Piston Rings

[terrier060]

To throw the cat amongst the pigeons I have always used nitrile 'O' rings on my pistons and my loco has done hundreds of miles without changing them. I am using them in the 'Terriers'. The only thing you have to do is to use the manufacturers dimensions, which allow the rings to 'float'. They fit exactly from side-to-side, but have clearance ID, and fit the bore exactly. Pressure does the sealing. As long as they are well lubricated, I have had no problems with friction. I use hydrostatic lubrication with a sight feed.
Ed

[Roger]

Feb 15, 2018 15:36:39 GMT terrier060 said:

To throw the cat amongst the pigeons I have always used nitrile 'O' rings on my pistons and my loco has done hundreds of miles without changing them. I am using them in the 'Terriers'. The only thing you have to do is to use the manufacturers dimensions, which allow the rings to 'float'. They fit exactly from side-to-side, but have clearance ID, and fit the bore exactly. Pressure does the sealing. As long as they are well lubricated, I have had no problems with friction. I use hydrostatic lubrication with a sight feed.
Ed

Hi Ed,
Now that's something I've never heard of being done. I've only ever heard of 'O' rings in a bore being used with a positive compression between the inner and outer and a clearance on the sides. I really like this idea, it seems about the simplest I've heard of so far. If PTFE rings are as elastic as Joan claims, this ought to work with PTFE 'O' rings too. The only concession you would need to make would be to make the piston in two pieces because you wouldn't be able to stretch a PTFE 'O' ring over the outside without permanently deforming it.

Am I right in thinking that the locomotive isn't superheated though? That would explain why you can use Nitrile rings since the temperature is unlikely to get very high. They're supposed to be good up to 120C, and if you have wet steam at 90PSI that's only 160C at the regulator. By the time the pressure has dropped and it's reached the cylinders, I don't suppose it's anywhere near that.
I don't think Nitrile rings would work with Superheated steam, but Silicone rings would take that up to 300C which ought to cope with most things as long as the rings don't want to adhere to the bore.

[joanlluch]

Hi Roger,

Before anybody gets me wrong let me say that I always meant to use the physics definition of "elasticity", which is "the ability of a body to resist a distorting influence and to return to its original size and shape when that influence or force is removed. Solid objects will deform when adequate forces are applied on them. If the material is elastic, the object will return to its initial shape and size when these forces are removed". This is not equivalent to saying that PTFE is or behaves like rubber. So I think Ed's approach would not work well with PTFE o-rings because it lacks the ring compression that an initial interference fit would provide. Without that, any small pressure would not be enough to provide the required seal and leaks would occur until pressure is high enough.

Silicone rings are generally not recommended for hot steam. It's not a matter of temperature alone. As strange as it may seem, steam is a very chemically active substance and most kind of rubbers do not like it very much. Look at that table:

mykin.com/rubber-chemical-resistance-chart-6

Silicone o-rings are relatively cheap and can stand high temperatures in chemically unaggressive environments, but they are not a panacea. They are not generally good for dynamic applications because they tend to wear prematurely, and they do not like some oils (steam oil?). At low temperatures EPDM is better for steam but not oil. Nitrile is better in the presence of oil. At higher steam temperatures properly Peroxide cured Viton will outperform Silicone, but unfortunately most cheaply available viton o-rings are manufactured with poorer (faster/cheaper) curing procedures. If nothing is mentioned then chances are that Viton is not the right kind for steam. When temperatures get even higher only Kalrez will do it. Still, temperature conditions are probably not that extreme anyway in the cylinder of a steam loco, which may explain why a number of different solutions seem to work just fine. It's a pity that PTFE is so good at chemical resistance but it's not rubber like. Well, I guess that's why someone invented Kalrez (if only it would cost like PTFE...)

