Target Points Help

[[De...]]

We have a part that calls out (4) 0.25" target points on each corner of the part for a datum structure. The print allows us to eliminate one of the points so that only 3 need to be used. We are having a hard time getting Calypso to filter out one point. This is a bigger box type part with components that go together, so there tends to be a little bit of a bow in the part. So eliminating one point will help us tremendously here.

I have it set up currently to take a plane within each 0.25" targets. I then create a space point (Outer Tang) within each of the planes to create 4 individual points. From there i recall those points into a plane for a plane of 4 individual points. I have this plane set to outer tang with just the standard filter on. This is not eliminating a point for us. I am assuming there is a filter setting that I am missing or maybe I am just doing this completely wrong. Any help would be greatly appreciated.

Thank you in advance for any suggestions!

[[Ke...]]

You are speaking of Target Points, but describing Target Zones; and I would approach each of them differently. Can you clarify which is specifically called out on the print?
I believe that if you are recalling 4 points into a plane, and utilizing Outer Tangential evaluation method, then Calypso would be using the 3 outermost points, which would be "masking" the 4th point. Filters & outliers don't do much for limited data points.

Datum Taget.png

[[De...]]

My apologies. It would be the middle one, so a target zone. Thank you for catching that

[[Ke...]]

Please sign in to view this quote.

So, for a target zone, I would use the entirety of each zone (instead of reducing each zone to a point) when recalling into the Datum Plane, and use Outer Tangential evaluation.
If your part is too warped/bowed, the CMM may have trouble with the stability of the feature; just like if you are mounting the part on 4 posts, and the part is so warped that it rocks. I think that you would just need to agree on which 3 datum zones to use, and move forward from there(?)

[[De...]]

So essentially it will have to be manually removed each time due to the gap changing location each time they bring a part in. Thank you for the help, I really appreciate it!

[[Ke...]]

Please sign in to view this quote.

Personally, I would want all 4 pads measured anyway, to see what the overall flatness looks like.
Flatness of the "pads" could be something that you check first, and include as a reference check, to see what the trend is. If the faltness could be controlled better, it would be a more stable alignment, and you would get more consistent results. A preliminary review of the flatness may piont to if/when you should use 3 rather than 4 datum pads.
If it is up to you to decide which 3 to use, you could create alternate alignments using different patters of 3.

Just some thoughts/ideas.

[[Ke...]]

Heres a helpful thread, too:
viewtopic.php?t=10268&hilit=datum+targets

[[Er...]]

During machining, they same target zones should be used to make the parts. Whatever was used, is what I would use for inspection. If these weren't used, just like Keith mentioned, you may have to inspect with the different alignments to find which produces a conforming part, if any of them.

We had a similar situation where the machinists put in a machined surface, and then located to the machined surface. Everything machined was great, except for where everything was supposed to be located to the target zones of the casting. It was an expensive lesson needless to say.

Good luck!

[[Er...]]

Try Inner Tangential on the four points and see if it agrees on the point removed as you run?

[[De...]]

Erik, you are exactly correct. The machinists will shim whatever corner contains the gap to square it up and keep everything as "true" as it can be.

I will try Inner Tangential as well.

Thank you both for your input!

[[De...]]

So to reiterate, we are trying to use the 3 highest target points (created from the target zones) and the CMM has a hard time evaluating Outer Tangential with ISO5459. What I am understanding is if the twist/bow is too great, Calypso will have a hard time eliminating the lowest target. (4th target)

Is there a way we could use a formula or method in Calypso to automatically eliminate the low point and not try to fit to it?

[[Ma...]]

Can you use Min coordinates to get min in Z for example? I use that for curves.

[[Er...]]

Derrick, Are you sure it's the 3 highest points? All you know is based on the machining fixture you have 2 points that is making contact, and 2 other points where the part is "rocking." They proceed to touch one of them, and shim the other to eliminate the "rocking" of the part. It seems to me, it would be easier to just pick the 3 zones used, shim the 4th, to keep the machining/inspection standardized. Otherwise, I would think they would need to tell you which point was shimmed, and eliminate this zone from the plane.