Filter for roundness measurement

[[Kh...]]

I use MICURA and Taylor Hobson TR130 to measure roundness. However, with the same filter, I can't get the same result on two machines. Is anyone know how should I set filter on CMM to get the correct roundness result?
I've already tried a few different configurations but can't get consistent matching result with TR130.
Thanks in advance.

[[Da...]]

What level of difference are you seeing? Are you within the accuracy of the machines? (ie a result difference of a few microns)

What sensor does the Micura have? How fast are you scanning? Point density?

If your point density is not high enough, Calypso won't apply the filter. IIRC, you need at least 7 points/wave.

Are you scanning in the same spot, and fixturing the parts the same?

You need to give a lot more information to get a good answer.

[[Ti...]]

I am interested in this topic as well.

I use Duramax and separate roundness testers. I have not been able to quite match up the performance between the 2. Understanding of course that Duramax and XXT are not top of the line hardware, it does surprisingly well.

In my view, a cmm can do many things but there is a reason for roundness testers to exist as a separate specialty instrument.

[[Da...]]

There are some basic things you'll have to do to get similar results. but first a few questions.
What diameter size are you checking ?
Can you get the probe sizes to match ? (same size on both machines)
What are the filter settings on the Roundness tester? Should be displayed on the print out.
Is the part tapered? Is it a cone ?
What is the surface finish of the part ?
What is the final roundness tolerance ?

[[Kh...]]

Please sign in to view this quote.

If I apply the same filter, the difference can be as high as 50%.
After tested a few different setting, best result I got is about 15% higher than TR130. I took 3000 points, which should be more than enough. The speed is set at 20mm/s but I only run at 50%.
I measure the same height with diameter ~ 50mm not too big or small. The surface is honed cone. Stylus tip diameter is 3mm.
I'm using low pass, Gaussian filter for TR130 and Low pass Spline filter on CMM.
Is there any thing I'm missing?
P/s: If you have document on this topic and ready to share, I appreciate that. Thank you

[[Se...]]

Hello Kahn,

Please tell us how big are the differences. The percentage you´re giving is depending on the tolerance. We should know if you are talking about 0,001 mm or 0,05mm!
I see 2 more important points that have not been addressed here: [list=2]

Do you have the same outlier elimination on both systems?

[list=2]

Is your evaluation of the roundess according to the Chebyshev diameter at the TR130? Calypso´s standard setting is Chebychev but I know some other CMM softwares need a manual setting

[[Da...]]

The steeper your cone is the smaller the chance that you will get a correlation. We have seven roundness testers from several manufacturers , 3 three micuras , and 2 cylindricity testers. I have researched this problem for a few years. Talked to the programmers and engineers in Germany, and unless there was a major change that I didn't notice roundness on a cone is not possible. You'll get a number but it will never match what you get with a roundness tester.

[[Kh...]]

Please sign in to view this quote.

Wow, Thank you. I though I had done something wrong.

[[Kh...]]

Please sign in to view this quote.

Good day Sebastian,

The tolerance is <10um. I will check the other two points you said. Maybe TR have some kind of auto filter I'm not aware of.
Thank you.