Mini closed cycle steam engine, possible?

[AeroJetParticipant @aerojet]

Hi,

 

I’ve read bits of the answer elsewhere on this forum, but nothing identical, so I’ll ask here:

 

I’m toying with the idea of building a small steam engine for a project I’d like to kick off this year.

 

I want to make it a small steam turbine, but I also want to make the system closed cycle.

 

I know how to generate the steam, run it through a turbine then and then let it condensate to a steady state (presumably).

 

The turbine will be fitted to a small electric motor as a generator, all sealed up.

 

I’ve read the theory of closed cycle engines, and most of it makes sense, but from a small scale practical aspect, how would I pump the condensated water back into the boiler at small scale (about the size of a toaster)?

 

Obviously without talking about temperatures an pressures and configuration, I don’t expect a specific pump make and model. Just some general guidance.

 

I’m aware of this engine, but it isn’t closed:

http://maier-mechanic.com/en/gas-steam-turbine/

 

 

So what sort of pump would I go for?

More importantly, is a close cycle engine doable at small scale?

 

Regards,

Adam

[JasonBModerator @jasonb]

Have your turbine also power a small boiler feed pump to pump the condensate back into the boiler. Just needs a small throw eccentric on the shaft to drive a ram up and down in a cylinder with a couple of non return valves.

[bernard towersParticipant @bernardtowers37738]

Or use the generated leecy to run an electric pump

[SillyOldDufferModerator @sillyoldduffer]

The main problem I think on a small engine is the amount of energy available from the turbine to run a ram pump may not be sufficient, and an electric pump will be even less efficient.   It’s because nature doesn’t scale linearly, so what would be fine on a big turbine won’t cut the mustard on a little one.

I think the best chance of success would be to use the energy in the boiler to refill it via an injector.   This description may help.  Injectors are common on model steam locomotives, and the forum is quite knowledgeable on them.   Basically, when the driver sees the water gauge is low, he opens a valve that delivers boiler-pressure steam through a set of cones and nozzles that accelerate a high proportion of the flow to above boiler pressure so it gets back inside the boiler via an automatic one way valve.    As the steam accelerates through the nozzles, the high velocity flow creates a vacuum behind that pulls cold water from the feed-tank and injects it into the boiler.   The energy needed to shift the water comes from the boiler, not the turbine or generator.

The hard part is probably finding a way of automatically detecting low and high water and opening and closing the injector valve as necessary.

Dave

 

[BazyleParticipant @bazyle]

You are presumably inputting energy in the form of a gas flame or equivalent. So from that point of view some electrical energy input is also possible. The model boat fraternity sometimes just use an old servo set to run right round (ie feedback pot removed) to run the pump on demand.

[Nigel Graham 2Participant @nigelgraham2]

Ah, but what sort of pump, if using that option?

Turbines run at very high speeds so considerable reduction-gearing would be necessary to drive the pump if that is a reciprocating type. I think power-station and marine plant has used multi-stage (pressure-increasing) centrifugal pumps as they run fast, but I do not know how feasible this would be in miniature.

Using an injector is perhaps the best approach though going away slightly from the fully closed-cycle concept.

D.A.G Brown’s book on making injectors, although aimed at locomotive use, may be worth consulting because the ideal would be seem to be an injector matched to the turbine’s steam demand; perhaps with a second, slightly larger one for occasional “topping up”.

[HopperParticipant @hopper]

Have a look at some YouTube videos on how coal fired powerstations work. They are closed cycle (except the heat lost on cooling towers) so lots of good ideas on there on how to maximise heat reclaim (and thus energy conserve) at multiple stages of the cycle.

Things like using a superheater to add maximum heat to the steam. On a model, even if it is not truly superheated to any great degree, you will have nice hot dry steam at least. And using the warm water coming out of the condenser to pre-heat make-up feed water or boiler draft air etc . Likewise using boiler exhaust flue gas heat to warm up feedwater and so on. Thus maximising efficiency. Not a lot of energy to be saved on a small model, but not a lot to waste to start with either.

Plenty of designs around for the kind of small reciprocating piston boiler feed pumps Jason refers to. As Nigel says, major reduction gearing would be required, which of course sucks up energy in frictional losses etc.

