O-ring for piston ring

[sniffipn]

Been trying O rings, quad rings really. Having some trouble getting reliable compression.

Seems there is a difference of opinion on how much pressure the o-ring should exert on the cylinder wall.

Tom Stewart calls for .005" clearance between the back wall of the piston and the ring. This is about what others, having success have told me.

Others (Pete Twissell) call for 15% compression of the o-ring. Can both be correct or is there an ideal?

I'm now going to try a regular o-ring as I suspect the quad rings are twisting in the oversize groove and loosing compression.

What a great hobby!!!

John in Wyo

axial (along length of piston) clearance. for a 0.070" seal, the axial length of groove is typically 0.089/0.099"
Pete refers to radial (across section of seal) squeeze. for a 0.070" seal, typically 12% to 15% 0.060/0.062" housing radial width. different manufacturers quote slightly different numbers.

dont worry about minor differences, its helpful to understand the principles - as extract from a leaflet above.

both radial squeeze and axial clearance are required to optimise seal performance

Does Tom refer to the diametral clearance between piston and bore? for a 0.070" seal, typically 0.005"

quad ring less likely to roll and twist, than an o-ring - of the same section width
David

 

[Vietti]

David,
Thanks for the information. Did I misread or misinterpret Tom's statement when he said:

The piston ring groove is cut to provide 0.005" clearance top and bottom and 0.005" clearance on the back wall of the groove.

I'm about to remake the pistons and after 4 sets of them I'd like to get it right. The engine is Bob Shores Silver Bullet. 3/4" bore and stroke. I wonder if other builders might comment on their experience with this engine?

I seem to remember when a good friend, now deceased, Randall Cox, designed the Hoglet he used o-rings. I will have to dig out his build article and see what he says. I've seen many Hoglets at the various shows here in the US and most all ran really well.

I"m not trying to be argumentative in any way, I"d just like to get it right and move on!

John in Wyo

 

[sniffipn]

David,
Thanks for the information. Did I misread or misinterpret Tom's statement when he said:

The piston ring groove is cut to provide 0.005" clearance top and bottom and 0.005" clearance on the back wall of the groove.

I'm about to remake the pistons and after 4 sets of them I'd like to get it right. The engine is Bob Shores Silver Bullet. 3/4" bore and stroke. I wonder if other builders might comment on their experience with this engine?

I seem to remember when a good friend, now deceased, Randall Cox, designed the Hoglet he used o-rings. I will have to dig out his build article and see what he says. I've seen many Hoglets at the various shows here in the US and most all ran really well.

I"m not trying to be argumentative in any way, I"d just like to get it right and move on!

John in Wyo

0.005" clearance on what ...

1. the piston outside diameter - in comparison with the cylinder bore, but Toms explanation is 0.002" clearance, between piston and bore
or
2. axial length of groove, in comparison with the O-ring diameter section, for example?

top?
bottom?
back?

to avoid misunderstandings, Petes suggestion of downloading/looking online at a catalogue, is sensible ...

as would providing a simple sketch, showing dimensions/geometry that you intend machining.

looking at the catalogue will also help you to select an o-ring, with the appropriate inside diameter. the o-ring will typically have an inside diameter, that is a little smaller, than the root diameter of the groove (in the piston).

 

[Peter Twissell]

I would somewhat qualify my suggestion to use standard / catalogue O ring compression.
The standard compressions are intended for hydraulic and pneumatic applications.
I could well imagine that the application in an IC engine piston may be different.
My approach would be to start with 15% compression and reduce progressively as required to achieve the balance between adequate sealing and acceptable friction and wear.

 

[Engine maker]

I have many engines running with just off the shelf black Bunan "O" rings. From 3/4 inch up to a 2 " bore FarmBoy I just built. No problem with the "O" rings. Last forever. Most people try to make it too tight and it binds and doesn't float. I use the sizing that Jerry Howell uses on his regular size Farmbow with a 1"bore. Piston .998" dia with a .996' dia above the ring towards the piston face. Them for a 3/32"(.094") "O" ring I plunge cut the groove .110" deep by .105" wide. Should work fine on a 3/4" piston.

