solar closet diagnostics SOLAR.KnowledgePublications.com

solar closet diagnostics
22 dec 1996
meanwhile, in arizona:

>well after adding the blower, water, and square feet of surface area, the
>closet didn't seem to improve at all. it's still 70 f to 72 f in the mornings.

hmmm.

>so now i'm thinking that the insulation and thermal bridging is the problem.

air leaks might be part of the problem as well. an 8' r20 cube would have a
thermal resistance of r20/(6x64ft^2) = 0.052 "ohms" if it were airtight. with
a 10 cfm leak, ie a 0.1 "ohm" shunt resistor, this drops about 35% to 0.034
"ohms." the natural "stack effect" airflow inside an 8' tall chimney with
2 a ft^2 holes at the top and bottom is about 16.6 a sqrt(8'dt), where dt
is the inside/outside farenheit temperature difference. if dt=70f-30f=40f,
10 cfm=16.6asqrt(320), so a couple of holes or cracks with area a = 0.034 ft^2
ie about 2" x 2.5", or 8' (one side) x 1/16", or 32' (the perimeter) x 1/64"
might let 10 cfm leak through the closet...

air leaks can be reduced by covering the inside of the box with large pieces
of plastic film, carefully sealed (taped, caulked, etc) at every edge. it's
also important to keep moisture out of the sunspace, as we are discovering
in the hostel cabin project, where moisture is coming up out of the ground
in the sunspace and condensing and freezing on the inside of the glazing,
reflecting sun out from the glazing, reducing the solar gain, and letting
more heat flow out of the glazing via condensation than convection.

this might be a good thing, if there were a steady supply of ground-heated
water vapor, and we were mainly concerned with keeping the temperature inside
the hostel sunspace to not much less than 32 f, which all this condensing and
freezing would tend to do. but the frost on the inside of the sunspace plastic
glazing melted at about 10 am yesterday, and the unfinished sunspace with no
insulation behind it heated up to about 75 f when it was 30 f outside (a good
sign), so we lost an hour of sun. it would help to seal the edges of the vapor
barrier under the gravel floor better. this is probably less of a problem in
dry arizona...

the energy-crafted home spec contains a diagram showing two 10x10' painted
walls, one with no holes, passing 1 cup of water per year via water vapor
diffustion, and one with a single 1" hole in the middle, passing 20 gallons
of water per year via air leakage. if this were an r20 wall, and that 160
pounds of water passed through the hole over a 160 day heating season, an 
extra 1000 btu/day or 42 btu/hr of latent heatflow would pass through the
wall, about 20% more than the (70f-30f)/0.2 = 200 btu/hr that would flow 
through the wall's basic thermal resistance (r20/100ft^2) in january, or
50% more than the 80 btu/hr that would flow through a perfect r50 strawbale
wall. also, that water might condense in the insulation and rot the wall...

all these should be diagnosable problems, like fixing a very simple car,
with some understanding and prediction of how the system should work, and
some measurements and perhaps some experiments, like moving the water
containers out of the closet and putting a 600 watt electrical heater with 
a fan inside overnight, with it all closed up, and measuring the inside and
outside air temps, to estimate the thermal resistance of the closet. a perfect
r20 8' cube would have a temperature difference of 600x3.4x0.052 = 106 f. an
alternative is to measure the self-cooling rate with the thermal mass inside.
or exhaust air from the closet through a large hole with a window fan on a
cold day and find incoming air leaks through cracks, etc, by feeling all over
the inside for cold drafts or moving a cigarette (or incense, for new age
solar enthusiasts) slowly around the inside edges, and watching the smoke.

another possible problem is heat that radiates back out through the closet
glazing. this wasn't a problem in last winter's system with 2 layers of black
aluminum window screen under polycarbonate glazing, but it could be more
serious with polyethylene glazing with no mesh absorber. steve baer says it's
easy to tell how efficient a solar collector is: the surface should look dark,
and the glazing should feel cool. one might measure the input/output closet
air temp difference and flow. how can slow airflow be measured accurately
and inexpensively? last winter, we used our fan pressure/volume curve and a
$50 0-0.25" h2o magnehelic air pressure meter to estimate airflow volume. we
were pleased that the pressure difference across the fan was only equivalent
to a column of water 0.05" tall, which corresponds closely to the specified
free airflow spec of 560 cfm from a 4c688 grainger fan. that pressure didn't
change at all when we opened the large door to the closet, so the water
containers were not constricting the airflow at all, compared to the airpath
with 4 90 degree turns.

> straw bales would probably be the best solution to this problem, but they
>are expensive this time of year...

there are other ways to prevent thermal bridging. for instance, we will use
fiberglass insulation in the hostel cabin, stapled between vertical 2x4 studs
on 2' centers, with some poly film over that and some horizontal 1x3's on 2'
centers over that, on the inside to act as nailers for the eventual drywall,
so instead of having the 8'x1.5" length of the studs (r3/1ft^2 = 3 "ohms")
as thermal shunts every 2', we will only have 5 little 1.5"x2.5" points of
contact (r4/0.13ft^2 = 31 "ohms") in parallel with the larger resistor
consisting of the 8' x 22.5" r20 insulation (r20/15ft^2 = 1.33 "ohms"),
increasing the equivalent parallel resistance (rp=r1xr2/(r1+r2)) by about
40% from 0.92 "ohms" to 1.28 "ohms."

something has to be done about the 72 ft^2 of single pane glazing in the
cabin too, which is a huge thermal shunt. all this will take time, and
winter is now half over... :-( 

>btw, i got tired of waiting for the closet to improve, so i started using
>it last night. (i was cold :)) so i drilled a couple of 4" holes through the
>wall from my room into the closet, at the top and bottom. my room which has
>been getting down to the low 50's a night, was a nice 60 f this morning when
>i got up! the closet dropped 2 f from its normal morning temp.

sounds nice. you are way ahead of us at the moment...

> i also got the grainger catalog, and will be looking for a little 4" fan
>when i get time.

a 4" fan sounds wimpy and noisy for this job. a 50 cfm fan will only transfer
about 500 btu/hr, about 150 watts of heat, with a air temp difference of 10 f.
and fan efficiency increases with diameter: grainger's 4" 1765 rpm 4c548 fan
uses 9 watts to move 55 cfm, with a max air temp spec of 120 f, while their
10" 1650 rpm 4c688 fan uses 36 watts to move 560 cfm, with an air temp spec of
149 f. how about using that $61 fan, with a $16 5c343 speed controller, so you
can vary the speed and power and noise, and enlarge the hole if you need to?

grainger's $139 3c691 56" 265 rpm 110 w ceiling fan moves 25,500 cfm :-)

>gotta go now...

good luck :-)

nick