Alternative solution or improvement to this sealing design

[Znjmech]

Hello All

Apart from what the functioning and conditions are. I'd like to ask you guys , if you could tell me whether this solutuon needs to be corrected to avoid leakages. Currently, the cover is to be screwed down up to the point that it doesn't press the O ring ( why ?)
while the o - ring installation is quite a nasty job, since the inside plating might be to high and the O ring might not slide at its correct position. Mean while there might occure other problems like the ovality of the container or the impreciseness of the threading.
Is there any improvement in the geometry of this design that could reduce the effects that I mentioned to reduce the leakage?

 

[btrueblood]

If the o-ring doesn't get squeezed (by typically 10-30%), it won't seal. See the Parker O-ring Handbook, which can be downloaded in pdf from many sources.

 

[Znjmech]

Even if we say it's squeezed horizontally it's not enough? Doesn't it become statically indeterminate if pressed also from the top? Is it secure to be pressed by the screw? Cause the screwing doesn't arrive always to the same point

 

[btrueblood]

Sorry, wasn't clear that you had squeeze applied radially.

What do your tolerances and/or measurements of actual parts tell you, i.e. how much deflection of the o-ring occurs?

How much pressure are you asking for the seal to resist? The geometry at the corner (between "cover" and "container") is a path for an o-ring to extrude into and fail the seal.

 

[dgallup]

As btrueblood said, look at the Parker o-ring handbook. Everything you need to know is there for 99.9% of sealing applications. You need to actually compress the o-ring in the radial direction and allow excess clearance in the vertical direction to allow for thermal expansion and swelling. You need much better lead-in angles and radii, you need proper surface finish. It's all spelled out in the handbook along with material recommendations based on the fluid(s) being sealed, temperature, pressure, etc.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[Znjmech]

well , Thank you very much, reviewing the handbook , I haven't seen a similar case , mostly I see that the O-rings are encastrated into a slot over the diameter, they are not similar to this case.

 

[btrueblood]

Well, close to your case is shown in the section that describes SAE- and MS- boss seals. Those require a flush contact between the mating parts, unlike the corner geometry shown in your sketch.