Coaxility Zeiss Vs. DEA

[[To...]]

One our our customers have a lot of small coaxility tolerances on threads. Tolerances are usually 0,05mm.
I think that a tolerance that tight on a thread is not possible to verify by CMM.
They also tend to use threads as datums which I find impossible to verify by CMM.

They do the verification on DEA Global and say that this how it should be verified.

Is there anyone who know if DEA does something different in the calculation that will make it "easier" to verify coaxility on a DEA?

[[Ke...]]

I am doubtful that they use a significantly different calculation method.

I would recommend reaching out to the customer to find out what their measurement method is:
-are datum simulators of any type (e.g. True Pos Pin) or are they probing the feature?
-are they actually measuring the pitch diameter, or are they scanning the major/minor diameters?
-is the diameter of the stylus that they are using the same specification as the over/between pin specified size or are they using the closest standard size?
-are they scanning or are they taking discrete points?
-How many points are they taking?
-what are the evaluation settungs they are using for their features?

It would also be beneficial to have a sample of serialized parts with results from the customer to verify correlation of results.

[[Ch...]]

I too have small parts with very tight coax tolerances on threads. It is possible. I don't like it, but it is more accurate than I expected. In fact, we have about the exact same tolerances and size of features as it sounds like you have.

A few things:

Read the attached document. I use the helical scan, sometimes self-center and sometimes I don't. It all depends on what the MSA tells me.

Make sure to correlate strategy with supplier/customer. Make sure thread measurements start/end in exactly the same place, have the same number of points, same helical pitch and rotations, etc....

Let me know if you find a solution to CPK and outliers. Out of 30 parts, I will always have at least 1 that is an outlier. In my opinion, this is due to the nature of threaded features.

Threaded_Holes_metrology_conf_2011.pdf

[[To...]]

Please sign in to view this quote.

We actually have tried coordinating how to part is measured in terms of above, so we should be as close to each other as possible.