"O" Ring Unconventional Use

Are you moving accross the port with the O-ring? Looks like this in the picture. If so there is a chance of failure of o-ring due to extrusion and shear.

If the rotating is in the pane the o-ring is installed that can be done. Back up rings might be necessary.

Pressure below 14.7 psi is getting into the vaccuum stage. did you mean differential is 1 psi?

That makes it somewhere 15.7 psi on the other side.

I realise that dragging the "O" Ring across the port could cause problems, I'm hoping to radius the edge, to minimise this. We are talking gauge, <15.7PSI absolute. A small air leak is acceptable and maybe 2 rings could be used. I may plunge the hole twice tangential to the curve.

Tony

Why not use a plastic liner in the bore to turn the spindle in letting tight fit seal it off. Like a ball valve does.

The best solution would be to use a "washer" some low friction polymer and use the O-ring behind as spring and seal at same time. Even if you will round the edge the local change in compression for the ring will generate strains which will lead to cracks and failure. If the "washer" is thin it will deform and take the cylinder radius due to its form it will most probably not rotate around own axis.

Assuming this is a side view:

To give adequate compression at parts of the oring that are not shown (away and toward the viewer) the compression at the top and bottom of the oring (shown in cross section) will have to be relatively high. With the valve half open, the torque generated on the oring will tend to twist the oring out into the port. Perhaps it will catch, perhaps it will not. The less lubrication available, the more likely that it will catch.

I'd make the oring counterbore perhaps 12 mm deep and insert nylon tube (6mm OD 3mm ID) , with its outer end machined to match the valve bore. This piece would need to be keyed to prevent rotation (or careful assembly would no doubt work). The oring would be used at the end away from the valve bore to provide sealing force.

Gut feeling (purely) tells me that if the valve bore were 60mm, your existing system would work just fine. Maybe just build one slice (I gather there are six like this) and test it -- it could be just fine as is.

You might also go to a DIY plumbing store and observe (or buy if budget supports) several washerless faucets, some of which work on this principle. You'd get some ideas, I suspect.

Thank you, using an "O" Ring as a spring and seal is good and maybe 6mm nylon air tube would do, I think it would align itself without a key. I have restrictions with the diameter as I have an existing unit to modify and the port hole has to be at an angle, so the recess depth is restricted. The spool is shown. I'll see if I can pull apart an old mixer tap as well. I'm concerned about using a full sleeve as I think the friction would be too high for the motor to overcome. Tony

You should look at a Delta® single-handle faucet valve. It uses a perforated ball that is moved around by your faucet handle, in contact with spring-loaded rubber valve seals. Check the following link for a view of the parts:

http://www.amazon.com/Delta-Faucet-RP70-Ball-Assembly/dp/B000BQWBGY

The ball sits in a concave spherical socket, with the inlet pipes having the spring-loaded seats (look at the "repair kit" at the top) that let water into the ball when you slide it around. The water leaves through a third hole after mixing in the ball. Your local hardware store might be able to show you an example pulled apart.

So instead of having the seals on the inner, moving portion of your valve, you may want to consider putting them on the outside, fixed unit, inlaid into countersunk holes the way you show here. The faucet companies have figured it out for you - you may even be able to use the same spring-loaded seals they do. The seals are commonly available, and about the size you want.

Teflon will beat nylon hands down in this application. But if you are in a position to manufacture parts to very close tolerances, the bore and inner drum can be made of hardened and ground tool steel to fit close enough that no seal is needed, especially at such low pressures. When I first built a sealless moving air valve I was challenged to make the parts fit so perfectly, and thought I might "cheat" and use a fine sand paper (800 grit) to sand those last .0001" off to fit. WRONG! The ground surfaces must be kept pristine and true for the seal to be maintained. And these valves never wore out in millions of cycles in a high pressure air inflation system (300-600psi.) Needless to say I was very impressed as a young man with this engineers design and was careful to follow his instructions to the letter for the next 13 years that I had the pleasure to work with him. But if this is not an option, be sure to make the outer plastic seal area match it's mating surface as closely as possible, not only for improved sealing but to keep consistant support of the seal to glide smoothly across the openings. And minimal loading on the o-ring will surely be sufficient to seal at such low pressures also reducing wear.

I have seen the Delta Faucets use this technique.

However, I remember and old gas torch that used a tapered "plug" with a hole through it as the valve. It had no appreciable leakage at 10 PSI.

I think some of the old gas stop cocks on the lab benches were also just a tapered brass to brass plug with a hole through them.

It reminded me of the glass stop cocks for an assay pipette. the glass was ground to match each other using valve grinding compound. I think the same technique could be used to lap a metal tapered plug to its seat.

that would likely also require more torque to position than by the use of O-rings and tubing - IMHO.

That would depend on fit and finish, lubrication, and tension. An oilite bushing or Teflon parts would probable have very low friction.