Remember the actual OD of a 1" "0" ring is 1.005" and for a 3/4 "O" ring it's .755". Keep the bore as smooth as you can get, keep it cool, and keep it lubed and you won't have any trouble.

Jim G

 

[Vietti]

Thanks for all the reply's. They open up some options.

I'll remake pistons with 15% compression and then back off if need be as suggested by Pete. Also I've been using 1/16" o-rings, perhaps try the bigger 3/32 ring. Also will forgo quad rings till more comfortable with the technology. By the way, reading up on the Hoglet Randall specifies a relaxed, no compression fit for the o-ring.

axial length of groove, in comparison with the O-ring diameter section, for example?

is axial length of groove same as groove depth?

Welcome additional comments and will up date with findings in the future

John in Wyo

 

[MikeG]

Hi John
Here's my 2 cents (probably what it's worth) I'm currently making a .375 bore 4cyl engine. I'm using one 1/16" quad O ring per cylinder. I went out to the shop and measured the piston ring groove. The groove allows .005 total clearance for both ring thickness and ring width With my thumb closing the top of the cylinder, I can pull the piston almost out of the cylinder and then let go and it will snap back. I assume that when the time comes that will give me enough compression. By the way putting a quad
O ring into the piston groove is VERY frustrating because it ALWAYS wants to twist!

Hope to see you again at 2021 Rapid City show.

Regards
Mike

 

[PoppyAnn]

Are you not bothered about the o ring melting and sticking too the cylinder walls?

 

[MikeG]

No, viton O rings can take the heat of ignition and the presence of petroleum without "sticking". At least that's how understand it

Mike

 

[Vietti]

By way of an update I have now installed o rings on two of my recent engines. In both cases the compression improved as did the oiling problems. So I'm sold.
I just searched o rings on the forum and read most of the threads and found interesting, contrasting ideas on how they should be fitted. Most everyone agrees Viton rings are the best.

There seems to be two schools of thought concerning fitting the o ring. One school recommends fitting the ring loosely while the other school suggests some compression of the ring. I suspect both will work. I subscribe to the loose fitting theory. I notice that some prominent builders/designers like Tom Stewart, Randall Cox and Jerry Howell used the loose fit approach with a .002" piston clearance below the ring and .005" of clearance above the ring. I think what happens is when the compression pressure hits the ring as the piston moves up for compression, the ring expands as the pressure forces the ring against the cylinder wall and down in the groove. Once the compression pressure is relieved the ring shrinks back to original size and there is then minimal drag. Drag seems to be a problem with tight fitting o rings resulting in premature o ring wear and shorter coast time on hit and miss engines.

In the past race engine builder drilled a series of small holes down thru the piston top that intersected the back of the top ring groove to help force a better ring seal. Interestingly Total Seal has introduced a new compression ring that has shallow scallops ground on the top surface of the ring with the intention of getting more compression pressure behind the ring and creating a better seal. Would it work on a small engine?

For me the jury is still out on Quad rings and X rings. I tried them and my problem was they swelled up after a couple days to where they fit so tight they rolled in the groove. These rings started out about .071" in cross section and after swelling were .078". They were not viton. I believe everyone is correct and viton rings are the way to go. I did notice one builder who reports his o rings shrank.

Lastly I've heard and read that two o rings on a piston will either not work or the compression is excessive. I have no experience with this but am going to try it at some point. What got me thinking about this was there is a YouTube video of some one overhauling or at least disassembling one of those inexpensive Chinese hit and miss engines. There in all its glory were two rather thin o rings on the piston that looked to be tightly fitted but hard to say. He called them fluorocarbon or some such but I think it is another version of viton.

It is interesting that rings are such a vital part of our small gas engines and yet there is so much divergence of opinion re how to make, and fit cast iron rings, how to fit o rings and which is superior. I suppose that this divergence of opinion is what helps keep the hobby interesting.

