True Position Issue

[[Ja...]]

I have a question concerning the default Position evaluation in Calypso. I have a bore (Z+) called out to three datums, namely a plane (Z+), a bore (Z+), and a cylinder (X+). Everything goes great with the Position window until I add the third datum. After the third datum is added, my Position value jumps from roughly .002" to .080". I have gone over the evaluation of the third datum and decreased scanning speed. The values in the feature windows themselves do not come close to the values output by the Position function. When I manually calculate the Position it comes out fine. Is there something going on behind the scenes of the Position function?

Thank you.

[[Tr...]]

Have you checked and make sure the nominals are correct for both the timing datum as well as the feature? Something else you could try as just a cross check is to substitute the timing bore in the X+ with the face on the same side as that bore. This would just help determine where the problem is.

When you manually calculate the true position where are you getting the deviation from? From hard checks or from the cmm? If its from the CMM feature menu you have to be cautious that you are using the same datum reference frame as the print, it defaults to your base alignment which Id assume is different than your position you are trying to achieve.

[[Ja...]]

Hi Travis,

The nominal values are all as they should be. I will try substituting the timing datum with another feature.

I am calculating the position by using "Result Element" in Calypso - I guess this would be using the feature menu?

When I set the base alignment to the datum reference frame that I am working in, I get this large error. I also get the error when the base alignment is anything else and I set up the datum reference frame correctly. When I set the base alignment to a different setup and I set my datum reference frame to Base Alignment, the error no longer exists; but this can't be the correct evaluation.

I also notice that when I change from a diametral tolerance zone to a rectangular zone, the error goes away.

I must be ignoring something simple, I'm just having a hard time seeing what may be wrong.

Thank you.

[[Tr...]]

More than likely you are using whatever your base alignment is in the result element. Rectangular Zone as opposed to Diametrical Zone shouldn't change results just the Zone shape. I would also make sure you have the correct coordinates being evaluated in the true position callout (XY, XZ, YZ). Just know substituting only helps troubleshoot, it should be per the print.

[[Ja...]]

Something seems to be off. When I change the zone shape the results do change wildly. The zone is using the correct coordinates for this application.

The only thing that remains consistent and gives me an accurate result is when I use the Result Element. There seems to be some calculations in the background of the True Position calculation that are throwing the result off.

Do you have any idea if this will throw things off?
-When I select a feature, it'll often have it's own trihedron in the CAD window that does not line up with the alignment trihedron.

I am more knowledgeable with PC-DMIS and Calypso seems to be throwing me for a loop with some of these things, I apologize.

Thank you.

[[Tr...]]

You can try an alternate alignment outside of the datum reference frame inside the true position and see if it helps. Just make sure the evaluation of each of the features are Outer Tang. It may help you diagnose.

[[Tr...]]

You can upload that alignment to the true position and in the feature menu if needed.

[[Ja...]]

Travis,

I have tried experimenting with all sorts of different alignments (which I was wondering how to do anyway so thank you) and different features but nothing is seeming to resolve the issue. I do not believe the part could possibly be that far OOT so I believe there is something deeper at play here.

I have put in a ticket with Zeiss and am in the process of speaking to somebody that can hopefully discover what is going wrong. I will post the result, if any, when I discover what it is.

Thank you for your help.

[[To...]]

What I am seeing is that when you apply the Datum C to the tertiary datum, the DRF is rotating around the Secondary Datum axis until the axis of C is straight along the X axis. Prior to that, Calypso was best fitting the position of the first cylinder. To prove this, re-process the data without the Tertiary Datum, then look at the Default Printout. You'll probably see a reference to "Best Fit".

For curiosity sake, what is the actual X/Y angle of Cylinder C? Are you using C in your Base Alignment?

So, you either have to make sure C doesn't have any angular issues, or simplfy C down to a single point. For example, create an intersection of C to the plane face where C starts. This may help your answer but it's not really "legal" in my opinion.

[[Aa...]]

Let's get some perspective, here. How far is the axis of the controlled bore from the axis of the secondary bore? How far is the axis of the tertiary bore from the axis of the secondary bore? What are the length and diameter of the tertiary bore?

[[Ja...]]

I spoke with Sam Hoffman at Zeiss and he said that the angular range in which I can pick up the tertiary datum is too small to create a repeatable result. He went ahead and created a theoretical cylinder by recalling the tertiary datum feature and forced it to stay nominally straight. This is now working fine and I am seeing the results that I should. He mentioned that this is not always the appropriate way to solve this type of issue but in this case it may have been the best option.

Tom, I think you were on the right path in thinking that the DRF was getting rotated. It was just getting rotated in such a way that it was not repeatable and giving strange results.

Aaron, just FYI: The axis of the controlled bore was approx. 16" from the secondary bore. The tertiary bore was approx. halfway between the two bores and only about 1" long.

Thank you to everyone that offered your advice.