Mailing List Archive Mailing List Archive
Just adding my two cents to the discussion on designing O-ring joints: One consideration for using O-rings with non-metallic materials such as plastic battery cannisters is the compatibility of the O-ring compound with the plastic housing materials. We talk a lot about Viton o-rings as well as Buna-N (nitrile), butyl, ethylene propylene, etc. but what we should be including is the specific compound of the o-ring material. For example some compounds of viton and nitrile can cause crazing in polycarbonates (LEXAN). Crazing is a micro-fracturing in surface of the plastic due to a chemical reaction with o-ring compound, including the various plasticizers which add resiliency to the o-ring. Various lubricants can cause the same effect. One should consult with several of the design guides and call the o-ring manufacturers (Parker, National, Apple) to get advice on the specific compounds to use. Call any of them and they will send you a design guide (Parker Seals 606-269-2351). Fortunately acrylics, which most clear plastic cannisters are made out of, are very tolerant to non-agressive chemicals typically found in 0-rings and the diving environment. While I use a lubricant on all of my oceanographic instrument cannisters, I agree with George Irving on specific o-ring installations in dive gear. For example I do not lubricate the relatively thick o-ring on the top of my battery cannister; I simply wipe it clean and inspect for damage. Thicker o-rings tend to be tolerant to dirt and scratches on the sealing surfaces than thin o-rings. Although thinner o-rings have better strength properties, they require tighter tolerances and are more suseptable to damage and leakage due to dirt and scratches. If the tolerances in an o-ring connection are close, then a lubricant will assist assembly without damage to the seal such as rolling, shearing, abrasion, etc. While lubricants add some chemical protection to the o-ring, you should not depend on the lubricant to provide this protection. The o-ring material should be compatible with the environment. I have seen rotating pumps seize due to swelling of an etylene propylene (EP) shaft seal just from the absorption of hydrocarbon oil from seawater in harbors. Parker has done tests with EP in o-ring grooves subjected to oil and have measured pressures in the o-ring gland as high as 300,000psi! Imagine that in your tank. Rubber is nearly incompressible; if you squeeze it it will deform by getting laterally wider. If you restrict this lateral growth you will not be able to deform it much. So you need adequate reserve volume in the o-ring gland to permit deformation during sealing and any anticipated swelling. As mentioned, thinner o-rings require closer tolerances to effect a reliable seal. Thicker o-rings can be used with wider tolerances; however considerations must be made for extrusion into the gap. The standard gap should be around .003 to .004" for radial (barrel) seals for pressures up to about 3000psi (static and non-rotating or sliding). Above about 1500 to 2000 psi with these gaps you should consider a harder o-ring material (Shore A 90 v.s. the standard Shore A 70) to minimize extrusion. Backup rings, the little white teflon or nylon rings you see in tank valve stems or Scubapro Mk15 pistons) can be used in wider gaps at higher pressure to minimize or eliminate extrusion. I have used thick o-rings (apx.1/4" at several hundred psi with gaps approaching 0.010 to 0.015" with success. Each seal is designed for the intended application. Compression set is a term that describes how far the o-ring cross-section diameter "springs" back to its original un-deformed shape. Look at an old o-ring and it will be deformed => compression set. All o-rings will undergo compression set somewhat. Thicker o-rings tend to have more favorable compression set properties than thinner o-rings. In your design you should have at least a .007 to .008" squeeze on the o-ring. This is because almost all o-rings will have at least this much compression set. For example when you install an o-ring that squeezes only .007" when the two housing surfaces are joined (Radial or barrel seal) it will initially exert force on the sealing surfaces due to the elasticity of the o-ring. After a short while the material will relax and compression set will occur. If the o-ring takes a .007" compression set then there will be no elastic force exerted on the sealing surfaces from the o-ring. The pressure difference acting on the o-ring (outside to inside the battery cannister) will have to deform the o-ring to create the seal (this is how the o-ring seals anyways). Actually a static seal will typically seal with zero o-ring elastic force, but usually only reliably with a constantly applied pressure force. How much squeeze should your o-ring joint have? For static seals you should size your mating parts such that the o-ring cross-section diameter will be squeezed at least 13 to 15% with 30% being the maximum. Higher percentage will make the parts hard to assemble (30% will be hard to assemble). Dynamic sliding seals should see a maximum of about 15 to 16% with a minimum of about 7 to 8%. You need to realize gap between the mating parts will change when the housing sees a pressure loading (submergence). Sometimes the effects of pressure on the housing will aid in closing the gap. In other cases it may tend to open the gap. You must design your dimensions for the worst case (max. depth, widest gap, etc). Add structural analysis here. Obviously machined surfaces will offer the best control on gap tolerances. Acrylic housing bodies are usually not machined; the lids are sized to fit the body tubing, with or without deforming the body tubing into being round. I prefer to use an o-ring which is tolerant to large gaps rather than force the body into shape (for the relatively low pressures in battery cannisters). Machined is better though. Surface finish is important to o-rings in that if it is too rough it will damage the o-ring. If the design is not proper, the o-ring may migrate back and forth in the o-ring groove when pressure is applied and released. If the surface is rough, it will act just like sandpaper. Surface roughness should not be greater than 32 microinches for liquid systems, and about 16 microinches for pneumatic systems. O-rings are very versatile seals with which a little thought will go a long way to effect long term service. If you are interested in this technology then again call one of the o-ring manufacturers and they will send you a comprehensive design guide with lots of reading, tables, and design procedures. I did an o-ring design today for an oil well borehole instrument for 20,000psi. My tank pressure gauge doesn't even go that high! Thanks, Doug Chapman
Navigate by Author:
[Previous]
[Next]
[Author Search Index]
Navigate by Subject:
[Previous]
[Next]
[Subject Search Index]
[Send Reply] [Send Message with New Topic]
[Search Selection] [Mailing List Home] [Home]