 

[TSutrina]

The O-ring in the space shuttle booster rockets were Viton with packing materials trying to prevent the hot gas from touching the O-ring. Leakage around the O-ring resulted in burning of the O-ring and failure. Internal combustion temperature and solid propellent temperatures are similar.
The only reason that the O-ring doesn't burn is that the gas is cooled in the gap between the piston and the cylinder. An oil film on the cylinder wall adds to the cooling by evaporating. So long as the amount of oxygen is low the evaporation cooling is significantly greater then the burning of the oil. This didn't happen in the case of the shuttle boosters.
O-ring is hydroplanes on the oil on the cylinder wall. So the pressure of the O-ring against the wall determines, the pressure plus the pre squeeze, the thickness of the film available to be evaporated.
One thing to consider is backup rings instead of X or square profile which has two a longer face against the cylinder wall thus the final oil fill for the same pressure will be thinner. The back up ring will help prevent high pressure from extruding the O-ring into the gap and will add more resistance to decrease gas leakage.

 

[Brian Rupnow]

The figures I gave for piston ring groove width and depth are correct. I have never had any issue with compression on my engines running a Viton ring. The largest bore size I have used Viton rings in is 1" bore, the smallest is 1/2" bore.

 

[Kasey]

No, viton O rings can take the heat of ignition and the presence of petroleum without "sticking". At least that's how understand it

Mike

Would "tight" o rings work ok on a 2 stroke engine, or just chop up?

 

[Brian Rupnow]

In a 2 cycle engine the viton style rings get sliced and diced as the piston passes the ports in the cylinder wall.

 

[wespete66]

Some have mentioned using two "soft" seals on a piston. The problem with doing this is pressure trapping. Trapped air/oil in the space between the seals will begin to create pressure in that space, due to thermal expansion or from the seal scraping the oil film from the cylinder wall and trapping it in that space. As the piston cycles that pressure increases, causing increased seal force on the cylinder wall. It is possible that this can cause forces great enough to actually stop piston movement. It can also cause premature seal wear or failure.
It is also possible for this to happen between a dynamic o-ring and its backup washer (if one were to be used). There are special backup washers made to remedy this situation.
(edit: soft seals only, doesn't apply to cast iron rings)

 

[coulsea]

Some have mentioned using two seals on a piston. The problem with doing this is pressure trapping. Trapped air/oil in the space between the seals will begin to create pressure in that space, due to thermal expansion or from the seal scraping the oil film from the cylinder wall and trapping it in that space. As the piston cycles that pressure increases, causing increased seal force on the cylinder wall. It is possible that this can cause forces great enough to actually stop piston movement. It can also cause premature seal wear or failure.
It is also possible for this to happen between a dynamic o-ring and its backup washer (if one were to be used). There are special backup washers made to remedy this situation.

The first time I used O rings I thought that two must be better than 1, it would lock solid after a few turns.

 

[almega]

Should the wall of the cylinder be honed smooth or should there be a cross-hatch pattern if using Viton o-rings?

I have read that when using two o-rings, you need to provide a small vent hole which vents the space between the rings out the bottom of the piston to prevent pressure build up between the rings. the lower ring only provides for alignment in the cylinder and does nothing to enhance compression.

 

[Ken I]

With regard to using more than one "O" ring.
Firstly an "O" ring behaves like a liquid - it is displaced by pressure and the internal pressure of the "O" ring is always the applied pressure plus the preload.
It simply cannot leak - no matter how much pressure you apply - other than due to poor surface finish or excessive clearance.
This image from the Parker-Hannifin manual :-

I have used "O" rings to 16000 psi in hydraulic burst testing of tubing - by filling the test piece with oil and using two "O" ringed plungers, squeeze in a tensile test machine (applyling tonnes) until the tube swells & splits - great fun. Used zero clearance, very slow speed - but never lost an "O" ring.