Joan

[rwilliams]

Hi Ed,

It's useful to have that operating data from the larger scales, in the Garden railway world it's almost entirely sealed using Nitrile O rings, again as you describe floating with the ID gap between piston and O ring.
These are used on locos up to about 70PSI (plus superheat) in home builds, but commercially I can't think of a manufacturer who doesn't use them, and most applications are either 40psi superheated or 60psi wet.
I have no idea how many miles one of my locos has done, but over 20 years of use the cylinders are still tight, even if the valve gear and bushings have had it!
As I'm a modeller turning model engineer, I'm not sure how (if it does) the effects scale up?

[rrmrd66]

Morning everybody.

I think we have flogged this one to death. Well, until the next time!

All commercial seal manufacturers employ copious amounts of smoke and mirrors where claims of their ptfe based compounds are made. Caveat emptor.

Perhaps, if you have time, you might like to read the attached from the originators themselves, DuPont. It goes into a lot of detail about the properties of the material and the addition of fillers.

www.rjchase.com/ptfe_handbook.pdf

Back to the Hunslet!

Cheers

Malcolm

[robmort]

Feb 15, 2018 15:36:39 GMT terrier060 said:

To throw the cat amongst the pigeons I have always used nitrile 'O' rings on my pistons and my loco has done hundreds of miles without changing them. I am using them in the 'Terriers'. The only thing you have to do is to use the manufacturers dimensions, which allow the rings to 'float'. They fit exactly from side-to-side, but have clearance ID, and fit the bore exactly. Pressure does the sealing. As long as they are well lubricated, I have had no problems with friction. I use hydrostatic lubrication with a sight feed.
Ed

Unless you have different manufacturer's data, for this type of "floating" o-ring piston seal a small side-to-side clearance is still recommended because there is otherwise no easy way for the ring to expand or contract radially as air would be trapped beneath the ring, and to allow for ring cross-sectional expansion with temperature and steam absorption.

[suctionhose]

I found this drawing today on an SKF site. Well established practice of PTFE ring and nitrile energising element underneath. The pertinent detail is the "step cut joint".
I believe this negates the double ring of Roger's design (Sorry Roger!)



The "cat among the pigeons" design is interesting. Mainly because I can't see why it would work with any reliability under genuine working conditions! If it does then one wonders whether drag and piston movement combine to position the oring where pressure is applied to only 270 deg of surface thus squeezing it into a seal at the critical moment.

[dhamblin]

Out of curiosity what are the other two rings shown on that diagram?

Regards,

Dan

[joanlluch]

Feb 20, 2018 6:39:06 GMT suctionhose said:

I found this drawing today on an SKF site. Well established practice of PTFE ring and nitrile energising element underneath. The pertinent detail is the "step cut joint"
...

Hi Ross, That's an interesting finding and as a concept I suppose it may be applied to our needs, but that particular product from skf is not made of PTFE but Polyamide based (a.k.a Nylon), which is why I guess it must be split. On the other hand we can't use it as it is because it will not stand temperatures above 110 ºC. The choice of material and the availability of imperial sized only diameters suggests to me that this is an old product possibly produced as spare parts for existing equipment only, but others may give more insight into this.

If you look at the list of piston seal products from SKF you will note that the ones based in PTFE are NOT split, such as the GH series ones. SKF Piston Seals. Interestingly, none of the ones in the SKF range seem to be designed for relatively high temperatures, but this is possibly only because of their choice of nitrile rubber for the o-ring underneath.

As a comparison, I looked at the list of piston sealing products from Trelleborg, which is the world leader in such applications, Trelleborg piston seals. The following link is to a page with actual pictures for metric sized designs: Trelleborg piston seal catalog metric. and imperial ones Trelleborg piston seals catalog imperial.

My understanding is that the Trelleborg equivalent to the one you mentioned from SKF is the "Turcon Glyd Ring C", which again is only made in imperial sizes.




Note that none of the currently produced seals from Trelleborg are split any longer, because they tend to be made either in PTFE or polyurethate. This includes the supposed equivalent to the SKF one. In the case of Trelleborg, this particular product appears to be a niche one as it is not even listed in their main page. You need to look deep into their web site to find it. Again possibly only produced for compatibility purposes. I think this has been widely replaced by their "Turcon Glyd ring" and "Turcon Glyd Ring T" series.

