re: pv cooling for optimal output
28 jul 2002
a friend back from australia says "it's a great place, like america in
the 1800s! you think of some new idea, and everyone encourages you, vs
making lists of reasons why it can't or shouldn't be done."
solar guppy wrote:
>14gpm is allot of water , my 220 volt 12 amp well pump flat out makes only
>18gpm , yes it has about 40ft of head
and about 40 psi (92') on top of that. these $20 12 watt 11 gpm
bilge pumps are different. as i recall, they cut off at 3.3'...
>how much of the water is evaporated and would need to be replaced
>each day ??
making 10 kwh/day at 12% efficiency wastes 10kwh/0.12 = 83kwh of heat, ie
an upper limit of 284k btu, eg 284 pounds or 35 gallons or 4.4 ft^3 of
water, if evaporation is the only way to lose heat. we might lose some
by sky radiation or convection at night. or ground conduction. my old
hand-dug well has water 9' below the ground. we might need more water, if
we cool the pvs below ambient temp. a nuclear plant only wastes about
a gallon of water per kwh up the cooling tower. conserve water.
use nukes vs pvs.
>i live in florida and have a 48 panel kc120 system (www.djdigital.com) and
>have thought about spraying the panels , but with your calculations
>you are saying i would be in need of 42 gpm ??
that might maximize the net output.
>no way a 3 little 15 watt pump is going to push this amount of water.
god has spoken to you? :-)
>heck my solar hotwater panels (16 foot head) uses about 25 watts
>at about 2gph ...
heck. you mean gpm, nitwit.
>also , there is thermal resistance in that the evaporating water maybe
>cooling the top of the glass , but this doesn't transmit 100% to silicon
>cells on the other side of the glass.
with a thermal conductance about 0.6 btu/h-f-ft^2, ie a resistance of
1/(12x0.6) = r0.14 per inch, a 1/16" layer of glass is about r0.0087.
the slowly-moving water film conductance is about 60, so the other side
of a piece of glass with t (f) water on top in 250 btu/h-ft^2 of sun
might be t + 250x(0.0087+1/60) = t+6.4 f. not much warmer.
>1) i would suggest you check the actual pump power need , my actual numbers
>are off by a factor of 100 to your numbers.
you are wrong.
>you will have at a mimimum 2-3 feet of head summer to 5-6 feet fall/winter
>2) pump power needed to refill the "pond" since having and unlimited water
>source (pond) is unlikely
your 220 volt 12 amp 18 gpm well pump might run 2 minutes per day.
>3) adjust for the thermal resistance of the glass covering of the pv panels
>4) make you calculations include year round averaging , since in the cooler
>months the evaporative cooling will have no effect but
> the extra panels will be producing more power
you might make you calculations. cooling might help most of the year.
we might use 39 f water in wintertime.
>once the above is taking into account i would be interested to see the real
hey, try it!
>firstly, water isn't completely smooth and flat especially as it hits
>even the smallest of bumps (ie the wire connects to bring electrons
>back into the silicon), this will create a multitude of "funny" angles
>making the sun reflect off it more so than it would at incident, even
>though water is less optically dense than glass the imperfections in
>the surface would cause more of a reflection.
>second, water is not perfectly clear, it will drop sediment on your
>cells, not to mention probably causing the contacts ontop to corrode
unless the contacts are on the other side? sediment might not be
a big deal, if we use rainwater with a particulate filter.
>third, you have a glass covering over the cells, in which case the
>water isn't going to cool that much.
>finally the power required to pump water up to whatever height you
>need will most probably far outweigh any gain you'd get by cooling
>with a water system.
>to be perfectly honest if you must cool i'd say use a passive cooling
>system like heat sinks...
>although for a 120w panel, it'd cover a large area, making heat sinks
>probably not very cost effective.
>or if you want to go to some sort of active cooling, might want to think
>about small low voltage fans, it'll take a lot less juice than a water pump
>and would be a lot more efficent in that the above mention problems
>won't be there,
this ain't rocket science. then again, we ain't rocket scientists.
we need to tinker. how can we assure the water flows over a lot of
the panel surface, vs a tiny trickle down the middle? what's the right
pump and plumbing arrangement? how big a pond surface do we need to
dissipate the heat of the day? how deep, ie how much heat storage is
needed? can we make hot water for showers too? how about some solar
>depending upon the thickness of the cells you probably could cool them
>from the back, if you can make some sort of space between the framing
>and the cells you could do exactly what a fan on a cpu heat sink does,
>just circulate the air. of course a panel of that size most likely will
>be quite thick, and this won't be an option.
some people look for problems, and some people look for solutions.
pete prossen wrote:
>evaporation will leave behind crusty layers of calcium and magnesium
>compounds, so you'd have to spend some more money and energy to deionize
>that pond full of water.
rainwater might help. this thing is outdoors. we get about 45" of rain
per year where i live, ie about 0.123"/day, or 0.08 gallons/ft^2, so a
35/0.08 = 426 ft^2 roof might supply the water that's needed. wellwater
might require softening or an occasional acid wash for the panels. this
could be less of a problem with some concentration and a layer of plastic
film over the panels. it could be a non-problem, with a closed poly film
duct over the panels.
>but still you'd accumulate even thicker layers of algae, so you'd have
>chlorinate that pond like a swimming pool.
copper sulphate kills algae at 5 ppb. so does dryness and heat.
>of course, the continuously exposed area of water would collect dust that
>must be filtered to prevent mud deposits on the panel faces.
>don't forget to net the whole pv array so it isn't accessable as a bird
birds could be more of a problem with unwashed panels.
>i think it might be less expensive to just buy the extra pv panels.
>infinitely less maintenance headache as well.
>however, the calculation as an academic exercise is quite worthwhile. i
>wish there were a practical way to cool the panels.
wishing won't do it.
>here in north san diego county, where we have perfectly clear skies day
>after boring day, i can observe very significant energy production
>differences that can only be the result of temperature variations.
some also serve by merely observing.
sylvan butler adds:
>5) negative effect of deposits left behind by evaporating water
perhaps none, with rainwater.