However it will try and "leak" through the clearance (behaving like a liquid itself) and the combination of sufficient pressure and too much clearance will allow it to "extrude" this leads to seal failure or "nibbling". Nibbling also occurs because of a sharp edge on the "O" ring groove - hence the need for radii.
A bad case of extrusion / nibbling / melting :-

When you use two "O" rings the pressure between them builds and deflects the rings outwards - called a "pressure trap" due to dynamic forces (boundary layer and film strength etc.) this pressure can be incredibly large.
The pressure building in the "O" rings causes them to bind with the cylinder walls (as pointed out by Coulsea - above) Since the active internal "O" ring pressure keeps on building so does the pressure of any material (oil, air etc.) leaking past via surface finish or film strength into the "trapped" volume - its a vicious circle and absolutely can only end in failure - typically very quickly.
If you want to use two "O" rings for guidance or stability, then Parker-Hannifin recommend splitting one of the "O" rings (now referred to as a wiper) or by venting the gap between the "O" rings.

For an IC engine use piston rings for guidance and one Viton "O" ring to seal.
What should also be apparent from the fluid dynamic nature of an "O" ring, the higher the pressure, the lower the tolerable velocity and importance of lubrication / cooling if burning / melting is to be avoided.
Surface finish should be a very fine cross hatch (A'La honing) in order to carry oil through the seal on film strength. Mirror finished surfaces seal well - which is fine for static applications - but will "wipe" clean and cause lack of lubrication to the "O" ring and consequent "melting" type failures in dynamic applications.

Regards, Ken

 

[Vietti]

Interesting discussion. The diagrams shown by Ken show the o ring constrained by both the piston and the cylinder wall. I kind of think they work best when there is some clearance behind the o ring so the pressure of compression forces the ring out to seal and relaxes on the other 3 strokes.

I suspect the factory literature is focused on the sealing of a piston/rod in a cylinder and the pressures are from above and maybe below and the 0 ring is expected to do its job by always providing a nearly perfect seal, speeds are relatively low so drag is not an issue. We desire an o ring in our application to seal on one stroke and then relax, speeds can be quite high.
I am not certain of any of this and only started using 0 rings in the last few months so no expert. I can say that I now have real world experience with loosely fitting the o ring which works well. Next time out I'll try the tight fitting scheme.

I understand the logic behind the notion that two 0 rings will not work without modification. I will attempt to attach a link to the you tube video of two rings on a small Chinese hit and miss. If anyone has one of those engines maybe they can chime in and describe how they are fitted.https://www.youtube.com/watch?v=0oYgB3FBqjc&t=167s

 

[Ken I]

Vietti - the link you showed (Tip when inserting a link use the link icon - 9th icon from the left - as I have done below).



Is for two "O" rings on the water jacket which are static - that you can get away with - but not for a dynamically loaded "O" rings. (Looks to me like he replaced two Nitrile Rubber rings with Viton replacements.)

The dual rings on the piston are a bad idea - in fact exactly what you shouldn't do. It's possible there is a vent hole between the rings but it doesn't show in the video.

You are also correct in that a lightly loaded "O" ring is best - running an "O" ring "tight" to overcome poor surface finish or fit up problems may solve the problem at the expense of running life. If it works with little preload, there is no reason to go for more.

Stick to the manufacturers recommendations for interference, grooves, radii, surface finishes etc.

However if you run minimum preload then you would also run minimum groove width - you want the width to be just wider than the "squeezed" "O" ring - more lateral play results in the "O" ring rolling or twisting unnecessarily - which in rapid cycling can fatigue the "O" ring or cause it to rotate (or twist) to a different position on forward and return strokes leading to a dual flat and point wear pattern which ultimately fails to seal.

As regards the "pressure trap" problem lets just imagine a small quantity of lubricating oil (hydraulic fluid, compressed gas - take your pick) is squeezed out from between the running surface and the "O" ring - it is ejected at the same pressure as the "O" rings' internal pressure plus the film or boundary layer pressure - into the "trap" raising the pressure of the trap which raises the internal pressure of the "O" ring - obviously a vicious circle which ratchets up the pressure in the trap with each cycle - ultimately extruding the "O" ring.

Regards, Ken