Half Dovetail Groove Orientation

[EricHill]

A company I am working with has vacuum chambers as part of their equipment design. They use half dovetail o-ring groove seals fairly often, instead of full dovetails. One thing they do differently than what I have done, and seen, in the past is that they orient the groove so that the angled face is on the vacuum side, i.e., the o-ring will be pulled against the angled face rather than against the flat vertical face of the groove.

My understanding had always been that it is prefereable for the o-ring to be pulled on the flat face,so that the seal behaves like a square walled groove - the preferred geometry.

I be darned but I can't find any references anywhere to the preferred orientation of a half dovetail o-ring groove. Anyone here have any insight?

 

[btrueblood]

I'd agree with you Eric, FWIW, but can't put hands on a reference to state that either. Parker Handbook gives a reference to the "Parker Vacuum Seal Design Guide", ORD5705. You might be able to find it with a search of their website, and it might have the info. you are looking for in there.

 

[EricHill]

Thanks for the suggestion about the Parker guide - I had requested that from them some time ago. It doesn't have the information I am looking for here, it's mostly just some adjusted groove dimension charts for vacuum service.

 

[Timelord]

EricHill,

I regularly use half dovetail o-ring grooves. The angled face must face to the inside of the groove for it to retain the o-ring when the o-ring is streched into the groove. Putting the angled face to the outside renders it useless for retaining the o-ring because the o-rings is pulled towards the straight side by its tension. The Parker handbook doesn't mention single sided dovetails, but they work on vacuums and are easier to machine. A vacuum is such a low pressure that I doubt it even moves the o-ring to one side of the groove if you have the correct percentage squeeze. The seal is made by the bottom of the groove because practically that is the only surface you can control the surface finish of.

Timelord

 

[hydtools]

Half-dovetails are in the Parker Handbook, table 4-5. But show no preferred orientation.

Ted

 

[Timelord]

Hydtools,

I stand corrected, I never looked at the .pdf of the handbook until I saw your post. My old hardcopy version of the Parker handbook doesn't show single sided dovetails.

Timelord

 

[EricHill]

Thanks for the feedback!

I finally got hold of a Parker applications engineer and he confirmed what I had suggested - the groove should be oriented so that the o-ring is drawn to the flat side. This makes it behave like a standard groove seal.

Another related issue is that it reduces the trapped volume of gas in the lower corner, which is important in vacuum applications.

I have put a few o-rings into half dovetails before - the o-ring cross-section is so much larger than the groove corner width that you really have to force it in there. Retention does not seem to an issue as long as you have the right amount of stretch in the design.