while with this one). I'll look tommorrow in the standard to see if I can find something - but I am about 90% certain I will not.

I do not believe it to be correct to add them both together however...

KB

[Got-a-life]

What is datum -E- and -F- ?

[Jan d.]

What is datum -E- and -F- ?

-E- -F- and -G- are datums located somewhere else on the part. By the way: they also have dual virtual conditions....

This part is one big chain of datums.


Jan.

[Rich P]

need more info than shown on drawing
ie datums not all are defined
how about some basics maybe
is this 1 view or 2 ? part is not clear to me

[craiger_ny]

I have not looked at the standard but here is my opinion. I would say (opinion) that a feature is constrained to the datums in the FCF only. If the datum location is constrained to other datums so be it. However, when measuring a feature at basic dimensions from certain datums that feature is constrained to those datums at the specified basic dimensions only, regardless of the location deviation of the datums to other datums. In my opinion the VC of the datums is their (in this case) smallest size (MMC) and allowable orientation. In this case 0.1870. It is a little perplexing picturing a gage though depending on how much your datums can deviate. It sounds like a situation where stack-ups could lead to enormous deviation of the last features in the cycle. Like Kevin said as a rule of thumb I've always gone with the more restrictive when I interpret questionable stuff. Interesting problem.

Craig

[Jan d.]

need more info than shown on drawing
ie datums not all are defined
how about some basics maybe
is this 1 view or 2 ? part is not clear to me

Sorry, but my customer does not allow to show more. I had to run this by him first.

Basically, you have one flat surface, your primary and 2 cylinders as your secondary and tertiary. Both these cylinders have dual VC's, as shown on the sketch.

Then I measure the hole and evaluate it against these 2 datums with their VC. That leads to the question how much VC is used on the datums. I question PC-DMIS adding both. I think it ought to be the most restrictive one (like Craiger and Kevin argue).


Jan.