Your biggest enemy on a model might be leakage. Loss of steam and water will require the addition of makeup feedwater. So the better you can seal your turbine housing and shaft seals the better. Again, beware of frictional losses on shaft seals. The full sized ones use labyrinthe seals with no physical contact, just tiny gaps. Not sure that would work on a model size?

The other critical loss may be of steam from the condenser and contradirectionally, air entering the system and stopping the condenser from condensing steam into water. The air comes out of the water, in which it is dissolved. It is entrained with the steam and travels through the turbine into the condenser. There it accumulates and will displace the steam and stop it contacting the cool condenser surfaces and condensing into water.

Full size rigs use an air extraction pump to remove this air from the condenser. (And contrary to somewhat popular opinion, in modern systems, the air pump does not create the partial vacuum in the condenser, the 1400-times shrinkage of steam into water does that.) So you would have to have some kind of air extraction pump that did not also suck out steam. Again, no idea how this would work at model size as I only speak from experience in full sized powerstations. But it is something to look into.

Plenty on YouTube about powerstation condensers, economisers, and full cycles. As usual, some of them are even correct. Others, maybe not so much! Always look at your sources and decide how reliable they are. The videos by large power and turbine companies etc for their staff training are probably the best. Others by CAD jockeys working on their engineering undergraduate degree projects, beware!

The ultimate closed cycle steam systems are probably marine applications where they try not to dump any heat overboard for the sake of fuel economy. (The discharge from the “head” (toilets) is said to be the warmest overboard discharge on a well run ship) But with the decline of steam turbines at sea there do not seem to be so many training videos on that side of things. Plenty of old books on the subject though.

[DiogenesParticipant @diogenes]

It’s in the back of my mind that small but efficient feed pumps were a subject of great interest to flash-steam boat racers, the reciprocating type described by Jason I think coming out on top – a peer down that particular rabbit-hole might show that much effort and experiment has already been done for you by men very concerned with minimising losses.

[AeroJetParticipant @aerojet]

Hi ,

 

I did not expect this much feedback on this topic, so thank you all for your input.

@ JasonB

Are the small feed pumps you mention an off the shelf part for model engineering or something that I’d have to craft (quicker to buy, perform ideally if built )?

@bernardtowers

Not sure if I’ve seen a electric pump of the sort JasonB is suggesting.

@sillyoldduffer

Useful link to steam injectors. I’ll have a dig around and see if I can find something small.

However, the feed tank is  vented to atmosphere is it not?

I can possibly workout water levels using a metal ball float and some sort of magnetic sensor. Cross that bridge when I get to it.

@nigelgraham2

Looking at injectors again, thanks for the book reference. I’ll see if I can get acopy. Will be usefull for other projects.

@hopper

I’ve looked at power station designs. TBH was my first point of reference. They have the luxury of using much larger centri pumps that make some of the losses seen at small scale irrelevant.

Wonder if NASA ever built any experimental kit that was compact enough for deepspace/plantary work where sunlight would have been insufficient? Like some sort of RTG with turbines and working fluid.

Perhaps I could use some of the boiler pressure to operate a high pressure one way piston valve in a i pilot configuration, that uses a solenoid to trigger it every so often.

I’ll look into the marine engines, i suspect designs from the 1950-1970’s might be the time to look at.

 

@diogenes

I’ll look into flash steam boat racers and see what has been done.

TBH, never heard of that field, so would be an interesting find.

Thanks

 

All, seems I have a bit of background reading to do based on the info you have kindly provided.

I’ll report back if I have any more questions after digesting what I find.

Thanks to you all,

Adam

[HopperParticipant @hopper]

Stuart Models sells kits for boiler feed pumps, although a bit hard to find on their website. There is a hand pump, easily converted to eccentric operated, and another option, a reciprocating steam-driven pump a bit like the old full sized Weirs pump that was once ubiquitous, on this page here https://www.stuartmodels.com/products/set-of-castings/page/3/

They also sell parts for a variety of pumps elsewhere on their site but full pump kits or completed pumps seemed elusive to me (which doesnt mean much these days!).

I would expect that other suppliers of model steam engine kits could supply similar pump kits. Suppliers such as Blackgates, PM Research, EJ Winter and all the usual suspects who advertise in ME mag.

