Thinking out load and brain storming I have never run up on this large of a seal.
But that said I believe a go and no go is not practical.what is the elasticity of the seal. In other words how much does it stretch.
I would say make a restraining fixture
To hold it in the restrained condition. And
Inspect it on a coordinate measuring equipment.
Do sampling plan . And maybe a functional gage to make sure it will assemble. Food for thought

mfgenggear,
Thanks for the idea. It's flexible. the hardness is about 75 shore A. The goal for the go fixture is to go in without any compression on the seal, would it be better to cut the fixture in half to expose the cross-section? that way we have a better view and perhaps can use feeler gauge? But it would slow the inspection down for sure.

I will also think about the CMM route, although, I doubt the management would like it since it's for production lots, the CMM is often occupied by other tasks.

Ya management cringes when CNC coordinate
Machine is requested. Well darn. You could try to put a slght high limit Diameters, no go and low limit diameter go gage. On the fixture.
Make snap style gage check the ID to od.

The main idea is to make sure it assembles

Why not use CCD to check the diameter?

I have never seen a go/no-go for a rubber seal.
Maybe make a test fixture that has min/max diameters of part it seals, then do leak test?

Chris, CSWP
SolidWorks
ctophers home

What are you checking?

Thickness?
Roundness?
What does "profile" mean? Have you got a drawing of the profile?

ID? OD? to what sort of tolerance?

On the basis that its the ID and OD you lay it flat into a cut out piece of ply and then have another piece of ply to go into the inside?That would give you max OD and max ID but you might need two other for min OD and min ID

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

Forgot to update the thread. I later realized that the profile is only on one side of the seal, so the fixtures that I initially proposed would only be checking the max/min material condition of entire part, which is not exactly useful. See link below for the actual drawing:

https://imgur.com/a/5mc1RrY

Below is my interpretation of the profile tolerance zone:
https://imgur.com/a/6YCafQ9

Currently, I am leaning more towards using a scanner or a mold and inspect it on a vision system. Is it reasonable to ask the supplier to provide a segment of the seal along with the production lot? That way we can just put it on an optical comparator directly.

The only way I can think of measuring those points is to cut a cross section, lay it flat, then measure points and profile.

Chris, CSWP
SolidWorks
ctophers home

What are the critical bits you need to be no go? Clearly not OD if you've got 0.5" variation, but angles? locations of those very small little keys, radius of the ends?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

I think I'd lean towards making a jig to present the part for inspection "as installed". I'm assuming you're concerned about the ID functioning to seal. Get it "installed" and have a band of steel or something to measure the ID based on the ring gap or something, and also have a "go" gauge pin/cylinder that's centered and is min required diameter to seal against. Then you could check it'll engage and also have a reference of the diameter that's quick to get.

Is the angle on either side of the rounded ID lip critical? you could also have a drop down ruler to get an angle measurement.

Depending on the application pressure, you may want to make a hydrostatic test jig that can also pressurize while checking the sealing function.

We used to inspect elastomer parts by using fixtures that stretched them slightly.
If they slipped on easily then they were too large.
Once they were on the fixture we could measure at a few locations and confirm that width and/or height were correct.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

The tiny keys are the allowance for the mismatch, the fillet radius is not part of the profile per my interpretation. Since the profile callout has no datums, its only controlling the form and size, so maybe I could break up the profile into length and angle which is easier to measure? To directly measure the profile without a CMM, I think you would need to mold it and then put the mold on an optical comparator with a mylar? Or similarly, use a scanner to scan the profile and then import the CAD drawing (nominal) and then use the software to create the .50 profile tolerance.

The request came from Quality, they just want to inspect the profile without having to use a CMM.