A solar tank shed
nick pine
> My house heats itself when its sunny. I already have 150 sq ft
of glass...
With another layer of shrink wrap window film inside the single
sunspace glazing and some twinwall solar siding, you might collect
0.8x580-6h(80-32)1ft^2/R2 = 320 Btu/ft^2 instead of 0.9x580-6h(80-32)
= 176 and keep the house warm on an average vs sunny December day
in South Bend.
And the solar lawn box could have a lot less glazing. If the house
has a 200 Btu/h-F cloudy-day conductance and uses 300 kWh/month of
indoor electricity, it can stay 65 F with 696K Btu for 5 cloudy days.
A 4'x6'x8' plywood box with a $156 15'x20' folded EPDM liner
from http://www.pondliner.com with R30 insulation would have
a 208ft^2/R30 = 7 Btu/h-F conductance.
It could live in a 12'x12'x7' tall shed with 8'x12' of twinwall
polycarbonate south wall like this, viewed in a fixed font:
8'
epdmepdm epdmepdmepdm
S<-- t siding e e
t s n n
8't 6' s ~7' d 8' d
t tanks w tanktankw
twinwalltanks a tanktanka 4'
t 4' | tanks l tanktankl
ffffffffffffff lfffffffffffff
12' 12'
It could be heated with a 12'-wide x 8'-tall bladder collector
made from a folded 12'x16' piece of EPDM behind the twinwall with
fiberglass window screen inside to help spread out the water flow
and 1.5" of foil polyiso board underneath. The endwalls could be
twinwall, with no insulation in the roof or the north wall.
If the tank temp is T (F) and the south wall collects 0.8x717x8'x12'
= 55.1K Btu/day = 6h(T-29)96ft^2/R2 + 24h(T-29)7, T = 150 F. With
C = 4'x6'x8'x62.33 = 12K Btu/F, the tank can warm the house for
about (150-76)C/139.2K = 6.4 cloudy December days.
Nick
nick pine
polycarbonate south wall like this, viewed in a fixed font:
8'
epdmepdm epdmepdmepdm
S<-- t siding e e
t s n n
8't 6' s ~7' d 8' d
t tanks w tanktankw
twinwalltanks a tanktanka 4'
t 4' | tanks l tanktankl
ffffffffffffff lfffffffffffff
12' 12'
But then the water can't drain back into the tank...
A shed with 8'x16' of south twinwall with a 45 degree slope and
an 8'x11.3' footprint would cost less and work better, like this:
0123456789012345678901234567890123456789012345678901234567890123456789
t tttttttttttttttttt
t s e e
tbi s n n
S<-- tbi s d d
tbi s w w
/ tbi s | a a
16'tbi s 11.3' l l
/ tbinsulation s | l l
tbi s e e
tbladder siding n n
tbi 6' s d d
ti tanktanktanks wtanktanktanktankw
twinwall tanktanktanks atanktanktanktanka 4'
ti tanktanktanks ltanktanktanktankl
fffffffffffffffffffffffffff lffffffffffffffffl
11.3' 8'
It could be heated with an 8'-wide x 10' tall bladder collector
made from a folded 8'x20' piece of EPDM behind the twinwall with
fiberglass window screen inside to help spread out the water flow
and 1.5" of foil polyiso board underneath. The endwalls could be
twinwall, with no insulation in the north wall.
The south wall receives 580sin45+430cos45 = 714 Btu/ft^2 of sun on
a 29 F average December day with a 32 F daytime temp in South Bend,
with Qt = 0.8x714x8'x16' = 73.1K Btu for the twinwall including 6'
below the bladder and Qb = 0.8x714x8'x10' = 45.7K for the bladder.
If Qt = 6h(Ta-32)8'x16'/R2 + Qw and Qb = 6h(Tw-Ta)8'x10'x1.5 + Qw
and Qw = 24h(Tw-29)7, with Ta (F) air behind the twinwall and Tw (F)
water and a Qw water heat gain, then Qb = 720(Tw-Ta)+168(Tw-29), so
Ta = (888Tw-Qb-4672)/720 = 1.23Tw-70 and Qt = 384(1.23Tw-70-32)+Qw
= 472Tw-39.2K+168Tw-4872, so Tw = 183 and Ta = 155 F.
A differential controller could run the pump at night when/if
the water temp exceeds 170 F, to make the EPDM last longer.
Nick
nick pine
an 8'x11.3' footprint would cost less and work better, like this:
t s e e
tbi s n n
S<-- tbi s d d
tbi s w w
/ tbi s | a a
16'tbi s 11.3' l l
/ tbinsulation s | l l
tbi s e e
tbladder siding n n
tbi 6' s d d
ti tanktanktanks wtanktanktanktankw
twinwall tanktanktanks atanktanktanktanka 4'
ti tanktanktanks ltanktanktanktankl
fffffffffffffffffffffffffff lffffffffffffffffl
11.3' 8'
tf wttttttttttttttttw
S<-- t f a a
t foil l l
12't f | l l
twinwall f ~13' e e
t f | n n
t 6' f d covercollector d
ttanktanktankf wtanktanktanktankw
ttanktanktankf atanktanktanktanka 4'
ttanktanktankf ltanktanktanktankl
gggggggggggggg gggggggggggggggggggggggggground...
8' 12'
The tank could be 6'x10'x3' tall, with a 10'x16' folded EPDM
liner inside 1' of insulation. The horizontal cover collector
could be 2 10'x12' pieces of EPDM. The upper north wall could
be foil-faced foamboard to reflect sun down onto the cover.
The south and endwalls could be polycarbonate twinwall.
This would require less pumping power than the previous version,
with a simpler collector (no internal felt or screen). It would
have 192 vs 128 ft^2 of south glazing and 136 ft^2 of more useful
endwall glazing, which should make up for the fact that the north
wall will reflect some sun back out of the south glazing.
A larger version might have a circular swimming pool under
a geodesic cover. Cloudy day heat storage doesn't require much
glazing compared to average-day house heating, just lots of water
and insulation. Average-day house heating isn't hard either, with
enough airtightness and insulation and mass and solar siding.
Nick