|
Post by Doug on Mar 23, 2015 19:13:06 GMT
I made them a sliding fit in the bore then cut them diagonally, the theory is that the lateral movement in either direction will push the two surfaces together, it seems to be very effective I could tell the difference as soon as I moved off it felt so much more powerful. During driving I couldn't notch up any more, it ran fine almost in mid gear and with very little regulator open. its such a simple fix I am happy to continue with speedy now as it has exactly the same fit on the rings. That's a great result, it clearly works well. I'm assuming that the saw cut is just about the right amount to accommodate the expansion when it gets hot. If the gap wasn't quite enough, it might be able to cope with that, and if it was too much it would noticeably leak. Did you use a standard hacksaw to make the cut? Hi Roger no I used a stanly knife, it cut it very cleanly. Oddly I didn't leave any room for expansion so I am lucky in that the excess went somewhere. I guess worst case it extruded down the piston sides which will give a rather neat side effect of giving a better bearing face to run on. I will hopefully do better on the piston valves on speedy and have some space for expansion in the design.
|
|
jma1009
Elder Statesman
Posts: 5,901
|
Post by jma1009 on Mar 23, 2015 19:14:52 GMT
whilst i entirely agree that doug's PTFE valves appear ok, doug really needs to have a heavy load with the loco performing hard with the fire white hot and consequently the superheaters providing steam to the cylinders at a much higher temperature! a 1.5 hr test with just driver and plus a trolley with some bricks on is pretty lightweight stuff for a 'Butch'!
i didnt think much of the ME 'crankpin' article (referred to by joan) at the time. doug has very effectively explained why.
cheers, julian
|
|
|
Post by Roger on Mar 23, 2015 19:24:58 GMT
As an additional note, I found that PTFE is not working right for everyone. See this (piston valve attempt) www.modelengineeringwebsite.com/PTFE_valve_rings.html . In this case I think the problem could be that no clearance was allowed at the inner bore of the rings and that the rings were probably too deep to adapt to temperature expansion. [/quote] That's an interesting article and he clearly started off with the false assumption that the thermal expansion was insignificant. It does sound like the rings don't have internal clearance. He doesn't describe the shape of the ports, if they have long straight sections they could snag the PTFE rings which are pretty soft when hot. It seems to me that it's difficult to get the right initial fit for solid PTFE rings, but once it's right it would appear to last well from everything I've read.
|
|
|
Post by Roger on Mar 23, 2015 19:35:47 GMT
That's a great result, it clearly works well. I'm assuming that the saw cut is just about the right amount to accommodate the expansion when it gets hot. If the gap wasn't quite enough, it might be able to cope with that, and if it was too much it would noticeably leak. Did you use a standard hacksaw to make the cut? Hi Roger no I used a stanly knife, it cut it very cleanly. Oddly I didn't leave any room for expansion so I am lucky in that the excess went somewhere. I guess worst case it extruded down the piston sides which will give a rather neat side effect of giving a better bearing face to run on. I will hopefully do better on the piston valves on speedy and have some space for expansion in the design. I'm surprised that worked but pleased all the same. It's interesting that you noted such a change in performance over 'O' rings which one would assume to have sealed well. As an aside, I think this is where anecdotal evidence of people saying there were no leaks etc is very shaky evidence. I don't think it's possible to tell how much something leaks in a running locomotive unless it leaks badly. The lack of obvious leaks isn't evidence of no leaks.
|
|
|
Post by Doug on Mar 23, 2015 20:29:35 GMT
whilst i entirely agree that doug's PTFE valves appear ok, doug really needs to have a heavy load with the loco performing hard with the fire white hot and consequently the superheaters providing steam to the cylinders at a much higher temperature! a 1.5 hr test with just driver and plus a trolley with some bricks on is pretty lightweight stuff for a 'Butch'! i didnt think much of the ME 'crankpin' article (referred to by joan) at the time. doug has very effectively explained why. cheers, julian You have as usual have hit the nail on the head Julian I won't claim full success until I have done a steam day run with the proper load. The silicon orings worked well (although not this well) until I did a full load, which as you said increased the temp significantly. It really didn't get a workout at all even the towing session didn't really give it any trouble. My one good thought is though that I am now running a much more conventional setup as a lot of the "regular runners" at our club run with PTFE. another positive point is that I will never be a regular runner so what works for me won't necessarily work well for someone who wants to run most weekends. a good outcome though IMO.
