solar air pump?

[MC_Hamster]

Greetings, all.. newbie here with a bit of a project in mind and some questions I can't seem to find answers for that I wouldn't mind knowing the answers to before I just dive in and find out experimentally :)

Idea in mind is to make an LTD stirling that'll be able to perform just a tiny little amount of work. The idea came from wanting to make a bit of a vertical garden and keep it watered... looking around here and there, there's little PV solar pumps around that'll do, but I'd like to have a crack at something a little different.

The designs I've seen often make use of an air-lift pump powered by a small aquarium air pump to bring water from a reservoir up to a height to trickle-feed the garden. The air is injected into a column of water and as the bubbles rise in it, it creates a current which draws more water in from the bottom.. Not a huge amount of lift height normally available, but if you narrow down the pipe a little, people seem to do OK. There's adaptations of that I've seen referred to as a 'geyser pump' (commercially produced and otherwise) which create a small chamber to 'buffer' the air before releasing it in bursts, which I figure would be able to make better use of a lower input flow rate to still provide a lot of lift, just more sporadically.

Anyway.. I'm wanting to use some form of stirling design to power it. Needs to be LTD, needs to be self starting. I've mucked with a very limited fluidyne, but couldn't get it running - I'm assuming the aquarium hosing I was using was just too restrictive.. plus from what I've seen of it, they're really not that suited to either LTD or actually producing any sort of usable power, though I do like that they have a ready-built pumping action.



So.. to mechanical stirlings.. The simple little gamma-style LTDs look like they have potential, but a: they're not self-starting, and b: I don't know just how much work you'd be able to pull out of them. I keep seeing that multi-cylinder arrangements can self-start.. is that just a matter of essentially hooking a multiple of them together on a common crankshaft and they'll run? Can I create an 'inline 4' with each cylinder (power/displacer pair, that is) 90 degrees out of phase with the next? (Which would mean I could run the power of one cylinder off the same crank pin as the displacer of the next, yes?). A power take-off running off the crankshaft could drive another piston working like a mini bike pump to give me an air stream.. Actually, it'd perhaps be possible to extend & seal the top of one of the power pistons, add the appropriate ports and valves and that'd work as a pump itself, no?

Since there's also a built-in water supply there, that could also be used to help on the cooling side.. The engine sits on top of the reservoir, open to direct sunlight on the top plate, with an aluminium heatsink bonded to the cold plate underneath, sitting in the water?

Anyway, to ask the experts here.. is such a thing even feasible?

Oh, also... 'Siemens' style double-acting engines.. Seems like there's less moving parts & hence less friction losses and probably simpler to build than the typical kit LTDs you generally see. I'm thinking a single 'block' with channels linking the top of one cylinder to the bottom of the next (filled with steel wool/aluminium strips/some other form of regenerator), which is then sandwiched between a hot and cold plate.. that should be fairly simple and also pretty efficient, no?

So many questions, every one just raising a bunch more. Anyway, would love to get some feedback from the experts.

Thanks, all.

[Geoff V]

Hamster

I would have thought the application you have outline would be best served with a Fluidyne, silent, no moving parts, what more could you ask for?

Having built one many years ago I do remember having to spend some time 'tuning' the pulse pipe but that's just a matter of patience.

GeoffV

[MC_Hamster]

Thanks for the reply, Geoff. Yup, I could certainly take another (better) look into a fluidyne.

Are there any resources around online that anyone can point to as far as design decisions and parameters go?

Though I've got myself all curious about the mechanical ones now, too.. Seems there'd be something very cool about a little 4-cylinder gamma spinning away...

[Geoff V]

Try this.

http://youtu.be/7Gtins-mMRg

GeoffV

[MC_Hamster]

Yup, already came across that one (and every other I could find on youtube), but I'm somewhat concerned by the fact that that example is using a blowtorch for a heat source, while I'm hoping to get there with unconcentrated solar.. The power in versus output of that one seems horribly out of proportion.. to extrapolate that to the handful of degrees I could accomplish in solar, and there's barely _anything_ there to work with.

[Geoff V]

Hamster

I'm getting the impression that buiding a gamma is of more interest than pumping water for your vertical garden?

The video link I offered was a poor choice on my part although I'm sure the operator was using a propane torch for convenience.

I would recommend reading the article by Dr Colin West before dismissing the Fluidyne and I would also respectfully suggest you do some 'broad brush' calculations on the amount of, unconcentrated, solar radiation you are likely to realise.

http://www.ornl.gov/~webworks/cppr/y2001/rpt/27113.pdf

GeoffV

[MC_Hamster]

:) Oh, I've gone wayyy beyond this being anything based primarily on practicality. If I were doing that, I'm pretty sure I've already got a solar powered pond pump that's perfectly capable of doing the job just hanging around somewhere. Being perfectly honest with myself, this is providing a convenient excuse to be a full-grown kid and play with stuff.

