Sterling Engine- Pump variant: will it work?

[Glurth]

I've been wracking my brain for a simple way to implement a Sterling Engine as a water pump, rather than an engine, and using solar power as a heat source, rather than a flame.

The attached diagram is the simplest implementation I could come up with: Rather than pushing/pulling a piston, like in a Sterling engine, the gas in the outer cylinder pushes/pulls a column of water. Rising water column should suck in water through the bottom valve; descending water column should push the water out the top valve.
(Alternate design: Rather than an inner and outer cylinder, it could be configured as two separate cylinders, one for water (with the valves), and one to heat gas in. These two cylinders would be connected by a tube at the bottom.)

If someone already came up with this, which I suspect is the case, please ignore the rest, and just let me know what term/name I should look-up.

I don't have the physics knowledge to figure out if this design will actually work like a sterling engine, and pump the column of water up and down. Will it? Or, will the dynamics of water somehow mess up the cyclic nature of the expanding/contracting gas?

If it WOULD work, would it self-prime? (i.e. Starting from reservoir water level, would it fill the inner cylinder with water?)

Any other things you think I should consider?

 

[AlephZero]

See http://en.wikipedia.org/wiki/Fluidyne_engine

 

[Glurth]

Thanks Aleph, amazing what the right search term will do! Looks like the same principle indeed, but any idea why that design is so much more complex (pistons, spirals, floats, etc..)?

 

[AlephZero]

A "conventional" Stirling engines have two pistors that work 90 degrees out of phase with each other. One generates the power, the other one pushes the working fluid between the hot and cold regions of the engine.

With the fluidyne, one of those two pistons is the oscillating fluid level. That has its own natural frequency of oscillation, depending on the shape of the fluid container. The fluidyne engine will only run properly at the same speed as the fluid oscillates.

The reason for the "spiral" is to change the oscillation frequency. Forcing the fluid to go round several turns of spiral lowers the frequency, so there is more time to heat and cool the working fluid, and the engine works more efficiently for the same amount of heat input.

You should be able to find plans for a simple fluidyne engine on the web, but they are a bit trickier to get working than a conventional Stirling engine. I would suggest you make a conventional engine first, even if it is a very simple "bean-can" design (i.e.the cylinder is made from a food or drink can, not an accurately machined cylinder.)

 

[Lok]

Hi Glurth,

Your design is simple enough and it will work with day/night cycles (so not a true Sterling).
If there is a way that you could stop the Solar radiation when the water has been pumped up, in order to cool the gas you could get a working cycle. Maybe a sleeve which can be raised via magnets (:P the nuclear option) from a floater inside so that some-kind of stripes block another stripe pattern at a given height.

 

[Glurth]

Hi Lok,
You are of course correct, I don't have a thermally separate cooling chamber in that diagram! DUH!
The idea of using a movable shade maker, to turn the heating chamber into the cooling chamber, and back again, is great. Using the up and down movement of the air/water column seems pretty simple indeed, but I worry that it will cause problems when it's say 1/2 way shading the hot chamber, and the air is only 1/2 way expanded (less heat available to complete the expansion). Or, am I missing something there (like the stripes pattern you mentioned- would they eliminate this)?

 

[Lok]

Using the up and down movement of the air/water column seems pretty simple indeed, but I worry that it will cause problems when it's say 1/2 way shading the hot chamber, and the air is only 1/2 way expanded (less heat available to complete the expansion). Or, am I missing something there (like the stripes pattern you mentioned- would they eliminate this)?

Well a normal Sterling does have the same issue with the 1/2 shading, in the phase where the piston is halfway between hot&cold. The way it still works is simple, it uses a the momentum of it's flywheel to get through the difficult phases of it's cycle. You do not have to worry much about that as the momentum of the moving water works for the cycle and not against.

The stripes will sadly cut your solar in half. But there are other ways to cut the radiation. Mine was just a simple example. Go crazy.

Have fun.

 

[Glurth]

After your input, I would indeed like go crazy; for now, by amending my initial diagram with three notes: and ask a couple of questions about the result. Madness!
Diagram amendments : (or let me know if I should just redraw it)
1) The diagram represents a box, rather than a cylinder. So, there are TWO air chambers.
2) An air-hose connects the two air chambers at the top.
3) The left chamber is cooled, rather than heated.

Questions:
1) According to what I've read so far on fluidyne, this should cause an oscillation between the water columns that partially fill the air chambers. a) Is that correct in this setup? b) In all the diagrams I've seen of a fluidyne, the connecting air-hose is narrower than the air chamber, why is that?
2) Let's say there IS an oscillation between those two water columns, would that actually force any water through the one-way valves?

I guess what's bugging me is that the fluidyne uses a separate "working fluid", rather than the fluid it's actually pumping. Is this mandatory for some reason?

 

[Lok]

1) The diagram represents a box, rather than a cylinder. So, there are TWO air chambers.
2) An air-hose connects the two air chambers at the top.
3) The left chamber is cooled, rather than heated.

Connecting the two chambers will not give any result, it will be just like the cylidrical setup where the chambers are already connected. Cooling one chamber and heating the other in this setup will not generate a cycle. You still need the expand=>cooling=>contract=>heating=>... part. The expansion will need to condition cooling which will result in contraction and so on.

While it is a fun experiment to do at home, keep in mind that the energy generated by a working setup is small and it might not be able to pump too much water at a high level.

Also the gas being in direct contact with the water will at some point dissolve away or grow. The direct contact to the water will cut some of the efficiency as it will act as a thermal sink for the gas. That is why you should have a different "working fluid" although it is not mandatory.

About the Questions on the fluidyne: 1a. If I understood the Question right, yes. 1b. Air can be easily transported thorugh thin tubes without significant friction unlike liquids. 2. For the fluidyne it will.

 

[Glurth]

>> Cooling one chamber and heating the other in this setup will not generate a cycle.

hmm, so what do I have wrong? I tried to do something similar to what I saw on this linked paged (link below). I do indeed have a different output setup, but the air/water columns look the same as in fig 1, no?

http://jesseenterprises.net/amsci/1985/04/1985-04-fs.html

 

[fleebell]

Glurth,
I've been working on some ideas kind of in the same area you might want to look at for ideas. The page on my site is : http://www.packratworkshop.com/minipwrwater.htm
I've been trying to come up with all sorts of variations on the theme of using solar heated hot water for an engine. I'm currently working on a design based somewhat on a Minto wheel that is only about 3' in diameter and horizontal versus vertical.