Best probing method

[Anonymous]

Hello,

We are validating a process of reverse engineering of custom precise parts. We are using a Helmel manual CMM, with PowerINSPECT.

We were assessing the best way of probing points for a circle feature and I wanted to have an external opinion.

So, when measuring a circle feature, is it more precise to probe a circle by: 1- probing points opposed to one another (doing a star) or 2- probing points aside, like going clockwise all around the feature?

The calculation behind measurements are depending on the machine itself, or is it done by the software?

Thanks in advance for helping out.

Regards

[prashant.gurav]

Hi @Anonymous, 

Welcome and thanks for contacting Autodesk PowerINSPECT Community.

In PowerINSPECT the order you probe the points dosn´t matter. The result will be the same.

Do you mean, how many points you should probe.

Usually diameter is measured with minimum of 4 Points. Accurately one can probe a minimum of 8 points

Please accept as solution if my post fully resolves your issue, or reply with additional details if the problem persists.

Thank you.

[Anonymous]

 Thank you.

Actually, we probe 6 points to complete our feature. Our challenge is, we need to perform circle measurements on cylinders that are not in the axis of the machine (our CMM is a standard 3 axis, without indexing head).

We know the diameter of the probed cylinder, but we want to obtain the information of its position in the workspace. So we start by measuring a plane on top of the cylinder, then do a circle projected on that plane. It usually gives us reliable results but the diameter isn't always consistent while circularity is top notch. Then if we focus on getting the right diameter, now we have a wrong circularity.

Would you rather focus on getting the right diameter, or lowering circularity error to achieve the best precision on the position of the circle?

Thanks again for the help.

[prashant.gurav · Accepted solution]

Hi @Anonymous

Thanks for getting back to us.

If after 4 or 6 or 8 probed points the circularity is bad, then you need to scan the circle and you'll receive the final value so the full amount of the circularity of the circle. If the circularity is bad, then you can't expect super precision and repeatability for the diameter, because the diameter is a best fit, so PI tries to best fit a circle into an 'ellipse'.

Make sure the reference plane is perpendicular to the cylinder's axis, because if not, then the cylinder may have very good cylindricity but the resulting circle will not, as the inclined plane cuts an ellipse from the cylinder.

Please accept as solution if my post fully resolves your issue, or reply with additional details if the problem persists.

Thank you.