Water cooled Air

[Aggie]

I concur, and for the installation I'm planning, that's where it will be for any hot part of the day.  There just seems to be a lot of anecdotal evidence suggesting that these sort of systems cool things down 'in the sun'.

Now, in a system using cloth-covered coroplast panel with very little thermal mass and a high enough surface area to make wind a significant factor, do you think it'd matter?  I might try this for cooling the loft, by placing it outside of one of the windows.  It could be mounted to the north, south or west.  West would allow for rooftop mounting, but north has a roof over top of the front stoop that'd also likely work.

[Bob in a quantum-state-of-faith]

I'm assuming a "coroplast panel" is one of those plastic things that are like corrugated cardboard, only made of LDPE?

I've often thought of using a black one as a solar water heater engine-- plumbing it into my domestic hot water tank, with a thermostat testing the water coming out of the heater-- if it's hotter than the tank?  The electric pump is on.  If it's cooler?  Shut down the pump.  I would have to use a heat-exchanger loop, as I don't think the panel can withstand domestic water pressure, alas.   And that's really the only thing keeping me from doing this project-- I routinely get black ones from major A/C installs (they use them to cover the fins during shipment).   I just cannot think of a suitable heat-exchanger that would also be cheap.

One possible suggestion, is tube-within-a-tube, in a giant coil, and the whole thing immersed in some form of insulation-- spray foam?    That is a smallish, say 1/4" tube for the low pressure hot water from the roof, inside a 3/4" higher pressure tube from the hot water tank itself.  The beauty of that, is that the loop could be laid flat, below the tank itself, and let simple convection carry the hotter water up into the tank (by way of the boiler drain port).  The return, cooler water would come out the top, via a T installed at the top of the tank.  Convection would circulate the hotter water from the tube, into the tank.  Insulation keeps the whole thing hot.   

The roof-loop could even be filled with antifreeze (non-toxic) so I wouldn't have to shut it down in the winter, at night.    I'd want to keep the panel pretty flat, to avoid high pressures at the bottom end.  The only thing I cannot solve, is how to make a manifold between the copper tubing and the plastic corrugated sheeting:  being made of polyethylene, nothing really sticks to it all that well, and I don't want leaks to have to make up-- I was going for a sealed heat-loop.

Sorry about the thread-jack....!

[Aggie]

You are correct on coroplast.   Look around the 'net... I've seen homebrew designs on heat-exchangers made with it.

[Bob in a quantum-state-of-faith]

I've been thinking about that again-- what I need, is a waterproof adhesive that is heat-resistant, but remains tacky-- similar to asphalt, only some that doesn't melt with heat.

I'm thinking of some sort of polyurethane, some that does not get brittle-hard, but sets up as pliable (and somewhat tacky).

For a manifold, I was thinking of using a heat-resistant plastic tube, which I will slit down it's length using a table-saw & a jig. CPVC ought to do the trick.  If I can make the slit small enough, the tube ought to "clamp" slightly onto the plastic sheet.  Then plug one end with silicone (or urethane) glue, and fit up the other end to my supply/return tubing.  

However, I suspect that if I rough up the plastic sheet with course sandpaper, and use simple silicone RTV, it ought to hold well enough for the low-pressure application I want.

I like your idea of covering it with terry-cloth (or similar) and allowing cooling/evaporative water to trickle down the outside-- catch it in an overflow basin (I'm thinking of a home-made trough made of 2" or 3" plastic sewer pipe, cut in half--or even just a strip cut out of the top to receive the panel & it's lower manifold.   A very small pump at one end would recirculate the water back to the top.  A trough made of plastic pipe with holes drilled along it's length would let the water drip out all along the width of the panel, soaking down through the terry cloth.   You'd need some sort of wire across the terry, to hold it down in the wind-- perhaps some coated chicken-wire fencing?  They sell fencing that is powder-coated (and vinyl coated too) that ought to resist the constant soaking it would get.   A wooden frame to hold the "sandwich" together along either side.  A side-channel of plastic pipe (slit to fit over the panel) would keep the water from rolling off either edge.  Or simply fold the panel up a bit--holding the fold with the wooden frame.