|
|
|
Post by Doug on Mar 23, 2015 20:37:50 GMT
Hi Roger no I used a stanly knife, it cut it very cleanly. Oddly I didn't leave any room for expansion so I am lucky in that the excess went somewhere. I guess worst case it extruded down the piston sides which will give a rather neat side effect of giving a better bearing face to run on. I will hopefully do better on the piston valves on speedy and have some space for expansion in the design. I'm surprised that worked but pleased all the same. It's interesting that you noted such a change in performance over 'O' rings which one would assume to have sealed well. As an aside, I think this is where anecdotal evidence of people saying there were no leaks etc is very shaky evidence. I don't think it's possible to tell how much something leaks in a running locomotive unless it leaks badly. The lack of obvious leaks isn't evidence of no leaks. Very true especially when thinking about piston rings as any leakage will just be lost performance, however I did notice a very free powerful surge every time I opened the regulator. The orings felt powerful although slightly "sticky" the evidence of a good seal must be that it ran without hardly any regulator and notched up as far as it could go. I could pull away almost in mid gear then back the regulator off to almost shut so really no leaks. The rod seals are still holding up as well.
|
|
|
Post by Roger on Mar 23, 2015 22:01:40 GMT
It's a very promising sign and hopefully you'll very good service from them. Can you remind me what rod seals you used? Were they Viton?
|
|
|
Post by Doug on Mar 23, 2015 22:33:00 GMT
It's a very promising sign and hopefully you'll very good service from them. Can you remind me what rid seals you used? Were they Viton? They were silicon from blackgates engineering, I would however recommend the design not necessarily the material of my orings.
|
|
|
Post by Deleted on Mar 23, 2015 23:11:45 GMT
ablesealanddesign.ca/Public/files/PPR-1000.pdfNote how simple the seals are and note in particular that the rings are solid . The designers have understood and used the special properties of PTFE very effectively . With suitable materials choice for DIY versions this design of seal would work well with wet or moderate superheat steam .
|
|
|
Post by Roger on Mar 24, 2015 4:36:33 GMT
It's a very promising sign and hopefully you'll very good service from them. Can you remind me what rid seals you used? Were they Viton? They were silicon from blackgates engineering, I would however recommend the design not necessarily the material of my orings. Thanks Doug, I'll be very interested to see how they hold up over time.
|
|
|
Post by Roger on Mar 24, 2015 4:55:21 GMT
ablesealanddesign.ca/Public/files/PPR-1000.pdfNote how simple the seals are and note in particular that the rings are solid . The designers have understood and used the special properties of PTFE very effectively . With suitable materials choice for DIY versions this design of seal would work well with wet or moderate superheat steam . Hi Michael, I love the simplicity if this idea too and it might just work if the steam is as you suggest. There's still a huge amount of thermal expansion to accommodate though and I think one or more split rings would cope with that better. It's all worth trying though. I think I'm going to go to the other end of the complexity scale and use two pegged split rings with concentric rings under them plus 'O' rings like I've used on the piston valves. It's not that difficult to make and it takes the ring gaps out of the equation. It ought to be an interesting exercise and if it doesn't work, I've lost nothing but little time.
|
|
|
Post by Doug on Mar 24, 2015 6:03:38 GMT
I think that will work well, what type of PTFE are you using?
|
|
|
Post by Roger on Mar 24, 2015 6:30:59 GMT
I think that will work well, what type of PTFE are you using? I'll probably use natural PTFE since I've got some of the right size. I'm thinking of making the rings about 2mm thick from inside to out, and pretty narrow, say 2mm wide each. Since there's no need to peg them to the piston, I may just peg them to each other. I'm sure the design will go through many iterations before I'm happy with it. One thing to bear in mind is that the wider the rings, the greater the outward force from the steam pressure and the higher the losses. I'm relying on adjacent rings to seal the ends of the slits so there has to be at least two with this arrangement. When these rings get small in section, they are very flexible so it's a compromise between them behind small enough to fit in a small space while being practical to make. If I can avoid having any pegs going into the piston, there will be more room for the 'O' rings. I've probably stated this before but the 'O' rings don't form a seal, they just provide the outward pressure. There has to be zero ends float on the rings to get a complete seal and this is the only place where the design falls short of the ideal. The hope is that zero ends float forms a seal when cold, and as the temperature and pressure increase, there's enough force and compliance to allow the rings to seal on the diameter while they try to expand width wise. This is another reason for keeping them narrow. It's an interesting challenge though, and that's what interests me.
