Need To Size an O-ring for Square Face Seal O-ring. 3

[Zibraz]

Is it reasonable to assume that the inner circumference of the O-ring roughly equals the inner perimeter of my O-ring groove? If so I can work out my required diameter and order a few O-rings of roughly that size to increase the odds of getting it right.

The above O-ring has an inner length of roughly 254, so if I select an O-ring of roughly this circumference I should be alright I suspect.

 

[The Obturator]

What is the component for? The ring and groove size may be a critical OEM component requirement, not to mention the o ring material itself and its corresponding hardness. If it's a joint holding pressure, then you may need OEM information if it is available to you.

*** Per ISO-4126, the generic term 'Safety Valve' is used regardless of application or design ***

*** 'Pressure-relief Valve' is the equivalent ASME/API term ***

 

[Zibraz]

What is the component for? The ring and groove size may be a critical OEM component requirement, not to mention the o ring material itself and its corresponding hardness. If it's a joint holding pressure, then you may need OEM information if it is available to you.

This will be for a prototype of a sealed box which will be purged with Nitrogen with other bits inside. I know the required shores hardness and material required for my application. I'm just trying to estimate the size of O-ring required. I'm fairly certain my assumption is correct, but assuming is the mother of all f*ck ups.

 

[robyengIT]

why the inner length and not the average length ? (I mean the length of the axis of the O-ring : in fact you are talking of "roughly the circumference")

 

[hydtools]

If pressure is inside, use the outer groove length to find the oring outside diameter. This 'preloads' the oring to the outside of the groove.

Ted

 

[EdStainless]

Use the outside dimensions, you are better off starting with more o-ring rather than less.

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

 

[3DDave]

I'm not sure about having too much being a good thing - if the o-ring has to stretch a bit it will tend to stay in place. If it's loose or too large, getting the lid on is a struggle as it will try to return to the round shape and be out of that groove. Don't ask how I know. I have no comment on that. A compromise is to have the groove be wavy to put some back-and-forth bends to capture it. It's also why dovetail grooves are popular.

 

[Scuka]

Use volume.

Volume of an ideally circular ring must equal to the volume of a ring formed into the rectangular groove. If that's the case, your O-ring will fit into your groove.
I worked with wire bending, and when figuring out what the starting length of the wire must be to achieve the desired bent dimensions, I would check whether the volume of a formed wire matches the volume of the straight wire on my 3D model. It's a simple application of law of conservation of mass. That way I never had to bother calculating circumefences of radii or rack my brains whether to use inner, outer or middle perimeters.

Of course, if the O-ring needs to stretch significantly (I don't have much experience with O-rings) as you put it in (i.e. the diameter of the cross-section changes), it's gonna complicate your math a lot.

 

[SnTMan]

See also the Parker O-Ring Handbook.

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[Compositepro]

Lengthy o-rings are often made or replaced by using spools of o-ring stock cut to the needed length. What is the needed length, you ask? It is long enough to fill the groove plus a small amount of overlap at the ends. The o-ring will then be under slight axial compression when installed, and no adhesive is required to keep the splice together and sealed. The o-ring diameter should be slightly larger than the groove width, and friction will keep it in place. The spliced ends will seal tightly due to axial expansion when the o-ring is radially compressed. This is why o-rings are extensively used as seals. They seal very reliably because they conform to sealing surfaces, and pressure causes them to seal even tighter. They don't have to be perfect to work perfectly.

 

[TugboatEng]

If retention is desired, the groove should be dovetailed.

As for dimensions, I see no reason you couldn't translate the sizes into a circular dimensions, perimeter to circumference and then use standardized dimensions from the aforementioned Parker document.

 

[hydtools]

Be careful filling the groove with oring. There needs to be space into which the oring can fill when it is squeezed.

Ted

 

[CWB1]

O-rings work by compression which directly alters ID and OD, so personally I would recommend sizing them based on the centerline and cross-section (volume works too as mentioned).

 

[robyengIT]

For pressure pipes according to AWWA C301 (prestressed concrete steel cylinder pipes) we have the same problem. Here attached the spreadsheet I use to design the correct size of the O-ring. For such a pipes the working pressure (in average) is 12 kg/cm2.
Modifications have to be done in few cells to fit your configuration : I think is valid to state the max diam of the O-ring to fill as max as possible the groove, but in any case < 100% as mentioned above by hydtools and according to what said by Scuka

 

[robertib]

We regularly use o-rings in this configuration, and our policy is to use the inner perimeter of the groove as equal to the inner circumference of the o-ring. From this we can set the o-ring ID. In our case, sometimes the pressure is on the ID, sometimes the OD, but we have had no problems with incorrect seals in the past - except when user calculations are wrong.

 

[FACS]

O-Ring gland design is readily available for reference.
SnTMan pointed to a great source. Research properly.