Checking stock...

[[Da...]]

We have a cylindrical part that has 2 different ID cylinders. The finished part has a total runout tolerance of .0004 (inch). The part was machined, and heat treated by our customer. The customer left us approximately .003 Stock on the IDs, but the customer ran each ID separately, so there is some runout between IDs. The Total runout spec is from one ID to the other. I've been asked to check to see if the parts will cleanup at final size. I've tried a few different ways to check this, and none have been successful. What are your ideas?

The part is 4.0 long, the ID's are each about half of the length and around 3/4 or smaller. I have a model. I have curve. I do not have PCM

[[Mi...]]

Diameter set to the Circumscribed Evaluation should show the amount of material available to cleanup.

Assuming final machining happens with perfect form, location and orientation so there is that margin of error.

[[Da...]]

That will not account for the runout of the cylinder. There is only .003 stock on diameter, so roundness and runout are very critical.

[[Da...]]

Is there a way to measure radius from the centerline of a different cylinder? If I use radius measurement it uses the C/L of the circle I just checked, not the cylinder of the other ID.
Space point might work but, Calypso can't seem to get my vector angles correct. (it presents them pointing into the material) .

[[Ja...]]

I think you can do a surface profile evaluation to check stock

One option is by scanning these ID's as a stepped cylinder (Part is 4.0 long so this will only work if can you can measure both ID's in same setup). you can setup the DRF of that profile characteristics depending on how the machining will take place or use the feature itself as DRF.

[[Da...]]

Please see the "Space point" post, Calypso is throwing a wrench in the problem!

[[Ja...]]

Please sign in to view this quote.

The method I suggested does not use space points. It uses cylinder/step cylinder features itself. You can set the nominals to finish diameters and do the evaluation using surface profile characteristics. Looking at the graphic you can figure out if it will clean or not and also which side has the least stock.

[[Ti...]]

Assuming you can CNC scan both cylinders in one shot with either a long probe or RDS, and your base alignment is based off of the OD, try this:

1. use helix and line strategies on both cylinders (just get LOTS of points)
2. profile the cylinders (tolerance doesn't matter)
3. turn on CAD evaluation at 1x mag
4. go to wireframe and use the standard views to look straight down the part and see if _any_ points drift outside the CAD nominal of the ID's and that they have plenty of cleanup room/material between the scanned points and the CAD nominal

bit of a hamfisted visual method, but boy will it look smart to the foreman...obviously you're limited by the point density, probe diameter and length, etc, so you won't be able to objectively tell if they'll clean up all the way to the intersection between the ID's but everyone should understand that it's an unavoidable shortcoming with probing anyway

just in general i've kinda soured on using gantry CMM's for tight runout measurements, the part should be rotating and the instrument static and that's backwards with a CMM, hopefully your customer lets you verify the final .0004 runout in the lathe spindle with a tenths indicator in the constrained condition because that would be pretty tough to measure outside of the lathe spindle you made it in

[[Ma...]]

Use radius measurement with coordinates of the circle constrained to the desired datum(s).

If you take more points than you care to report, mask the results then use Max Characteristic to find the greatest radius (last to clean up) point.

[[Da...]]

Please sign in to view this quote.

Mark, this worked perfectly! Thanks for the help everyone! As a double check I found the difference between the Max and Min radii, and it closely mirrored the radial runout of the test diameters!