Joan

[steamer5]

Hi Don,
Called rider rings, take the load of the piston & allow the rings to do the sealing with minimal load.
The gas compressor at work has these, there is another version were the ends of the piston have valves installed that then pressurize the the interior of the piston at the end of each stroke & the gas then escapes under the rider rings, helps keep everything centered. The piston dia is 180mm & about 750 mm long so not quite our size!

Cheers Kerrin

[joanlluch]

Feb 20, 2018 7:49:24 GMT dhamblin said:

Out of curiosity what are the other two rings shown on that diagram?

Regards,

Dan

Hi Dan, 

They are called "hydraulic wear rings" or "Hydraulic sliding rings" or "rider rings". They are usually made of hard plastic materials and are split to account for expansion, but they are not usually designed to stand high temperatures or steam.

Visit www.seals-shop.com/eu/, then click on "Wear Rings" for some available options.

Or look here for technical specs

I don't think they are needed for us, because they are used in high pressure applications to prevent metal to metal contact and where an internal guide of the piston is required due to transverse forces being applied externally to the pistons.

Joan

[suctionhose]

Joan,
Agreed the SKF page related to hydraulic pistons and yes, the materials were as you state.

That said, conceptually it seemed valid for our low pressure steam application.

Did you state earlier that the proof is in the actual use or something along those lines?

[joanlluch]

Feb 20, 2018 8:51:39 GMT suctionhose said:

Joan,
Agreed the SKF page related to hydraulic pistons and yes, the materials were as you state.

That said, conceptually it seemed valid for our low pressure steam application.

Did you state earlier that the proof is in the actual use or something along those lines?

Hi Ross,

Yes, I said commercial hydraulic seals are used successfully in many steam locomotives and locomotive steam pumps at least in Catalonia and Spain.

However, these locos are all non-superheated, so it's not clear or proved that they would also work on superheated steam also. In fact nobody here or in any other forum or FB group that I have joined was able to assert that PTFE based rings (any kind) will work for many years in superheated steam locos.

Joan

[terrier060]

Hi Ross

I don't really have an answer - only experience. My loco has done many miles of running on the Soton track and is now very old and the last time it was used there was no blow-by. As far as I understand 'O' rings, the seal is made by the ring pushing against the bore and the side of the groove in the piston farthest from the high-pressure side. As I have stated earlier, the rings were slightly worn flat, but still sealed perfectly. I did wonder whether there would be frictional drag, but never noticed any. Probably the most friction would have been at slow speeds in full gear with a heavy load. As I have said previously, I have a good hydrostatic lubrication system on it with sight feed and atomiser. I shall repeat all this on the 'Terriers' but may leave off the atomiser unless I can hide it. The 'Terriers' are not superheated unlike my other tank engine.

Sometimes one can theorise on these matters, but in the end it is whether it works in practice that is important.

Ed

Ed

[suctionhose]

[Source] terrier060 said:

Sometimes one can theorise on these matters, but in the end it is whether it works in practice that is important.

Youre right about that of course!

For over 10 years I used o rings to seal against the valve chest cover for balanced slide valves. It was a concept someone introduced me to when other ideas weren't working out.

In fact, it had features in common with your piston seal because the ring was free to move about and pressure on 270 deg of circumference wedged the ring into the gap making the seal.

They worked perfectly in terms of sealing, frequently with steam chest pressures up to 90 psi and radiant superheat. It was only long after installing them that I did some testing on the friction side of things - which was the reason for valve balancing in the first place - and found the load on the valve gear actually INCREASED due to the rubber/metal contact.

A metal to metal seal was a much lower friction. Scrapped the orings in 3 engines and started playing with PTFE (encountering all the problems we discuss on here!)

Thanks for sharing your experience. There's no single answer to anything. Just what experience leads one to believe is best!