I see no reason why this would not work-- again, I'd put it in the shade, if cooling is what you want, here.   And a white panel would be better than a black one, I'd wager, for cooling.

If the panel can withstand the pressure of being vertical?   (and you could help, by cutting it down to say 4 feet tall or even 3 feet-- cutting your head pressure at the bottom) I see no reason why you could not put terry cloth on both sides of the panel, sticking it perpendicular to the building's side, like a flag.   I'd hang it from a sturdy arm, braced out from the building's side.   Let water trickle down both sides, spread out by the terry cloth.  Then inside the panel itself, plumbed with manifolds top & bottom, you'd run either air directly from the room, or water going to a water-to-air heat exchanger inside the house.

One of the nicer advantages of just using air inside the panel?  No worries about head pressure!   Simply use slightly pressurized air flowing down the panel.... I'm thinking of a positive-pressure air pump in this case, or a fan strong enough to raise the pressure to 5 or 10 psi (silicone glue ought to be able to handle that, no problems).   This would increase your airflow tremendously.   Then, simply dump the output from the collection manifold directly into the room-- it ought to be pretty cool, if you've cooled both sides of your panel.

I would mount the air pump outside, in a weatherized box, to get the noise away from the living quarters.    For the intake, I'd use filters too-- to keep dust and debris from clogging your panel's tiny passageways.

Since it'd be suction at that point, you're going to want a larger diameter intake than your exit pipe, to increase your efficiency, and to avoid a "sucking" sound.   At least 2 to 1, here:  if your output is 1"?  Your intake(s) ought to be 2" or more.

This is kinda fun to speculate about, but I may just fab up a small panel to see how effective it could be.   I much prefer piping water through the panel, then a water-to-air heat exchanger with a very quiet fan blowing the room's air past a used car radiator.   Several advantages to that:  you are moving the room's air mechanically, rather than trusting to convection.    You can use an oversized radiator, to greatly increase the efficiency.   If condensation becomes an issue, it's easy to make a collection trough at the bottom of the radiator-- and since it's from a car?  It's already weatherized, and the water won't hurt it at all.   If you mount this near a window, you can plumb your condensation directly outside, and let it dump onto the ground-- it's pretty much distilled water+dust from the room here, no worries.     If you insulate the water pipes, there's no worry about condensation dripping from them, either.

I'd use an open system-- i.e. one that's not pressurized or sealed, but is open to the atmosphere.  To get that, you plumb the bottom, collection manifold larger than needed, and vent it to the atmosphere.  That's where you install the aquarium pump, and add a make-up line, if you like.  (but if the vent is tiny, you would not lose much to evaporation at all, so manual top-up would probably work well.)

I'm thinking out to the side, for the cold water-loop pump, let the collection manifold dump into that box via gravity.   So the loop goes from the collection/pump box (which is at the coldest point, and needs insulating), to the bottom of the radiator, up through and out the top opening to the top of the panel-manifold, where it trickles through the multiple channels and is cooled, to be collected by the bottom manifold and dumped back into the pump-box.

This puts the coldest water at the bottom of the indoor radiator, where if there's any condensation, it will happen here, and drip down into the collection trough, to drain into either a (closed) bucket, or outside.  I would blow the forced-air fan only across the top 3/4 of the radiator, letting the bottom 1/4 be convection air (where any condensation will happen).  This way you don't blow air past the wet part of the radiator, possibly blowing droplets into the room.

[Aggie]

For an air manifold, I would probably just use Tuck tape.  I concur with using a slit plastic tube to 'clamp' the tubes.  In the context of an air-based unit, you could eliminate an intake manifold and just let the coroplast be the intake: Using long fins of coroplast, it'd be possible to clamp a stack of them at the tube end, and then insert dowelling or a wedge-shaped strip of wood a few inches up from that to create a fanned-out set of panels   \|/  to keep the whole thing compact.  Set in a breezy location, this should work well enough.

