O-ring groove edge thickness 1

[davidlee97]

I am looking the calculation or empirical data about the minimum wall thickness of O-ring groove to backup o-ring against pressure. For example, to seal the end of pipe, an o-ring groove machined on the end face. How close the o-ring groove outside edge diameter to the outside diameter of pipe can be depending on pressure, material and groove depth.

Thanks!

 

[reverman]

I would recommend the Parker O-ring handbook:

http://www.parker.com/literature/ORD 5700 Parker_O-Ring_Handbook.pdf

 

[davidlee97]

Well with your reference, I went over the O-ring bible again. Unfortunately, it seems nothing related my question. Actually, I am not concerned about the failure of leakage. I want to know the strength issue of the groove. If the groove locates too close to cylinder edge, there will be failure problem of edge broken as pressure imposed. I can do FEA analysis, but I want to know some equation or common sense for it. Thanks!

 

[btrueblood]

Thin wall pipe theory, get enough meat to resist the full internal pressure.

Then add however much you need for machining tolerances.

Then add some more to compensate for Bubba mangling the parts when putting them together ...

 

[gerhardl]

..once one of my old professors said something like this about a strength calculation: --'and thus we would normally neglect the last five digits after the comma of the seven we have, and end with only two digits after the comma for this calculation (showing to a calculation with two digits before the comma!). And then we normally will multiply with a safety factor of five(!)

Unfortunately (or fortunately?) modern material technology, calculation methodes and economy gives another kind of safety factors.

On your question: if you look to constructions where 'un-normed' o-ring groves have to be machined, constructions or 'branch or factory standards' seems to be conservative. Refer for instance to large-sized valves at higher pressure classes where o-rings on flanges are placed well within any critical measurements. Eg.: any safety factors are taken on the safe side.

Are there any existing constructions or 'factory standards' for competitors you could look to?

You should perhaps also look to 'stability issues', the distance left to end should be sufficient to support alignment of the two surfaces, and correct pressure to form the o-ring to correct geometrical form to give sufficiant pressure to all sides in the groove, and not make it possible to give a skew opening when mounting giving 'blow-out, conditions for the o-ring.

Taking practical possible mounting and machining fault issues into consideration, material factors might well be taken good care of?

Distance not less than x times grove width, x beeing at least 5? (Just on feeling?)

 

[tr1ntx]

This is a Strength of Materials / Mechanics of Materials problem, and a tricky one at that. The thin-walled pipe or thin-walled cylinder hoop stress calculations aren't valid because the area in question is, first of all, at the end of the cylinder, and second, too short. The expansion of the material outside the o-ring groove is going to be resisted by the shearing effect at the cross-section at the base of the groove. Maybe if you can find equations for the expansion of a ring, you can use that as a worst-case since the shearing resistance will prevent it from actually being that much.

What kind of pressure are you talking about?

Like gerhardl said, look at other applications. Hydraulic fittings maybe. With valves and pipe fittings, you generally have so much extra material that you're not concerned with being too thin outside the o-ring. Everything I can think of where I've seen such an arrangement, the end of the "pipe" has been inside a larger cap or socket, and thus contained on its OD.

 

[gerhardl]

To davidlee97: I think we need your sketch with data for a 'typical
application' to progress further with advice and thoughts.

 

[bullsmitty]

Would be a fan of assuming full hoop stress on o-ring, pushing outward and look at shear of say 1 inch of wall length. You will be conservative already since the o-ring is not fully exposed to all the pressure... but safe is good....

Then I would shell test against it to your codes and maybe 10% beyond them ie 4.1X

I have put o-rings in seatrings and got them too close to edge where hydro pressure and squishing pressure caused a failure. Still it was surprising that you don't really need that much edge distance... even with brass to be sucessful. Yeah... watch the tolerances... run test so the wall is the thinnest that will be seen.

 

[davidlee97]

Hi all,

Sorry to late reply. Here a simply sketch attached. The blue area is pressurized. A cap (green) connects with body by thread. The pedestal is 1.5" square. This is a existent valve design. Obviously, it is not optimised design. What I need to do is calculating the Factor of Safety. Except the vulnerable cap wall, the O-ring groove is suspectable for failure too. I don't agree on the way of thin wall hoop stress. I think that way will devaluate too much. But I need to find a way to calculate and determine the highest pressure can be used. I found I can try Roark's Formula, but I need to find the fraction coefficient. I will try FEA to decide it anyway. Thanks!

 

[hydtools]

You might consider doing a fixed-free beam analysis. The internal pressure creating a uniform load against the gland OD tending to bend the beam element, the highest stress being at the corner of the gland.

You will have the additional constraint of the cap load against the face containing the seal gland.

You also have the condition of pressure tending to lift the cap off the seal.

Ted

 

[tr1ntx]

If that drawing is to scale, and if you're talking about reasonable pressures, it's OK as is. (eyeball test) I've seen end-face O-Ring grooves with that much OD material before.