using o-ring as pressure relief valve

[UBbaja]

I have a set up where i am trying to use an o-ring as a pressure relief valve. My idea is to put the o-ring over a cyclinder (there is a grove machined in teh cylinder for teh o-ring). Does anyone have any expiereince with this? I am trying to find an equation that I can input material properties (mod of elasticity), the grove /cyclinder dimensions, and will output to me the pressure at which deformation (ie- pressure relif) will occur. I have researched this subject but havent come across anything for tehis specific application. Most i have found have been for teh forces and stresses for common o-ring sealing applicaions. Thanks!!

 

[EdDanzer]

This would only work once then the o-ring will be damaged. Repeatability of relief pressure between o-rings of different batches or manufacture will be poor.

Ed Danzer

http://www.danzcoinc.com

http://www.dehyds.com

 

[kevindurette]

The deformation will always want to seep into the gap, and then the O-ring will extrude. You probably won't lose significant sealing until the O-ring is destroyed. You're probably better off with a spring-actuated mechanism in which the spring fights the fluid pressure. As the fluid pressure builds, it overcomes the spring action and forms some kind of opening. Look at how ball check valves are designed.

 

[UBbaja]

Thanks for the advice kevin. The issue with a spring-actuated mechanism is teh cost, that of which i am trying to get down. The application is for a product will only be used once. During usage if teh pessure builds too high, this would relieve pressure. Becuase of its one time usage over a 10-30 minute period of time I dont think deformation will be an issue. All out failure would be a problem though, as if it failed when it relieved teh pressure when teh pressure was above teh limit, then when teh pressure dropped too much flow would go through teh broken relief valve.

 

[israelkk]

It is not really clear from your description how you are using the O-ring. A drawing can help. This is a very unconvetional and risky approach and even if you will succeed once or few times the accuracy of such a solution will be very poor. You can expect large variations between the systems. How are you going to test each system if you can use it only once? When someone designs a burst disk he makes thousands of tests to create a statistical database. When you buy an o-ring from different suppliers there ia a large difference in the o-ring properties that will not affect the usual intended use but will affect the perfomance for your application. In contrary to metals, rubber properties changes with time, In fact the shelf life of a rubber is normally 5 years. Where metals last for 50 years and more.

 

[kevindurette]

The failure pressure won't be predictable to a high degree, so if this is a safety device, I wouldn't want to rely on it. When I say "spring" that doesn't necessarily mean it has to be metal. Polymers and elastomers are frequently used as spring materials in cheap applications. The tolerances for elastomer machining and molding are typically 30 times as high as for similar metal processes, so you'll find a wide standard deviation for such burst pressures.

How cheap does this need to be? Understanding your application will help.

I happen to work in the hydraulic sealing industry, but I'm still pretty new at the job. We reject stuff all the time in our first article inspections because it is true that there are variations even within a single batch.

--------

To my knowledge, the following is my idea, but I can't guarantee that nobody else has done it. I hereby release it to the public domain, and I make no warranty, express or implied.

You might want to look into designing a seal groove with a notch (on the axial+radial plane) on one surface that stops just short of causing a leak. When the seal deforms under pressure, a clearance is created as the notch is exposed, thereby creating a passageway for escaping gases. What do you think?

 

[UBbaja]

i must be confused... What you are describing sounds a lot like my original idea. Thanks for the input.

 

[EdTeach]

I would not try to design a pressure relief device. If you are protecting against a over pressure event, you need to have very good justification of what you are doing. This is not an application that you can justify cost cutting.

There are several existing products that should meet your needs. Your application sounds like a rupture disc would work, they are not very expensive and are 1 use items, since once they break you buy a new one. Also, simple relief valves are not very expensive.

You also need to take care in insuring that the size of your relief is adequate for keeping the pressure below the desired set point.