|
|
|
Post by joanlluch on Mar 24, 2015 7:31:24 GMT
I do like Michael's proposed design because of its simplicity, not to mention that such kind of rings may be obtainable off the shelf -hopefully in Europe- which is also something worthy to me. One concern with this design is that the working temperature range may be narrower due to thermal expansion of the ring. The manufacturer does not seem to specify a usable temperature range.
Roger, In case of the pistons (not the valve) I wonder if you thought on arranging the rings in separate grooves as it is done on automotive pistons. I mean instead of having adjacent rings on the same groove, keep them separated each one in its own groove. This of course will work better if the cut on the rings is just the right size to close at working temperature.
People in my club use modified off-the-shelf motorcycle pistons with their standard steel rings. This seems to work well, but I am not able to compare it with a properly designed PTFE rings piston.
|
|
|
Post by Roger on Mar 24, 2015 7:59:33 GMT
I do like Michael's proposed design because of its simplicity, not to mention that such kind of rings may be obtainable off the shelf -hopefully in Europe- which is also something worthy to me. One concern with this design is that the working temperature range may be narrower due to thermal expansion of the ring. The manufacturer does not seem to specify a usable temperature range. Roger, In case of the pistons (not the valve) I wonder if you thought on arranging the rings in separate grooves as it is done on automotive pistons. I mean instead of having adjacent rings on the same groove, keep them separated each one in its own groove. This of course will work better if the cut on the rings is just the right size to close at working temperature. People in my club use modified off-the-shelf motorcycle pistons with their standard steel rings. This seems to work well, but I can't compare with property designed PTFE stuff. Hi Joan, I don't think the commercially available seals are intended to cover the temperature ranges we have in mind. I think it's likely that they would have to be made smaller in diameter than they make them else they might be too tight when hot. Maybe if they are thin enough it might work... an experiment would probably need to be done to find out. My guess is that they would leak when cold just like the other solid rings that people have tried. I guess there's no reason why you couldn't have piston rings in separate grooves, but that would require the piston to be a very close fit in the bore to get a seal if there's a significant ring gap. With the exception if Clupet rings, any ring in a groove is going to leak at the gap, hence the need to control that. This is fine for cast iron but it's a problem for PTFE because the gaps are huge when they are cold. Putting two rings in one groove sounds like a good idea on the face of it, but doesn't really help because there's still a gap to the underside of the rings. This is why I've tried to seal the underside if any gaps with another ring. Even a simple Clupet ring allows steam to pass under the ring. So it seems to me that if you have a split ring system in PTFE, you either have to strictly control the gap and make it as small as possible when cold, or you have to stop the steam escaping with concentric multiple rings.
|
|
isc
Statesman
Posts: 708
|
Post by isc on Mar 24, 2015 10:16:57 GMT
Out of interest on plastic ring design: car shock absorbers use a stepped shape for the ring joint, this way the ring can have a large gap, while still sealing quite well. Admittedly these rings are made from sheet plastic, they are about 6 mm wide by 1 mm thick. isc
|
|
|
Post by joanlluch on Mar 24, 2015 14:22:47 GMT
Hi Roger, I understand your concept because I've been following your thread.
