Sneak Peak Video of the 
New Solar Hydrogen Home DVD
Coming SOON!

Download Over 100Meg of
FREE Hydrogen Video
Ride in the Famous H2 Geo
Click Here

re: help with passive solar hot water storage
30 may 1996
dan settles  wrote:
>(nick pine) wrote:
>> dan settles  wrote:
>> >(nick pine) wrote:

>okay, why not put the water heater inside the solar closet instead of
>upstairs?  the heater is now in the 130 f room, so you don't need to
>circulate water through it in order to keep it warm.

right, but that way, the water needs to be heated on the way in, in a
single pass. after it's in the water heater, it won't gain much heat.

>100 feet of 3/4 inch inner diameter polybutylene hot water pipe from home
>depot costs $60 and would have a surface area of ( 7/8) x >100 x 12
>= 3299 in^2 or 22.9 ft^2. how does this compare with the fin tube that you

10' of fin-tube pipe gains about 50 btu/hr per degree of air-water temperature
difference, and your pipe with 1/8" walls (?) might have a u-value of 10, with
a water-pipe-water interface, so it might gain 229 btu/hr-f. if the incoming
water only passes through it once, at, say 3 gpm, and the pipe has a volume
of 1/2 gallon, the water will only be in the pipe for about (1/2)/3 = 1/6 min
or 1/360 hour, so if it starts out at 60 f, say, it will gain at most
1/360(130-60)x229 = 44 btu, which will heat that 4 pounds of water to 70 f,
on the way into the water heater.

at that point, if the water heater is insulated, the 70 f water won't gain
much heat from the closet air. but say the water heater has no insulation...
say it's a bare galvanized tank... (which would make for a large backup water 
heating bill as the closet air temp drops to 80 f over 5 cloudy days.) if the
tank is bare, eg 6' tall x 1' diameter, it will have a volume of 4.7 ft^3,
holding 36 gallons or 292 lb of water, with a surface area of about 20 ft^2 
and a u-value of 1.5 btu/hr-f, so the tank will have an rc time constant of
292/(20x1.5) = 9.7 hours, so heating that 70 f water to 110 f (a nice shower
temp) in the 130 f room will take time t, where 110=130-(130-70)exp(-t/9.7),
ie 0.846 = -60exp(), or exp() = -0.0141 or t = -9.7 ln(-0.0141) = 9.6 hours. 
not exactly a fast recovery time for a water heater...

>i imagine the fin tube might make a more efficient air-water heat exchanger,

well, the water can move through it more than once... how long would it take
36 gallons of water to heat from 60 to 110 f with the fin-tube pipe? rc =
292/50 = 5.84 hours, so t = -5.84ln(-0.0141) = 5.8 hours. hmmm. let's try
20'... 2.92 hours, so t = 2.9 hours... still not very exciting. let's try a
fan coil unit (overkill), e. g. the $139 all-copper 2' x 2' shw 2347 duct
heat exchanger made by magicaire, attached to the $318 4c860 grainger fan
which moves 1840 cfm at 0.125" h20 of static pressure, and has an upper temp
limit of 311 f. this duct heat exchanger can transfer 45k btu/hour between
125 f water and 68 f air at 1400 cfm, with a 0.1" h20 pressure drop, ie
789 btu/hr-f. so now we have rc = 292/789 = 22 minutes and t = -22ln(0.0141)
= 22 minutes. but at that rate, we will run out of closet heat fast. we don't
need that much not water, on the average. so maybe we just need a bigger
tank, and, say 20' of fin-tube... or your plastic pipe setup with a small
circulator pump. but there's another bottleneck there, getting the heat
from the air into the drums. it helps if the drums are on top... 

>maybe in a water-water heat exchanger the plastic tubing might be a good,
>cheap alternative.  i agree that evaporation may occur where the tubing
>enters and exits the barrel, but i imagine that some type of gasket or
>pressure-type fitting could be made to minimize the water loss. 

perhaps it's really just a matter of taste... some fin-tube pipe with a
conventional water heater upstairs appeals to me...

>> >> now why not make that shallow box on the wall a bit bigger, say a lean-to
>> >> sunspace, 8' wide at the base, with a pebble floor and a railroad tie
>> >> foundation, and a picnic table, and use a bit more pvc pipe and keep the
>> >> sunspace temp at 80 f in january, to add some interesting daytime living
>> >> space and collect more heat for the house?
>> >
>> >this sounds nice, but i can't picture how this would work.
>> the usual way is some sort of bare solar collectors inside a sunspace,
>> eg zomeworks big fins, which are flat aluminum extrusions, about 4' long
>> and 8" wide, painted black on one side, that clip on to and are supported
>> by standard copper pipes running along the back side... then there's the
>> standard water heater upstairs and a thermosyphoning water loop from the
>> big fins to heat it, when the big fins are warmer than the water in the
>> water heater. so in this usual case, the fins and pipes are, say 130 f,
>> with the sun shining on them, but the sunspace is only 80 f.
>how much are the big fins?  do you have phone number for the company?

big fins cost a lot, $4/ft^2? because that's what zomeworks (505) 242-5354
charges for them... and i'm not even sure they make 'em anymore... but they
work very well. it wouldn't be too hard to make something up like this on
one's own. a big plate with a good bond to a 3/4" copper pipe behind it,
and some dark paint, or maybe some ovshinsky pvs glued to the front (or
electrodeposited on the front--hard to do at home.) bare water-type solar
collector absorber plates would work well in a sunspace, but they would
probably need a pump, since the water passages tend to be small.

or you could try making something attached to the north wall of the sunspace
with a single large piece of epdm rubber, a bladder, folded at the bottom,
with silicone caulk and a 2x4 sandwich to seal the side edges and a piece of
dark painted sheet metal to keep it from bellying out too much to the south.
sven tjernagel has site-built about 700 large collectors like this over the
last 20 years for commercial uses, for hotels and car washes, each being up
to 10' tall x 50' long, with a materials cost of about $1/ft^2...


I got ALL of these 85 Solar Panels for FREE and so can you.  Its in our Ebook

Site Meter