Some model locomotives use an axle-driven feed pump that could be converted to run off your model turbine possibly.

[Merddyn’s DadParticipant @merddynsdad]

Regarding the feed pump and possibly the air pump aspects of this project, I’d consider the use of gear type pumps.

The fact that they both run ‘wet’ is a bonus.

My previous attempts, may years ago, to achieve  a satisfactory result on a similar project were failures due to my inability to make miniature ones accurately enough, my choice of available materials and insufficient speed to drive them.

Nowadays, modern polymers/ceramics may fix the material issue, the turbine would provide the speed.

Lots of development work required though I’d suggest.

My experiments with an electrically powered pump for boiler feed converted me instantly to miniature injectors and off I went down another ‘rabbit hole’, beware there’s a whole world of distractions down there!

 

[duncan webster 1Participant @duncanwebster1]

We’ve no idea of the scale of this proposal, but a member of our club has an electric water pump under the tender of his 5″g Brit. It works really well, even if I reckon it’s cheating.

The labyrinth seals on big turbines are fed with steam part way along so that steam leaks in rather than air. You were meant to adjust the supply valve so that a whisp of steam came out at the atmosphere side. This is because getting a small amount of extra water out of the condenser is easier than getting air out.

Intuitively I’d avoid using a gear pump, I reckon in really small size they will leak as much as they pump.

The model boat guys are well versed in detecting water level. I believe they use a resistance probe into the boiler, bug I’ve seen and experimented with a light probe on the gauge glass. I think commercial systems are/were available

[Tim StevensParticipant @timstevens64731]

Just a thought: I am not aware of an injector system which always retains every last drop of the water it uses. A pump will leak a bit but it will be easier to run it in a closed space so the leakage is caught and recycled.

But what do I know about it? I just answered a different question here in general terms, so the answer would also be helpful in similar cases but not exactly the same, and was castigated for it. Just make your minds up, everyone, do you seek general advice which may be useful to others, or only advice specific to the  actual make and model and problem that you have in front of you? If the latter, most of us cannot help because most questions are not specific enough – are they?

At your service or not? You choose.

Tim

 

[Nigel Graham 2Participant @nigelgraham2]

An interesting project…

.

From several posts back:

”

However, the feed tank is  vented to atmosphere is it not?

I can possibly workout water levels using a metal ball float and some sort of magnetic sensor. Cross that bridge when I get to it. “

Do you mean the water-level in the feed water-tank or the boiler? Either way can you not use a simple gauge-glass, or is this for an automatic control circuit? (Still wise to add the gauge-glass as well, anyway.)

….

Tim –

Leaky water-pump? Could the pump cylinder be on or in the feed-water tank?

[Turbine GuyParticipant @turbineguy]

Hi Adam,

I can’t add anything to what already has been said about feed pumps or injectors but I am very interested in your steam turbine.  If you have already decided on a particular type to make or have a source to purchase from I would like to have a little more information. I started the Model Turbines thread in this same forum to gather and share information about model turbines and hope your project will be of interest and help others interested in making model turbines.  If you haven’t already decided what you are going to use for a turbine this thread might be helpful.

Good luck with your project,

Byron

[Mark RandParticipant @markrand96270]

Use a steam driven reciprocating boiler feed pump and a modified version of the same as the condenser air ejection pump. You could use a steam air ejector for startup, but probably not useful.

[Tim Stevens]

On Tim Stevens Said:

Just a thought: I am not aware of an injector system which always retains every last drop of the water it uses. A pump will leak a bit but it will be easier to run it in a closed space so the leakage is caught and recycled.

But what do I know about it? I just answered a different question here in general terms, so the answer would also be helpful in similar cases but not exactly the same, and was castigated for it. Just make your minds up, everyone, do you seek general advice which may be useful to others, or only advice specific to the  actual make and model and problem that you have in front of you? If the latter, most of us cannot help because most questions are not specific enough – are they?

At your service or not? You choose.

Tim

 

If it were a lifting injector, the overflow could be back to tank, on fact even if the injector is below the tank it might be possible to route the overflow to the tank, like a central heating tank. However, injectors don’t like hot water, and you don’t want to cool the condensate just to make injectors work, wasting heat

[Author]

Posts