How difficult would be to make a simplified version of a Clupet ring?. I mean, make a section of the ring where half the depth of the ring is cut in a way that one half slides above the other half. This should prevent any leakage (that's not that different than your idea)
|
|
|
Post by Roger on Mar 24, 2015 18:32:26 GMT
Hi Roger, I understand your concept because I've been following your thread. How difficult would be to make a simplified version of a Clupet ring?. I mean, make a section of the ring where half the depth of the ring is cut in a way that one half slides above the other half. This should prevent any leakage (that's not that different than your idea) Hi Joan, In a small size I think that would be tricky but in a main piston ring that could be worth exploring. The tricky part is that the ring has to be made significantly bigger in diameter during manufacture so that the details of the overlap can be machined. If the ring was held in a jig to support the inside and outside then a 0.5mm cutter could be used on the side of the ring to create the desired overlap and split the ring at the same time. I think the overall length of the overlapping section would need to be at least 6mm but that's only 2mm on the diameter which on a large ring probably wouldn't matter. Two rings like that side by side would still be required unless they were made as proper Clupet rings. I'll certainly consider doing it that way. At the moment I'm thinking along the lines of a single split inner ring that's 4mm wide i.e. the full width of the groove, and two top rings with notches on the inner edges to peg them to it at 90 degrees. That would require just one peg that's in the middle of the inner ring. That's pretty easy to make and gives a huge overlap between the inner and outer rings.
|
|
|
Post by chris vine on Mar 24, 2015 20:38:42 GMT
Hi Roger,
I think you will have fun experimenting with rings. However I don't think you will need to worry about pegging them to stop rotation. In full size steam engines they are not usually pegged (until someone says otherwise!) and don't seem to rotate.
You seem to be putting them in the same groove, next to each other? I guess that helps seal the gaps, however there is perhaps more to be gained by having them separate so that they act as a labyrinth seal as well. It is the gap between the rings which helps create the large pressure drop: as gas leaks past the ring it has to speed up because the gap is small, then it expands into the relatively large gap after the ring, then has to speed up to go through the next ring. Each time there is a pressure drop because it is a lossy process. After several rings or grooves, the pressure ratio or drop becomes quite high. Michaelw will be able to explain this much better than I!!
If you look at piston valve bobbins, they had several rings and always spaced apart. In the old days, they tried to use wide flat rings (round of course!) which took tremendous skill to fit perfectly to the bores. however they leaked a lot of steam, especially when worn. (I don't know how good they were when new..) The multi ring is much more tolerant of wear.
Chris.
|
|
|
Post by Roger on Mar 24, 2015 21:25:14 GMT
Hi Roger, I think you will have fun experimenting with rings. However I don't think you will need to worry about pegging them to stop rotation. In full size steam engines they are not usually pegged (until someone says otherwise!) and don't seem to rotate. You seem to be putting them in the same groove, next to each other? I guess that helps seal the gaps, however there is perhaps more to be gained by having them separate so that they act as a labyrinth seal as well. It is the gap between the rings which helps create the large pressure drop: as gas leaks past the ring it has to speed up because the gap is small, then it expands into the relatively large gap after the ring, then has to speed up to go through the next ring. Each time there is a pressure drop because it is a lossy process. After several rings or grooves, the pressure ratio or drop becomes quite high. Michaelw will be able to explain this much better than I!! If you look at piston valve bobbins, they had several rings and always spaced apart. In the old days, they tried to use wide flat rings (round of course!) which took tremendous skill to fit perfectly to the bores. however they leaked a lot of steam, especially when worn. (I don't know how good they were when new..) The multi ring is much more tolerant of wear. Chris. Thanks for all that Chris. It's mostly theory and guess work for me so I expect a few surprises and disappointments. It's a very interesting subject though with lots of room for experiment. I'm not sure how much of conventional ring layouts and designs translate when using PTFE, it's chewy stuff that expand so much more than metal rings. I hear what you're saying about multiple separate rings and that's certainly worth exploring. It may also be that the rings stay where they are assembled since there's no twisting to make them move. I'll see how much more work it entails. As I see it, the biggest difference between metal and PTFE rings is that the gap on metal rings is very small when cold where it's huge on PTFE rings. If you feel inclined to account for that large gap, much of the conventional wisdom goes out the window. Although the flexibility and lack of elasticity can be a pain, it does mean that concentric rings can seal against each other more easily than with metal rings. I'm not sure what the limits of widths are with PTFE rings, I might be able to get away with 1.5mm rings but that's pushing it. I reckon I could get two grooves with two rings in each if necessary. I think two rings per groove is better than one since the gaps are large. I know Doug is using virtually no gap and one ring so it may be that none of this is necessary. It's interesting all the same.
|
|