One thing I'd been wondering with the fluidyne, and it's nice to see it there in the equations, is the working fluid pressure. Noted in the text is "It is not convenient to operate this kind of fluidyne at a mean pressure above atmospheric (0.1MPa) because a higher pressure would expel the liquid from the tuning line."

However, by simply raising the end of the tuning line, you're creating a column of water in it. With the 'top half' of the system sealed and at atmospheric pressure, the water levels in the tuning line and both ends of the displacer column will all be equal.. but adding additional liquid to that line will cause a back-pressure in the displacer column, raising the water level and compressing the working fluid.. So with a tall tuning line and significant column of water in there, you could add non trivial pressure into that working fluid.. though.. I guess you'd also expend more work in raising that additional mass of water....



Hmmmm.

[MC_Hamster]

Well... attempt #2 at the fluidyne has been and gone.. Still not a hint of movement.. quite possible I've got a leak somewhere, also that my sizings are lousy.

Design cosists of two 500ml aluminium cans (like a regular old drink can, but taller... usually used for energy drinks around here, don't know how common they are elsewhere), from which I've removed the top with a can opener, loosely filled with some steel wool to provide more surface area (that'll cause corrosion in the long term, but for testing should be fine).. Small hole drilled through the base with 5mm aquarium air hose inserted & fixed with silicon sealant, that forms the 'breather' between the two chambers.. I then have 13mm (1/2 inch) ID hosing similarly fixed to some acrylic sheeting, which is again siliconed on to the open end of the cans.

Because I'm trying to get this (eventually) working by solar on one side and by water immersion on the other, my current setup has the 'cold' side immersed in a tank of water, with the displacer tubing rising up out of it and then dropping back down to form the 'U', going back up to the hot side sitting in the sun (or, for my testing purposes, in front of a 'paint stripper' style hot-air gun).


Does anyone out there have any other tips on troubleshooting one of these? I mean... I can't think there's much more to it than making sure it's sealed and that the geometry isn't totally broken, but even if it's less than optimal, I'd have expected a _little_ motion in the column..

[Ian S C]

Are you working from a prooven design, or close to it, the simplest way of making a fluidine pump is with laboritory glass ware, that is how it was origionally. Get a design, build it, then branch out and build your own design, after you get the first one working, and you fully understand what's happening. Don't worry, the fluidine is the only one of my motors that I have not got running reliably, must work on it one of these days. Ian S C

[MC_Hamster]

Well.. no, I'm improvising, which is no doubt an awful lot of the problem.. As to lab glass.. I'm working on a budget, here ;) would rather not spend up to buy gear I'll likely never use again, not when I'm really mostly prototyping at this point, anyway.

I have a few ideas, though.. I'm thinking maybe I have too much dead volume in there.. I'll give it a bit of a redesign so that I can get the cold side upright and cut out some of the extra length in the displacer column..

Will have to get some pics of the design, both the failures and (hopefully) working examples.



On the subject though... I had a though for a slightly different design.. obviously I'll want to get the 'simple' ones working first, but...

Rather than the standard 'loop' of them.. What if you used concentric cylinders? Hot end at the top with a narrow tube dropping straight down to an open end. That sits within a wider tube with a closed base, the displacer fluid filling both. Surrounding this is a third tube which forms the cold end. I'm just wondering about the 'breather' and regenerator.. whether you'd just be able to 'vent' directly between the tubes or would need some secondary loop around the 'outside' directly tieing the hot and cold. Anyway, that's another hypothetical altogether. First things first, self!

[fullofhotair]

Hampster;
Why not use the air pressure produced by the engine its self? You cant do this with a stirling engine but you can with an Ericsson cycle engine. Its similar to the stirling but is an open cycle, so the exhaust of the cycle is your air pressure. No reason you couldnt make a simple tin can model similar to the tin can stirlings on u-tube. Dont look for plans youll be the first.Probably want to make a simple stirling first just to get a feel. I had an idea a while back of using a stirling with a diaphragm as the power piston. The diaphragm on top would work a diaphragm air pump.Look at approtechie on u-tube.He uses diaphragm power pistons.The diaphragm would serve double duties.

[MC_Hamster]

I'd thought the same thing with using the top side of the power piston (be it a traditional piston-in-cylinder or diaphragm) to pump with.

I've never really looked at Ericsson engines much, but with a very quick skim I think I can see why you suggested it. Looks like I have some more reading to do!