I'm currently speculating on what the wicking height of various fabrics (cotton terry, cotton flannel, any other cheap and 'fuzzy' materials) is; I may have to just run some physical tests.  It might be quite reasonable to set up the cloth-covered panels without a trickle-pump, and just use the capillary action of the fabric to do the lifting.  Even if it limited your panel height to 1 m or so, it would keep the project even more low-tech.  All you'd need to do is sit the bottom of the panel, including manifold if necessary, into a vessel of water. With the radiator system, you could even return your condensation drip tray to that same tub to recycle the water.


Oh, based on a little browsing, I suspect what I'm aiming for in cooling my room is the equivalent of a small 5000 BTU/h unit, which translates to about 2.1 L/h of net evaporation.   I certainly will not get the efficiency I need for that to be the actual number, since the unit will also cool the air surrounding it, and most of the cooling power is needed to chill the water in the pot and balance heat gain from the ambient air temperature.

I'm still unsure of how much of a factor ambient air temp is, as it's the heat from the air, combined with the lack of humidity, that causes the evaporation. If the entire surface is working as an evaporator, then it seems that the pot should not actually gain any heat from the surrounding environment; the surface of the pot will be the coldest point on the entire setup.

I have no clue of how to calculate the actual required evaporation rate, and in any case will have to tweak that number by experimenting with covers and wicks. 

[Bob in a quantum-state-of-faith]

Fiber cloth as used in filter media might be a good choice for wicking material-- it has the advantage of being polyester (or nylon) and therefore not subject to bacterial attack (unlike cotton).

Which ever you use, be prepared to change it fairly often, as mildew and/or mold may be a problem eventually. 

How about old T-shirt material? 



We don't have "tuck tape" here in the US, but I know what it is from watching DIY tv shows from Canada and other locales other that here.   That's a good choice for an ad-hoc rig, which is what you have there.

Let us know the results of your experiments-- it's a fascinating project, to me.

[Aggie]

Ever used poly for exercise wear?  The stuff isn't subject to bacterial attack, but it hosts them something fierce.  Cellulose based material suffer more from fungi than bacteria (mildew, etc) and are less prone to UV damage.  

I'd love to use wool/felt as it's microbe resistant, but it's cost-prohibitive and some of the other properties might not be ideal.

PET-based fleece (yes, old soda bottles) is cheap and readily available, but is hydrophobic and unlikely to wick well (it's also too insulating when wet).  Cotton does seem the best choice; for a single coroplast panel open at one end and with a manifold-to-pipe at the other, a t-shirt would be ideal. Cut the panel to size, base the manifold on an old plastic clothes hanger, and pull on the shirt!

I'll need to get my hands on some fabric samples and jimmy up a wicking-height test station. I can start with towels; the piles of the terrycloth will assist greatly in the evaporation rate I suspect; hopefully the microclimate created at the surface will allow the cloth that's touching the cooling panel/pot to shed heat effectively.


PS - no Tuck tape? Srsly? For non-load-bearing applications, it's far superior to even the venerable duct tape, IMHO.  Black-market export opportunity?

[The Meromorph]

PS - no Tuck tape? Srsly? For non-load-bearing applications, it's far superior to even the venerable duct tape, IMHO.  Black-market export opportunity?

http://www.amazon.com/gp/product/B004HE1B4O/ref=olp_product_details?ie=UTF8&me=&seller=
Direct Replacement. Some Canadians apparently use it in preference, as it is slightly thicker and uses stronger glue.

[Bob in a quantum-state-of-faith]

That's the stuff I've seen on TV, but never seen in reality.

[Aggie]

PS - no Tuck tape? Srsly? For non-load-bearing applications, it's far superior to even the venerable duct tape, IMHO.  Black-market export opportunity?

http://www.amazon.com/gp/product/B004HE1B4O/ref=olp_product_details?ie=UTF8&me=&seller=
Direct Replacement. Some Canadians apparently use it in preference, as it is slightly thicker and uses stronger glue.

This, I will believe.  The original is a little thin.  I do like the fact that the glue is very adhesive but 'weak' enough to break cleanly off many surfaces.