Uniflow design theory

[Richard Hed]

Going back to Anatol's OP, the efficiency of a uniflow engine exhausting to atmosphere will likely benefit from auxilliary exhaust valves (controlled by valve gear).
A condensing uniflow engine exhausting into a good vacuum doesn't need them.

Without auxilliary exhaust valves, a non-condensing engine needs a large clearance volume to avoid over-compression.

Relief valves (which are different from exhaust valves) are needed particularly in the condensing case because they can have a smaller clearance volume of 1.5 to 2 percent of the swept volume without over-compression, but should the vacuum be lost the compression pressure could get dangerously high without relief valves.

I could not make out anything Dan Gelbart said over the racket the engine was making, especially as he appeared to have his back to the microphone.

Any sudden process is bad in thermodynamics as it is 'irreversible'. This is what the Second Law is all about. As Michael Flanders sings, "That's entropy, man."

Richard's long ramble about flow resistance in tortuous passages is generally on the right track, but I would suggest that in the case of a normal slide valve cylinder the thermal losses due to passing the steam and exhaust alternately through the same duct are considerably greater than the flow resistance, but that is no more than a well-educated guess. The Reynolds Number of the steam flow in model engine passages is much lower than in full size, mitigating the effect of the multiple holes and sharp corners.

As to the suggested problems with piston rings passing over ports, I am not convinced. Most main-line steam locomotives built after about 1925 had piston valves in which rings pass over ports, and in far more trying conditions than found in stationary engines, particularly as a result of pumping air through the cylinders when coasting.

Steamchick, in the 1909 edition of his famous 'Heat Engines' W Ripper* says, "The laws which govern the condensation of steam in the cylinder are not at present fully understood."

* William Ripper - Wikipedia

Yes, yes and yes. However, that frictional loss IS energy loss (heat loss). I suspect the back and forth thru the same passage, as you say, has some kind of loss like you are explaining--there is probably at some point a bit of "shock" in the sysstem somewhere which leads to inefficiency. also, the length of the passage doesn't help at all. In these typical slide valves a shorter tunnel/tube/passage way (I forget what it's called, sorry) could and should be supplied with very little design trouble and certainly easier to manufacture.

Thanx for mentioning my "ramble". You could have at least said "entertaining ramble". Did you like the main character?

 

[Richard Hed]

It depends on the back pressure. An atmospheric engine will exhaust down to a bit above an atmosphere. In a steam locomotive, where the exhaust blast is used to draw the fire, the back-pressure is a bit higher. A condensing engine cylinder is more thouroughly evacuated by the condenser vacuum. So, as the exhaust valve closes, you will have trapped in the cylinder whatever volume that represents of more-or-less dry saturated steam at whatever the back pressure is. Steam Tables will allow you to calculate the mass. Equally, if you know the inlet pressure, and cut-off, and the amount any superheat, you can calculate the mass of steam you are starting with (neglecting condensation). The difference is the amount exhausted. I suggest you look into cylinder P-V or 'indicator' diagrams as a starting point.

For another way of thinking about the quantity of steam exhausted, I am tempted to quote Tom Lehrer: "Life is like a sewer ... you get out of it what you put in".

That's great, that's too funny. Yesterday I watched Jay Leno's vid on his steam car collection. I didn't thimk I would enjoy it, but Leno is brilliant and he showed his two Dobles plus a cut-away of a Doble. the Main Doble used the steam right down to the last drop before being sukt into the condenser. It used th e steam about four times. I really recommend that vid. That Doble was a wonder and complex enough to take a look at the vid. Just look up Jay Leno on utub.

Well for these little toy engines, are the steam tables and what not, are they not overkill? Even tho' the principles are totally sound. But for designing a system that is meant to DO something, it is sound advice. I'm looking at between 2 and 5 HP. That one I mentioned to leap over tall buildings.

 

[Anatol]

Well, you wanted some critique so I can provide a start.

Not true - even if the cylinder dropped to atmospheric at BDC the pressure developed by compression could be very high, depending on the compression ratio of the cylinder.

yeah, that's what I said too

 

[Anatol]

Steamchick,
thankyou for that fascinating and knowledgeable historical view !
"So we should try and find some original texts to understand these "antiquated" designs"
wise words indeed. I have some of the old texts, notably -

Hawkin's New Catechism of the Steam Engine 1897/1904.
James+Doyle's Mechanism of Steam Engines 1914,
Holmes, The Steam Engine, 1895
and Harris' Stationary and Marine 1958 (recommended by fcheslop as
I recall. )
Do you recommend others in particular? (Re uniflow, I don't have Stumpf.

 

[Anatol]

This discussion about losses is a matter of the second law of thermodynamics.
I suggest an interlude for a brief explanation (bear with the intruduction):

and other favorites, like "I'm a Gnu"

My favorite succinct paraphrase of the 1st and second laws of thermodynamics
1. you can't win
2. you can't break even.

 

[Anatol]

Going back to Anatol's OP, the efficiency of a uniflow engine exhausting to atmosphere will likely benefit from auxilliary exhaust valves (controlled by valve gear).
A condensing uniflow engine exhausting into a good vacuum doesn't need them.

