Bore ID not measuring true

[[Jo...]]

Hey All,

I am coming across an issue measuring a Bore ID (75.06±.015) that does not correlate with the supplier or other measurement systems. 

We have tried various tests comparing different methods with the attempt to troubleshoot.  Below is a chart comparing different results.

• D probe = regular production probe (5mm)
• D probe new = replaced/recal (5mm)
• 3mm probe = choosing a smaller probe size per cookbook
• down probe = measuring the bore oriented upright instead of on its side
• smartscope A down/A up = measuring a circle on each side of the part
• MTM avg= average measurement of the suppliers' results
• Bore gage measurement

We measure the Bore as a cylinder at 3 sections @ cookbook settings.

Any thoughts or suggestions for what could be going on?

TIA!

 

[[Do...]]

Have you compared values when adjusting your evaluation method from LSQ to outer tangential? 

[[Ke...]]

What does the Form/Cylindricity look like?

What are the results for each individual scanpath?

What is the suppliers measurement method?

Edited 21 hours ago

[[Jo...]]

Please sign in to view this username.

 

Please sign in to view this username.

Outer tangential shows even smaller.

Cylindricity shows 0.009mm

Breaking the cylinder into individual Øs shows results within a few microns of each other. 

1. 75.0363
2. 75.0386
3. 75.0353

We have matched the supplier's method with no noticeable change. They use a Hexagon CMM. 

[[Ch...]]

I agree with Keith.  Throw in a roundness plot and cylindricity plot; then maybe even surface finish.  

Is your bore gage 2pt or 3pt bore gage?  Assuming your probes are clean and qualified, it sounds like a form issue. 

[[Ch...]]

I should mention, I will typically compare an unfiltered vs filtered roundness plot when I'm getting weird stuff like this going on. 

[[Jo...]]

Please sign in to view this username.

3 pt bore gage. 

Filtered cylindricity 0.009mm

Unfiltered cylindricity 0.016mm

 

[[Ch...]]

What does the roundness plot look like?  Is it oval shaped or more tri-lobed or something else?   

The min/max difference I see from you chart is about 35 microns or so.  A test would be to evaluate the circle using max inscribed and then evaluate using min circumscribed.  Then see what the difference between those two evaluations is.

 

[[Ke...]]

I suppose you don't have any certified masters of comparable size?

[[Do...]]

Could throw on a few 2-point diameter characteristics as well to show your min/max values at each level to confirm your roundness plots.

[[Jo...]]

Please sign in to view this quote.

Read my mind, this was my next verification.

We measured a 75.09mm ring gauge. The gauge came in at 75.073mm, so this is roughly 0.017mm under spec and aligns with the delta we are seeing vs the supplier.

So what now? Is the vast head or CMM off? Should we look to get it checked out by Zeiss? Anything else I should do to verify?

[[Ch...]]

The colors on your graph are messing with me, but are you saying 5mm, new 5mm, and 3mm probes were all measuring about the same?  

[[Ri...]]

What size does it show the MasterProbe radius is, as well as the radius of your stylus. 

[[Do...]]

Was the ring gage measured using the current stylus system or the MasterProbe? 

[[Ri...]]

If the radius is off on both the MasterProbe and your stylus, you most likely have the wrong information for the Reference Sphere. If they show correct, then you have something terribly wrong with your CMM. 

[[Jo...]]

Thanks for the feedback guys!

I will verify the masterprobe rad and measurment probe rad now. 

Please sign in to view this username.

 That is correct.

[[Jo...]]

Please sign in to view this username.

 stylus system used for the measurement.

[[Jo...]]

Update: I believe the issue is that the reference sphere is off in some way.

Please sign in to view this username.

 was spot on!

Could have gotten dropped, not 100% sure. I did not find any noticeable damage. 

I verified the probe radius on the measured stylus system was coming in at 2.493mm. Luckily, we have 3 reference spheres on our reference sphere tree. I swapped out the ref sphere and recalibrated. The radius changed to 2.5001mm.

When we remeasured the bore, it changed the result to 75.0518mm. This aligns very close to the supplier's measurements! 

 

Thanks for the help, everyone!

 

 

[[Ke...]]

Is it possible that someone calibrated to the incorrect reference sphere selection?

[[Jo...]]

Please sign in to view this quote.

No, there is a cal routine that selects what ref sphere to use; the operator only has access to use Autorun. They are all the same size (30mm Ø), just in a different orientation/elevation. 

[[No...]]

When was the last maintenance / calibration of the CMM? Was there anything out of spec?

Over the years we had two cases where the maintenance tech had to change the radius of the cal. sphere slightly to get a deviating axis into the tolerance band again. They were quite old machines and it was a last resort (and of course officially documented) before they would have had to be laser calibrated again (CAA). But we never saw any issues with deviating measurement results because of this.