Valve leakage

[Deleted]

John,
Please could you forward me a copy of the photo of the PTFE rings that Doug Hewson produces. I would like to compare them with my own modest effort.

Andy
I agree that making the bobbin the same size as the liner would not be a good idea. You would have no way of taking up wear other than a new bobbin. I seem to remember in the Ford Anglia days that you fitted oversize rings to deal with wear in the bore which seems logical and fairly straightforward. Not quite sure why Paul self-destructed because your e-mail made sense.

Regards

Jim

[baggo]

I'm still experimenting with the valves for Helen Longish to find the best material. I gave up with the solid bobbins made from Peek HPV as it still expands far too much (as suggested by Pondok). To get a good fit at working temperature, the bobbins need to be a loose fit when cold which means they leak like a sieve until thoroughly hot.

As an interim measure I turned up some good fitting bobbins from stainless which seemed to work well for a while. However, recently these have been tightening up and seizing. I thought it was the lubricator at first but not so. On taking the stainless valves out, they were badly scored and obviously had been running very tight. They were free enough when cold however and although the expansion should have more or less matched the gunmetal valve chests, I think the bobbins run much hotter than the valve chest and so expand more, causing the problems.

I had been toying with the idea of trying a stainless bobbin fitted with a thin PTFE sleeve and gave it a go a couple of days ago.

I don't remember reading Andy's (Pondok) description of his valves until after I had made one of my own but I may have done and perhaps subconsciously copied his idea.

As I am short of time (got a rally on Sunday!) I adapted the existing stainless bobbin by reducing the diameter of the valve heads and just leaving a thin bit at each end full diameter to support the PTFE sleeve. The PTFE sleeves were just plain turning jobs from rod. They were drilled out to give a clearance of about 0.01" on the centre of the bobbin, turned to a push fit in the valve bore, parted off and then faced off to a push fit between the ends of the bobbin. They were then slit with a sharp knife so that they could be fitted onto the bobbin. As the sleeves are very long compared to a normal valve head, I turned a few V grooves in the PTFE as well for good measure to hold a bit of oil. Incidentally, the sleeves are only 0.045" thick as I wanted to keep them as thin as possible so they would compress easily as they expanded with the heat.

Pic of a finished valve:



To begin with, I just changed the valve for the middle cylinder as this runs hotter than the outside ones. A test run on the rolling road showed that, although the valve did tighten up, the loco still ran well with no sign of seizure. I then took the valve out which still looked like it did before fitting. The only thing I noticed was that the diameter of the PTFE sleeve had shrunk slightly and it was no longer a push fit in the bore. I believe that PTFE does suffer from permanent deformation when heated so this may have been the cause. Perhaps it has shrunk itself to the correct fit?

Anyway, I then altered one of the outside valves and this proved ok on another test run. I'm now off to tackle the third valve ready for a proper run at the rally on Sunday.

I noticed on the test run with two PTFE valves that the leakage past the valves was much reduced and was now probably due to the one remaining plain valve. Also the starting and running seems much smoother. I also noticed that the loco would now run in almost mid gear with very little regulator which it wouldn't do before.

If the run on Sunday proves the valves to be ok, I'll make up some new longer bobbins so that the PTFE sleeves are the full width as in Andy's design. At the moment they are too short and the valve events will be a bit out.

John

[Deleted]

Doug Hewson has kindly agreed for me to post the photo that he sent me of his Winson Britannia PTFE valve piston rings:



As you can see, they look very similar to Baggo's. Doug tells me that they seal perfectly and the loco just glides away without a sound. He found the optimum fit by experiment and says that they need to slide in by hand without being forced.

Kind regards,

John

[baggo]

Just come back from the rally at Little Hay and the new valves worked perfectly. Ran for over an hour with no signs of any seizing. No leakage at all and the loco runs very sweetly with the regulator only just cracked open. Starting was very smooth but very difficult to stop the wheels slipping on the greasy aluminium track (dreadful stuff!). The steam consumption seems to have reduced as well.

Hopefully problems solved ;D

John

[Deleted]

Further to the previous pictures I am attaching pictures of my first attempt at the valve rings for my 7&1/4" Britannia. Also included is the original Winson brass ring. I hope to be able to test run on air within the next couple of weeks.
Regards
Jim

[pondok]

Hi chaps,

Sorry to hear Paul is gone, he's literally a pro.

John/ Andy - there's not much doubt that the valve events should be determined not by the rings alone but by the entire length of the bobbin end.

As you point out John, the rings alone are likely not even as wide as the ports!

My PTFE rings, as I described, replace the entire bobbin end, held in place by the coned "cheeks", as you pointed out John.

It's always worth the extra trouble doing some measuring of the valve spindle movement imparted by the combination lever with the engine in mid-gear. I have tried and tested this very simple formula 3 times and it's worked every time for me: bobbin end thickness equals port width plus half the total valve spindle travel (stroke) with the engine in exactly mid-gear.

Then the distance between the outer edges of the bobbin ends should be the same as the outer edges of the ports. Only sight/timing holes will then allow you to be 100% sure the bobbin is in the right position on the valve spindle, but you can also just use trial and error. Can't make out if you have sighting holes on your ID pic, John. You can also position the valve by working out where it should be on the drawings in relation to the end of the valve sleeve at, say, front dead centre.

The ones in my SAR loco work great, I was able to notch back to about 30% cut-off when running at the Guildford rally, and there is zero leakage past the valves, as you can hear from the crisp exhaust in this clip on youtube:

www.youtube.com/watch?v=HbPJ4g-LwTk

let us know John when the engine steams up!

all the best

Andy

[baggo]

Hi Andy,

Have you noticed if the PTFE shrinks slightly after the first run? I took one of the valves out again to have a look and it had shrunk in diameter slightly, although the valves still seal perfectly under steam. Perhaps it's because I have used very thin sleeves which are only 0.04" thick?

I'm sold on the PTFE idea and will be using it on all future piston valved locos! Definitely the way to go.

John

[pondok]

Hi John (Baggo),

The valves havent been out since first fitting;-) but i can imagine you're right about them resizing. Mine are about 3/16" thick though.

By the way, they're also brilliant as piston rings, giving an unbeatable seal and doing away with issues of cast iron rings sticking in the ring groove or graphite yarn compressing into the ring groove.

happy running, they've never let me down!

cheers

[brozier]

Hi,

The ME article is called "Plastic Piston Valves" by John Mercer it's in 4th December 1987 issue page 664.

He fitted his valves to size by heating the cylinder assembly to
340 F (Gas Mark ?) if the valve can be moved at this temp it's right if not skim a bit off and try again!

He was using a "filled grade" for PTFE called Florosint....

Cheers
Bryan