Without auxilliary exhaust valves, a non-condensing engine needs a large clearance volume to avoid over-compression.

Relief valves (which are different from exhaust valves) are needed particularly in the condensing case because they can have a smaller clearance volume of 1.5 to 2 percent of the swept volume without over-compression, but should the vacuum be lost the compression pressure could get dangerously high without relief valves.

I could not make out anything Dan Gelbart said over the racket the engine was making, especially as he appeared to have his back to the microphone.

Hello Charles
3 related remarks/questions concerning valves -

"a uniflow engine exhausting to atmosphere will likely benefit from auxilliary exhaust valves (controlled by valve gear)."

why 'controlled by valve gear' ? What would be wrong with a passive poppet valve with a weak spring that opened at low pressure (ie after exhaust to ports) and stayed open till inrush of supply steam, which would push it closed. I guess you'd lose that recompression effect, inducing 'wire-drawing' ?

you said something about double beat valves - I searched but found nothing useful - they seem pretty obscure.

"A condensing uniflow engine exhausting into a good vacuum doesn't need them."
yes, I'd thought of that.

You mentioned a 5% cutoff for uniflow engines, and opening before TDC. A bash valve seems like a viable solution (?)

A

 

[dazz]

Dan Gelbart's Uniflow Steam Engine

 

[Charles Lamont]

why 'controlled by valve gear' ? What would be wrong with a passive poppet valve with a weak spring that opened at low pressure (ie after exhaust to ports) and stayed open till inrush of supply steam, which would push it closed. I guess you'd lose that recompression effect, inducing 'wire-drawing' ?

Yes, and they would probably let quite a bit of steam go to waste before closing. Try it!

you said something about double beat valves - I searched but found nothing useful - they seem pretty obscure.

Not really. Mushroom poppet valves were the norm in smaller steam applications such as cars and lorries, but anything bigger tended to have double beat valves.
Do a search on Caprotti valve gear.
(And for an absolutely outstanding example of model engineering see this: )

"A condensing uniflow engine exhausting into a good vacuum doesn't need them."
yes, I'd thought of that.

You mentioned a 5% cutoff for uniflow engines, and opening before TDC. A bash valve seems like a viable solution (?)

Indeed. Benson & Rayman's book 'Experimental Flash Steam' has a picture of Jim Bamford's bash-valve, uniflow, v-twin, racing model hydroplane engine on the front of the dust jacket.

 

[Anatol]

Indeed. Benson & Rayman's book 'Experimental Flash Steam' has a picture of Jim Bamford's bash-valve, uniflow, v-twin, racing model hydroplane engine on the front of the dust jacket.

I believe I have a copy of that , I'll go look . thanks.
"Do a search on Caprotti valve gear."
ok
thx A

 

[Steamchick]

Thanks Charles. When I find that elusive Round Tuition I'll consider tracking down a copy for a bed-time read. Just have to get through the dozen or so similar books I inherited from father and 2 grandfathers, including the 1890s print of Maxwell's electromagnetic theory. So much to learn and experience, so little time....
K

 

[Steamchick]

Arrrgghh! Spell check has changed Tuit into Tuition. - I need that as well, but don't appreciate a machine telling me so!

 

[Richard Hed]

Thanks Charles. When I find that elusive Round Tuition I'll consider tracking down a copy for a bed-time read. Just have to get through the dozen or so similar books I inherited from father and 2 grandfathers, including the 1890s print of Maxwell's electromagnetic theory. So much to learn and experience, so little time....
K

Just be careful when you read in Maxwell's theory that when "the four equations" are explained--those are NOT Maxwell's equations. Maxwell had 20 equations. It was Oliver Heavyside who took those 20 equations when he found them and condensed them down to the well know four equations which Maxwell gets the credit for. If you can find it, Heavyside's explanation of that is out there in his published stuff somewhere. I get upset by Maxwell getting the false credit. Heavyside was very important but gets extremely little credit for the great stuff he didded.

 

[Steamchick]

Hi Richard,
I worked on Air "motors" - well single stroke actuators for Circuit breakers - back in the 1980s.
Very simple 2 piston assemblies, providing Huge forces for a short stroke - driven from an air receiver - with much lower forces for the return stroke. I am vague on how we did the calculations, but as we were dealing with non-condensing gases we would simply have used the gas laws... for adiabatic expansion.
But for the Uni-flow engines, the calculations or the steam driven stroke should be like any engine, and allowing for adequate passages and hot engine, with adequate super-heat, the "filling" should obey the gas laws for Steam expansion, with adjustment if it cools to condensing condition. (Insufficient superheat). Now as the steam in the cylinder is dry, you can use the gas laws to expand through the exhaust to a huge volume (the world) - which will be a constant pressure, to 3 decimal places! If however the steam has expanded to wet condition before the exhaust opens, then you will need to determine the steam pressure upon opening from steam tables, which (I think?) will account for the condensation...
I haven't sat down to write the maths, but it should not be difficult - if you have the brain to understand Maxwell, Heavyside and co! From what I remember of 50 odd years ago, Boyle & co were relatively simple by comparison... It was Plank and Schroedinger that blew me away...

