Seal rotating parts/reduce friction (hand-tight thread)

[thereddevil]

I am working on a design where I have a block which has a cap assembly fitted to it - see images attached.

The cap assembly is constructed of:
- Main body with two holes through (flowing water in and out)
- Locking ring with a threaded outside, grooves for tightening by hand and two o-rings (face seal and rod seal)
- Dowel pin for alignment (important that the flowing water comes in and out in particular orientation)

The block has the following features:
- Threaded entry for locking ring
- Alignment slot for dowel pin to sit in
- Flat face for face seal to sit against

It has been designed so that as the cap assembly is dropped into the block, the dowel pin must first be aligned to the slot in the block before you are able to screw the locking ring. Materials:
all to be Delrin/Acetal at present. Block will be fixed as Delrin due to other requirements, but cap's main body/locking ring could be other plastics. The pressure inside will be a minimal negative pressure (i.e. water is being pulled out of the space under the cap rather than pushed through). Locking ring outside diameter is Ø40mm for an indication of scale.

Now the questions...

The o-ring that sits between the cap's main body and locking ring must allow the two parts to rotate relative to each other. Previous experience from colleagues of mine show that there can be a significant amount of friction here with a nitrile o-ring as well as acetal on acetal mating threaded parts. Any lubrication of the o-rings will quickly be removed through use and cannot be relied upon.
- What can be done to reduce this friction?
- O-ring material change to say PTFE/Nylon?
- Rotary seals?
- What materials could the cap use to prevent binding to the block?

Thanks in advance!

 

[CWatters]

Could you split the part with the o rings from the threaded part so that the part with the o rings doesn't have to rotate? Eg it's just pressed in by the threaded part.

 

[thereddevil]

Could you split the part with the o rings from the threaded part so that the part with the o rings doesn't have to rotate? Eg it's just pressed in by the threaded part.

Reference https://www.physicsforums.com/threads/seal-rotating-parts-reduce-friction-hand-tight-thread.952106/

I would like to have the entire assembly captive so that there's no chance of losing the parts (why there's a shoulder and a dowel pin either side of the threaded part). The centre piece must align with the block so that can't be threaded.

You've sparked one idea: move the piston o-ring from the threaded part to the centre piece. The threaded part could then be made of a low-friction, fine surface finish metal (i.e. anodised aluminium). This, I feel, would dramatically reduce the friction since rubber on metal would be less than rubber on Acetal (maybe?).

I'd be keen to find out more about the option of moving to a PTFE/nylon o-ring - do you have any experience of this?

 

[Dullard]

Regarding PTFE O-rings:

My experience (as a poor, dumb EE who occasionally gets stuck doing some plumbing design) is:

It is really important to remember that PTFE is not an elastomer. In an application like this (where you're counting on the the O-ring to maintain the initial compressive force) you'll be disappointed. Over time, the PTFE will 'deform' to fill your O-ring groove and there will be very little radial sealing force. PTFE-coated elastomer O-rings might be an option?

 

[CWatters]

You've sparked one idea: move the piston o-ring from the threaded part to the centre piece.

Yes that would work as the center piece doesn't rotate.

I think you can move both rings or even eliminate one ring.

 

[thereddevil]

It is really important to remember that PTFE is not an elastomer. In an application like this (where you're counting on the the O-ring to maintain the initial compressive force) you'll be disappointed. Over time, the PTFE will 'deform' to fill your O-ring groove and there will be very little radial sealing force. PTFE-coated elastomer O-rings might be an option?/

That is a very good point with the deformation! I've no experience with PTFE coated o-rings - could be worth testing.

 

[thereddevil]

I think you can move both rings or even eliminate one ring.

Could you explain this point further?

 

[CWatters]

I was thinking of something like this but it's an incomplete idea. As drawn below the outer ring isn't captive to the inner. This can be solved several ways..

1) Using a screw thread. The outer ring would be screwed onto the inner until the threads become unmated. It would then look and feel captive but could in fact be disassembled by getting the threads to mate again. I've seen this done before somewhere but I just can't think of an example. If that's not clear I'll try and make a drawing to show it.

2) Length of chain. Common on dust caps for military electrical connectors.

3) Fitting a retaining disc to the top after assembly.

 

[JBA]

Taking a variation from the original design and CWatters modification, what about cutting off the threaded bushing at the bottom of thread to make it as short as possible to move his busing shoulder up in order to give the bottom bushing sufficient length to allow placing the non-rotation pin and body slot in the bottom most bore below his face sealing shoulder.

 

[Tom.G]

For volume production, instead of using a pin for orientation you could make the body hole 'D' shaped for instance to ensure correct orientation of the orifice insert.

For a captive assembly a slight modification of @CWatters approach would be the method used on some medicine (pill) bottles. i.e. the cap is held onto the body with a ramp in both pieces. To remove a medicine bottle cap you must line up the cap with a notch in the ramp on the bottle. Without that notch, the cap is secured.

edit: With proper geometry, the body hole and orifice insert could also be made rectangular.

Cheers,
Tom