2020 R2 True Position Datum Callouts

Good afternoon,

For a momentary fix I have resorted to using legacy mode to get this to work for now but I really don't want to use legacy mode for 1000+ feature reporting, I'm sure you can sympathize lol

I have been tasked with writing a program that has 3 shifts (part is 136" long and machine travel is 59.750"), and when programming the middle section I have to use a theoretical for Datum E and H (because they're unreachable). I did the typical thing and created a generic feature, with comments that the operator can input actuals for X and Y locations that turn into assignments within the dimensional data for the datums, based on results from the 1st program where they could be reached.

The issue I have is when trying to dimension a feature (circle feature, true position callout to D/H/E) it will not allow me to select the datums I assigned (Datum D is a reachable Plane, Datum H and E are both theoretical's from section 1 program data) and I have to adjust the alignment and use Legacy Mode to get results.

Is there a rule against Plane/Point/Point datum schemes in R2? I used to do this same thing in R1 all the time without fail. Does the math in R2 require something more than a point for a secondary datum? Do I need to create a 2nd point offset in X from the data of datum H to make a line, and then will be allowed to use the line as the secondary datum callout?

Attaching a copy of the print

Attached Files

KIRBSTER269 over 4 years ago

How about a line between the 2 points for rotation? A Plane, point, point? kinda weak, reminds of........
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
AverageInspector over 4 years ago in reply to KIRBSTER269

I ended up creating a line from my shift circle, to the theoretical point for H and offsetting the .0425" it is out of center, and it let me use that line as the secondary datum (basically what I assumed I needed to do in my initial message, but wanted a pro to nod or say nay to).

Thanks, Pro
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
JacobCheverie over 4 years ago

Are you seeing an error message when you select DRF [D|H|E] in the geometric tolerance dialog window or the edit window?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
AverageInspector over 3 years ago in reply to JacobCheverie

Yes, but I figured it out thank you!
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
JacobCheverie over 3 years ago in reply to JacobCheverie

AverageInspector out of curiosity, do you remember the error message that you received?
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
AverageInspector over 3 years ago in reply to JacobCheverie

"H (SIM_H): FEATURE NOT ALLOWED FOR SECONDARY DATUMS."

I think it was because all degrees of freedom weren't locked in with those theoretical's set as datums, I had to create a line from my shift circle to the sim datum point for the secondary datum, and then my other sim (Datum E) was allowed because it just locked translation in Y axis and everything else was already locked in.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
JacobCheverie over 3 years ago in reply to JacobCheverie

That's a good solution. I was just curious. A plane/point/point wouldn't have been a good idea anyway, as your secondary datum should constrain rotation per the print, but a plane/point/point would constrain rotation using the line from secondary to tertiary.

I was able to make the plane/point/point work in PC-DMIS though. You have to construct a cast sphere using the point and arbitrarily set the diameter. You can then use the sphere as the datum feature. Your workaround is more accurate.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
AverageInspector over 3 years ago in reply to JacobCheverie

Never thought of it that way, but thats a great idea.

I think my biggest issue is that this part is so long I have to do 3 programs, which entails 3 shifts back to datum E because its unreachable. Lately I have been sort of cheating the datum edge, and constructing an offset point in X axis (10 inches to the right of the theoretical point) and then one like .050 above that created offset point, then creating a 3 point plane. I know its really lying to the machine but R2 really likes to use Plane for datums I have found. 100% of the time the FCF doesn't error and the nominals on the results are always money.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
JacobCheverie over 3 years ago

If the two constructed points have the same Y value as the first, then constructing the plane shouldn't make any difference from the point (short of PC-DMIS not liking the point datum). A planar datum feature should be measured as a plane anyway because the high point of the plane should constrain translation. The arrow in the image below points to the point that should constrain translation in X. If you only probe the center of the plane as a single point, you could be quite off.

Have you checked out the new (2020 R2+) PC-DMIS help files? There is a large amount of information regarding the new Geometric Tolerance command, including error message diagnostics and information on datum features with and without surface data.

https://docs.hexagonmi.com/pcdmis/20...ves_Datums.htm

Check out the section "Datum Plane Samples". It appears that PC-DMIS treats a point as a planar sample and can only be used as a tertiary datum.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel
Matthew D. Hoedeman over 3 years ago in reply to JacobCheverie

equate alignment used to be SO MUCH better than it is now. It USED TO BE that you could measure any thing you wanted, anywhere you wanted, level, rotate, set origins, origin offsets, then equate that to a previous alignment that didn't measure ANY of those 'exact' features. for example, top of a fixture base, the edge, and a hole at the front (set up #1). Then shift it, measure a plane (all different hits, in new locations), measure the edge (again, different hits in new locations) then a hole at the back, and you could equate those 2 alignments. When they changed it to be more like LEAP FROG, that was a game-ender for me.
Cancel
Vote Up 0 Vote Down

Sign in to reply

Cancel