 

[Cogsy]

If you can find it, Heavyside's explanation of that is out there in his published stuff somewhere. I get upset by Maxwell getting the false credit.

Heavyside himself considered that it should be called Maxwell's theory and said as much in the first volume of his Electromagnetic Theory books. And Oliver Heavyside is reasonably well known in mathematics with at least a few theorems and functions named after him. For a self-taught guy he did some remarkable work (actually even for a fully-trained guy he did some remarkable work!).

 

[Richard Hed]

Heavyside himself considered that it should be called Maxwell's theory and said as much in the first volume of his Electromagnetic Theory books. And Oliver Heavyside is reasonably well known in mathematics with at least a few theorems and functions named after him. For a self-taught guy he did some remarkable work (actually even for a fully-trained guy he did some remarkable work!).

I didn't know Heavyside considered that as you say. Even so, I thimk that today, it should be called the Maxwell-Heavyside equations. The Theory is Maxwell's but the equations are not. Yes. It certainly surprises me that so many peeps on this forum are familiar with Heavyside. It is a pleasure.

 

[Richard Hed]

Hi Richard,
I worked on Air "motors" - well single stroke actuators for Circuit breakers - back in the 1980s.
Very simple 2 piston assemblies, providing Huge forces for a short stroke - driven from an air receiver - with much lower forces for the return stroke. I am vague on how we did the calculations, but as we were dealing with non-condensing gases we would simply have used the gas laws... for adiabatic expansion.
But for the Uni-flow engines, the calculations or the steam driven stroke should be like any engine, and allowing for adequate passages and hot engine, with adequate super-heat, the "filling" should obey the gas laws for Steam expansion, with adjustment if it cools to condensing condition. (Insufficient superheat). Now as the steam in the cylinder is dry, you can use the gas laws to expand through the exhaust to a huge volume (the world) - which will be a constant pressure, to 3 decimal places! If however the steam has expanded to wet condition before the exhaust opens, then you will need to determine the steam pressure upon opening from steam tables, which (I think?) will account for the condensation...
I haven't sat down to write the maths, but it should not be difficult - if you have the brain to understand Maxwell, Heavyside and co! From what I remember of 50 odd years ago, Boyle & co were relatively simple by comparison... It was Plank and Schroedinger that blew me away...

Well, to tell the truth, I never thot of that and I am very lazy but that doesn't look like a difficult thing to do. Also, for models, I doubt that superheat (that is, REAL superheat up into many hundred degrees) will ever be used. We aren't dealing with rocket science really, but if we wanted to, we could indeed go so far as to make rocket science calculations just because we can.

 

[Steamchick]

Last Uni flow I saw was in a fast boat (I think it was the UK record holder for whizzing around in a circle). Steam supplied at high superheat as the flash boiler was red-hot! Engine running at 50,000 rpm or whatever.... I think about 5cc? Sounded like an aero - engine 2-stroke I engine without silencer! No idea of pressure or temperature...
K

 

[Anatol]

Last Uni flow I saw was in a fast boat (I think it was the UK record holder for whizzing around in a circle). Steam supplied at high superheat as the flash boiler was red-hot! Engine running at 50,000 rpm or whatever.... I think about 5cc? Sounded like an aero - engine 2-stroke I engine without silencer! No idea of pressure or temperature...
K

50,000 rpm !?!

 

[Anatol]

So... while we're talking recompression and relief valves (or not), there's been passing reference to condensers. It make good sense to reduce exhaust port pressure to < atmosphere. I see in principle how this might work, but has anyone made one recently? Firstly, I'd like to know clever solutions to avoid lots of plumbing. As Stumpf says, it defeats the purpose to have any obstacles in the path of steam flow from port to condenser. (I feuded a free source for Stumpf 1922, link is in the designing steam engines thread.
I'm seeing a vessel with radiator fins or similar, with perhaps boiler feed water passing through to additionally cool the exhaust and so condensate can be recycled. Can anyone add details/design solutions? I've seen some images of 'jet' condensers but there doesn't seem to be much info, unless I'm missing the key search term.
thx!

 

[Steamchick]

50,00pm or maybe a factor of or 5 less... it was screaming! I çant remember, but the web will...
On condensers: for a small tug I used concentric copper tubes, steam and condensate outside, pond water drifting through the inside per outlet close to front of screw.very successful, as the pond had an adequate supply of cold.... (In Sunderland there is always an adequate supply of cold everywhere!).
But if you are pumping out a lot of kW of steam from your boiler, getting 1 or 2 kW from a high speed boat engine, then the rest (lots!) Of the kW need a Huge condenser... Look at steam tables for the heat in steam at 1 bar abs at 100C, versus water at 30C, based on the volume of water you boiler will use, and develop the heat flow required at the condenser.... Then look at the radiator on your car - which can take 10 or 15% of the power your engine can develop.... So if you have a 75 kW engine, the radiator can manage maybe 10 to 15 kW? It may give you a clue as to the size of radiator you could need, if you pro-rata size (areas) for your needs.
Does that help?
K