Profile of multiple planes

[[Na...]]

Hi everyone,

I am relatively new to Calypso and I cannot figure this out. I have spent hours looking through the forum and the help file with zero luck.

I have a part that is basically a castle nut, minus the threads. There are 30 teeth to this part and the drawing calls for 60x profile .0039 to D|E (each face of each tooth). Datum D is the ID and Datum E is a face of one tooth. I got a hold of Zeiss and they walked me through the freeform setup on this part and the results are what they are. Each face is scanned with 50+ points. The program runs fine, however I am questioning the results. We have a Hexagon machine running PC-DMIS 2023.2 and a brand new ZEISS Contura. The Hex machines are passing the parts, but when they are run across the Contura the results are almost double. Hex machine is giving 0.002 - 0.0035 and the Contura is pushing 0.007. The Hex machine that we use, and must use per customer requirements, is down and so now we have to use the Contura to finish machining the parts due to deadlines. I have tried my best to mirror the flow, probe size/length, and point counts on both platforms. Same holding fixture used on both machines. Can you please point me in the right direction.

[[Ma...]]

I don't want to doubt your experience in measurement, but could it be that Hexagon machine is giving you number per ASME ( two results ) and Calypso is giving you one result ( which is doubled the worst deviation )?

[[Na...]]

PC-DMIS is reporting per ASME 14.5, however I am not sure what standard Calypso reports in.

 

I did switch Calypso to bilateral - two results to perhaps get a better idea of what's going on. I know that I am missing something. Could this be that the setup of the freeform is incorrect?

 

I'm okay with you questioning/doubting my experience. That's why I'm here 🙂 

[[Ma...]]

I am not sure if you have correct alignment due (M) for datum D.

Now you have similar view like from PC-DMIS, but it seems like you have still something different.
How is your filtering and outlier elimination?

It would be helpful for you if you enable graphical evaluation and look where are those min and max deviation.
It should be at start and stop of scanning path, i believe.

[[Na...]]

I am using default filtering on both machines. I played around with filtering a little, but I am not that experienced with it. The Geometry Best Fit alignment was suggested on the forum on another profile thread. I gave it a shot, but when you switch to a feature alignment the results are nearly the same (-0.00368, 0.00331). Under the standard feature control frame the (M) option does not appear. There are 21 planes that show out of tolerance when graphical analysis is turned on. 20 planes show all out of tolerance, including starts and stops. 1 plane shows the middle third of the scan passes but the 1st and last third do not. Here is what the filters are set to for the freeform. I have also looked through the other features that are associated with this and I could not see anything out of place.

 

[[Ma...]]

Can you post your datums for this callout? To tryout purposes.

You can disable bestfit on your surface profile. It won't apply unless you use no datum reference or you make an alignment from it and then use it elsewhere.

I hope someone will join this thread.

[[Na...]]

Datum D is made up of single points due to the fixture interrupting the scanning process. Datum E is a tooth face. All 60 teeth use the same scan path, point density, scan speed, etc., etc. I have 4 days worth of Gear training next week so if we can't figure this out then perhaps next week will bring an answer.

 

 

[[Ma...]]

Is that part that small? It looks like 0.33mm length of a cylinder is really small compared to it's diameter.

Is that something to watches? 🙂

Would be acceptable to have main axis from plane perpendicular to datum D and datum D having only circle?

I think your problem is not having MMB on datum D.