Finding Centerline of a Keyway

[[Al...]]

Hello everyone,

I am fairly new to programming a CMM and was wondering if you have any advice for finding the centerline of a keyway back to the cylinder its located on. I have tried a few ways to stop the rotation of the cylinder, but always seem to come up with results that do not match with what I can show using a height gauge and plate.

Thank you very much,
Al Chrostowski

[[Ri...]]

Can you post something from your blue print?
it will provide information on what you mean by rotation of the cylinder

[[To...]]

Have you tried measuring the sides of the keyway as planes in a symmetry construction?

[[Re...]]

I would also like to know if anyone has a good method for checking keyways or keyseats on the cmm. This may not be technically the correct way to establish clocking of a keyseat on a cylinder. I usually find that the sidewalls are much too small or have too much form error in order to establish a reliable clocking with a symmetry plane. I find that the larger back face has less form error and is a more reliable feature to clock to, then I look at the position of the symmetry plane of the side walls to this alignment.

[[To...]]

What characteristic are you using now?

I was asking myself how you know the results were wrong from measuring on the surface plate. I assume the part may be sitting on a flat surface which is clocking the part. So, if I were to take another look at this, I would do a cartesian distance from the cylinder to a symmetry point, but add a datum feature to the primary datum field. Possibly the same feature the part is sitting on for the surface plate measurement. Or, create a theoretical plane feature that would be aligned to the base alignment and use it as the primary datum feature.

Screenshot 2021-05-25 090009.jpg

[[Er...]]

To get the granite and CMM to agree they have to be using the same references and assumptions.

On the granite... If using a v block standing upwards, leveling the bottom of the keyway, then flipping the v block on it's side to find the difference from top and bottom of the shaft...

On the CMM this will need to be duplicated. Taking measurements in nearly the same points as the v block and indicator would touch.

The assumptions are the keyway was cut perpendicular to the shaft on both axis, that cutting tool flex was taken into account, etc, etc.

[[Me...]]

To Erik's point width and depth of the keyway still can make it difficult to indicate in.
In the past because of limited depth and width I would find a gage block, or stack together blocks, that would fit snugly. That would project more indicating area.
Then, maybe, use that setup on the CMM.