|
|
a solar hayloft
28 oct 1996
a 40x32'x26' tall horse barn near philadelphia faces southeast, with a
16' tall uninsulated gambrel roof over a concrete block wall 10' high:
--------------- --- .
| | . s .
----------------- sppppp.
dd . ds . ---
----------------- 26' ..................
| . . . . . . . | . . . . 10'
| . . . . . . . | .horses. . . <-- new sunspace
| . . . . . . . | . . . .
....................... ..............................
| 40' | | 32' | 12'| 16' |
the tyrannical young horsepersons ........d........... . . .
desire to live in the hayloft. . ssssss.t1 . .
. s ...... .
trees t1 and t2 are now gone, . s . . r .
cut off about 28" above the . s 12' ...... e .
ground, so tables can be . s .t2 . f .
screwed to the tops, . s ...... l p .
in the new sunspace. . s pp . . e o . se ->
. s pp ...... c n .
so is the large shrub s. . s . . t d .
. s ...... i .
but lilac bush l remains, . s .l . n .
though pruned a bit, so far. . s ...... g .
. s . . .
suppose we built a straw bale . s ...... .
or haybale wall down the middle, . ssssss. . .
about 14' tall x 40' long, about .......d............ . . .
180 bales, as an infill wall with s
a 2x4 framework, with another 50 bales
lining the endwalls, and another 25 bales to cover the 2' southeast kneewall,
and 640 ft^2 of new 6" fiberglass insulation for the rest of the roof, with
a poly film vapor barrier or foil underneath, 12' in from the se wall...
thermal storage might be a ceiling pond above the new living area,
12' wide x 40' long = 480 ft^2. how deep? the new living space
would have a total thermal conductivity of about
14x40/r50 + 2x14x12/2/r50 + 2x40/r50 + 16x40/r20 =
12, middle wall 4, endwalls 2, kneewall 32, roof =
50 btu/hr-f, requiring about 24hr(68f-26f)50 = 38k btu/day to stay warm on
an average 36 f december day in philadelphia, pa. on the order of 1/3 of a
cubic foot of wood per day, if a 4x4x8' cord contains the heat equivalent
of 100 gallons of oil.
we need 5 x 38k = 190 k btu of stored solar heat for 5 cloudy days in a row.
if the pond were 110 f on an average day, each cubic foot of water would
store about 64(110-80) = 2000 btu of usable space heat, so the pond needs
to be 190k/480/2000 = 0.2' deep, about 3". or we could cover the loft area
with waterbed matresses on top of an epdm rubber liner. or use 190k/30/500
= 13 55 gallon drums full of water or 150 5 gallon plastic drywall or paint
pails, with lids, in a more compact heat store configuration with a lower and
more desirable envelope surface to volume ratio. a ceiling fan with a
thermostat might bring down heat to the living area, as needed. the water
might also serve for hot showers, and as a sprinkler system, in case of fire.
the floor could use a vapor barrier underneath (but no insulation, perhaps)
to reduce le parfum des chevaux, and serve as a return path for living space
air that flows into the sunspace on a sunny december day. the sunspace might
be built with 9 curved beams on 5' centers, each made from 2 $2 16' 1x3s with
1x3 spacers every 2', held together with drywall or deck screws, attached to
a 2x4 under the eaves, and wedged between the wall and pressure treated 2x4s
spiked to the ground through holes with 40" of 1/2" reinforcing rod every 5'.
the sunspace frame might be covered with a single piece of greenhouse uv
polyethylene film, 40' long x 16' wide, costing $32, which might be rolled up
in the summer to make it last longer and keep the sunspace cooler. the east
endwall might be white painted poly or an insulated wall, since no sun would
enter it from the east in december. poly film wind fatigue might be reduced
by strapping it down in the winter with some stretched 1/8" nylon ropes over
the film between the curved wooden beams, following the eave to ground arc.
the sunspace would have a layer of greenhouse shadecloth spaced a few inches
away from the block wall, to keep the sun from hitting the block wall, so
the block wall stores less solar heat during the day, and wastes less heat
at night, back out through the glazing. this would also keep the barn cooler
in summertime. the upper part of the shadecloth might have another layer of
glazing or shadecloth a few inches to the south, to increase the air temp
above the average sunspace temp. air might ideally flow in a figure 8 during
the day, clockwise below and counterclockwise above, in sw elevation, from
(1) north of the shadecloth up through some ductlike rafter spaces insulated
with foamboard or foil, through the loft area (2), then back down under the
floor (3), then out into the sunspace (4), south of the shadecloth, then
north and sideways through the shadecloth (5), then up through the rafter
channels and back into the loft area again (1).
.
. s .
. s .
s <-- 2
. --- sppppppp.
s
. 8' ds .
ds
.....................^. --- sun
. . 3--> .1 .
. . .^ . 10'
. . .| 4 .
. horses . . . <-- new sunspace
. . .<-5 .
. . . .
.................................
an approximate calculation for loft water temp: the sunspace gains about
1000 btu/ft^2/day, ie about 400k btu/day, and loses about 6hr(90-36)400ft^2/r1
= 130k btu of heat through the se glazing, and an equal amount through the
block wall, perhaps (some of which might be insulated on the outside, under
the shadecloth), leaving a net gain of about 140k btu on an average december
day, of which 38k is needed to keep the living space warm, leaving about 100k
btu/day. if the main heat loss from the loft area is through the r20 roof,
then 24hr(t-36)640ft^2/r20 = 100k btu, where t is the average water temp,
so t = 36 + 100k/(24x640/20) = 130 f. lovely.
nick
none of us let reality interfere with our thinking --dan gottlieb
|
|
