O-ring groove design for o-ring retention during assembly

[mkslice]

I am looking for some design recommendations on o-ring groove design to help retain an o-ring used in a face seal application.
With the standard rectangular groove the o-ring does not stay in the groove reliably and it can not be checked after assembly.
We have tried the dovetail type groove in the Parker handbook and found that the o-ring is very difficult to install and ends up with the o-ring being streched during assembly so that it appears the o-ring is and inch or more too long. This is a groove approx. 10" in length and is not a simple circle but a path around a rectangular part that has some bolt holes that have additional clearance for between the groove and the through hole.
Thanks in advance for any help with this issue.

 

[flexibox]

Have you tried using Silicon Grease to hold it all in position. If not try it you may be pleasantly surprised.

 

[israelkk]

Beware of silicon grease when the o-ring is made of silicone rubber or fluoro-silicone rubber.

 

[1gibson]

Possible to freeze the o-ring in the installed shape? Using a dummy part in the freezer. Then remove o-ring, install on part for assembly, use white lithium grease (lubriplate) to hold it in, assemble.

 

[mkslice]

Thanks for the ideas so far.

We tried the grease previously but the radius around the bolts and at the corners are usually where it pops out and grease is of no help there.

Freezing is not an option as this is a production line enviroment and freezers next to the line are not too practical.

Any other ideas out there?

We may try a single sided dovetail groove as a compromise.

 

[1gibson]

Is the part ferromagnetic?

Thin magnetic strips in strategic locations. Set parts together, pull magnetic strips, bolt together.


Or similar with not-so-sticky tape?

Seems like you might need to cheat a little on this one.

 

[israelkk]

What is the pressure and environment needed to be sealed?