Line profile on sphere

[[Da...]]

Hello measuringheros,

I'm a production / engineering guy who's developing tools and support the production. I can handle 80% of our measuring equipment, but unfortunately I have never gotten a training for our CMM and the Calypso software and obviously learning by doing is not an option here. So please forgive me if I have wrong assumptions or ask stupid questions.

We're producing a cold formed spindle and need to measure a line profile on a sphere:

Currently our quality department is creating a reference point on the right side of the part, then move a fixed distance to the left to measure the line profile. If the shaft with the small diameter changes in length, the measured line profile is (of course) not OK anymore. Now they asked our operators to control the length on the upper limit in order to have the sphere inside the tolerance. But this caused already a lot of machine downtime's to get the sphere and length within tolerance and even the customer complaint as some parts had the length of the small shaft exceeding the upper limit.

In my imagination it should be an (more or less) easy thing to measure the profile without depending on any length, it would be probably something like:
- Get the axis of the spindle
- Find diameter 14
- Create a sphere on the axis with diameter 20 that touches the diameter 14
- Measure the line profile between 41.427 and 47.427 degree

But our quality said that would not be possible... 🙄

Am I thinking to simple?
I would appreciate if anyone could give me an advise that I could bring to quality guys and gals.

[[Ma...]]

I would just scan for a sphere R10 with locked radius to get position, then from center of this sphere i would make simple 2D or 3D curve given by angles. That dia 14 is useless if basic angle is used. Because with given radius and angle will give dia 14 ( trigonometry - sinus / cosinus ).

But i am not sure if line profile must use center from shaft or sphere.

[[Is...]]

this is for a curve in a Sphere:

viewtopic.php?p=29138#p29138



here some background:
viewtopic.php?p=35328#p35328

[[SH...]]

Please sign in to view this quote.

I think radius R10 and angle are required for runout not for profile, since no datum in the FCF it is a form control.

[[To...]]

I agree. The profile of a line is form only.

[[To...]]

I agree. The profile of a line is form only.

[[Da...]]

Please sign in to view this quote.

Thanks for your replies Martin, Israel, Shabu and Tom!
Even I have no idea about CMM measurement, but that sounds reasonable to me. The Ø14 is probably only used for the run-out and as reference point for the length (with the ±0.08 tolerance) of the big shaft. I assume that the center of the profile is on the center axis of the spindle.
I will bring this to our quality and hope that they know how to do it. 😃

[[Da...]]

Please sign in to view this quote.

With other words:
The center of the SR10 don't need to be on the center axis of the spindle?

[[Er...]]

"With other words:
The center of the SR10 don't need to be on the center axis of the spindle?"

The Circular Runout may measure circularity and coaxiality per Y14.5 back to datum A. I would scan 3 circles on the sphere minimum. Use point recall to create the circles, add range limits, and pick the different paths.

The Line Profile is only two dimensional and the customer has a very specific area to inspect. If I am reading this right, no datum is specified so it is measured to itself. Thus the suggestions to "find" the sphere first. With an XXT I would have to measure a sphere with single points, set that as an alignment, then scan the sphere in that alignment.

Capture.JPG

[[Da...]]

Thanks for the explanation Erik!

As said, I'm not familiar with the CMM measurement, so

Please sign in to view this quote.

sounds good to me, even I don't know what it means. 🤣
I will bring it to our quality department and hope it will help them. 😃