re: searching for comments on omsolar space heating system
17 may 2001
steven tjiang wrote:
>i plan to build a home of about 2500-3000 ft^2
>in the san francisco bay area
that's one of the easiest climates in the us for solar house heating.
nrel says it's 48.7 f (vs 30.4 in phila) on an average january day,
1,050 btu/ft^2 of sun falls a south wall (vs 1,000 in phila) and 680
falls on a horizontal surface. a 1-axis ew tracker can gather 2.5 kwh/m^2
(793 btu/ft^2) of beam sun. a fixed solar trough can do almost as well
>i want to include passive (or near passive) solar as its primary space
>heating source. i think 40% doesn't seem adequate for my needs...
...40% "on an average january day" (or was that "an average winter day"?)
means the house gets 60% of its heat from some other source on an average
day, eg propane. if there's no heat storage, it needs 100% from some other
source on a cloudy day. and how will you heat water for showers?
>this would require a larger backup system, a cost that would probably
>more than offset the cost of a storage system.
especially when we consider the yearly fuel bill as well as the initial
cost of the backup system. a rarely-used 5 kw electric resistance heater
could be fine for a near-100% solar-heated house.
>i find passive storage systems such as the one used in omsolar.com,
>earth walls, or water sealed in steel tanks much simpler and less
>likely to require maintenance than an active backup source.
the omsolar system isn't completely passive, and it looks expensive
and heavily hyped (with the om association and the om institute :-)
it has serious thermal disadvantages--the roof panels heat 100% outdoor
air, which is less efficient than recirculating indoor air, and we live
inside the heat battery, so the room temp varies from day to day and the
battery is less efficient than one that is thermally-isolated from the
living space. it is not compact, and it has no insulation and a low peak
temp and a small temp swing. and concrete costs more than water, and
it stores 1/3 as much heat by volume. you might fill the space under
the wood floor with 4" pvc pipes or 2 liter bottles full of water.
>since i plan essentially a new home, adding an insulated 6" to 12"
>concrete slab to the foundation doesn't seem like a big diversion.
but then you need another wood floor over that, for omsolar heating...
two floors, and a thicker slab...
>as for omsolar, i wonder how effectively it will work.
they mention 70%-solar heating, with omsimulations, and so on.
why not ask them for a guaranteed yearly max fuel bill?
>can a system like that really collect enough energy, transfer and store
>the heat in a slab, and then supply the house with heat until the sun
>comes up again.
maybe, but there are simpler and less expensive techniques, and making
the sun your primary source of space heating means storing heat for
a few cloudy days as well as over a 24-hour average day.
>the collectors will take up about 500 to 800 ft^2 of roof space
>and the slab spans the whole house or about 1000 ft^2.
say it's a 2,048 ft^2 32'x32' 2-story house with 8% of the floorspace
as r4 windows with 50% solar transmission, 4% on the south side, and
2% on east and west sides, with 0.5 air changes per hour. suppose
we use 600 kwh/month of electricity.
with r16 insulation (4" sips), the ceiling has 1024ft^2/r16 = 64 btu/h-f
of conductance. a concrete slab over 4" foam insulation would add about
(1024ft^2/(r16+r10(soil)) = 40. the 164 ft^2 of windows add 41 btu/h-f.
the walls add 1884ft^2/r16 = 118. the house volume is 16,384 ft^3, so 0.5
air changes per hour makes 136 cfm of air leakage which adds about 136,
for a total conductance of 399 btu/h-f. the house needs 24h(65-48.7)399
= 156k btu of heat on an average january day.
with 50% solar transmission, the south windows provide 43k btu, with
0.5x82ft^2x445 = 18k from the east and west. electrical use adds 600/30
= 20 kwh/day, ie 68k btu, so the house only needs 156k-43k-18k-68k = 27k
btu of extra heat on an average january day. (it wouldn't need any extra
heat with r25 6" sip walls.) it needs 156k-68k = 88k btu on a cloudy day.
we might keep the house 70 f during the day and 60 f at night, when it
needs about (60-48.7)399 = 4500 btu/h of heat, of which 68k/24h = 2.8k
btu/h comes from electrical usage. a 4"x1024ft^2 hydronic slab with
4"/12"x1024ft^2x25 = 8500 btu/f and a 1024x1.5 = 1500 btu/h-f conductance
to house air and a 6 hour time constant might be 60+(4500-2800)/1500
= 61 f at dawn. absorbing 43k+18k = 61k btu/day of solar heat from the
windows, its temp rises 61k/8500 = 7.2 f during the day. with no direct
sun on the slab, the house air would be about 61k/6h/1500 = 6.8 f warmer
than the slab, ie about 61+7.2+6.8 = 75 f at dusk. some direct sun on
the slab would lower that.
we might use a fixed reflective solar trough in an attic or over a south
deck to collect the extra 27k btu/day of heat. we might collect hot water
for showers, too, say 50k btu/day altogether, at 180 f. a linear foot of
a 12' deep x 12' tall parabolic reflector with a 3' focal length aimed at
the southern horizon could collect an average 0.8x12x793 = 7.6k btu/day of
sun over about 3 hours in a 180 f 3'x12' water trough below the north wall.
with an r1 cover, it might lose about 3h(180-48.7)3'x1.5btu/h-f-ft^2
= 0.6k, for a net gain of 7k btu, so we might make the reflector 50k/7k
= 7.1' long. how about 12', so we have some margin, and it can be a more
usable space for people? the materials cost for a structure like this
could be less than $500.
we might use 50k btu/day for showers, with a simple greywater heat
exchanger (a couple of plastic 55 gallon drums) so most of this heat
eventually becomes house heat. the remainder might keep a thermal
storage tank hot.
the house needs 5x72k = 375k btu for 5 cloudy days in a row. with an
80 f minimum water temp after 5 days, we need at least 375k/(180-80)
= 3750 pounds or 469 gallons of 180 f water cooling to 80 f, eg a 4'
plywood cube or a 2'x8'x4' tall box lined with a single layer of epdm
rubber, somewhere inside the house.
we might use a $75 shurflo 12v rv pump with a built-in pressure switch
to make hot water for showers. in lieu of the radiant floor, we might
pump 1 gpm of hot water through a $150 copper shw-2347 2'x2' magicaire
horizontal duct heat exchanger or an old auto radiator on cloudy days.