re: solar barn update -- one last question
3 oct 2004
bill kreamer wrote:
>take a look through the 2 layers of screen at the sky, and notice that you
>can see through it easily. there are not sufficient fibers (wires) to
>occlude (intercept) the light, so too great a percentage of incoming
>sunlight passes through the screen and hits the back wall of the cavity,
>which then reflects, and/or gets hot and radiates back out through the
got numbers? even a single layer of screen absorbs in both directions
and reduces reradiation and helps keep warmer air away from the glazing.
if 2 screens absorb 50% of the solar spectrum and a dark back wall absorbs
95% of that, only 2.5% of the sun goes back out the glazing. reradiation
sounds more serious than reflection, but the screens also intercept that.
>your quite low stagnation temperatures (155 and 172 degf) confirm that you
>are losing plenty of energy back out through the glazing.
glass won't pass much longwave heat energy (> 3 microns), although
it can intercept and re-radiate. what's the glass temperature?
>this is basically an absorber problem. the absorber wire material runs
>way too hot - it is probably around 250 degf; it doesn't have nearly
>enough contact surface area for good air-cooling.
a dark back wall can make more surface. why is the wire temp important?
>i suggest you increase not only the fiber face area (not counting empty
>space area between wires) by a factor of at least ten, and at the same time
>increase the bulk surface area of the absorber, by a factor of at least
>twenty, compared with screen. to do this you need to go to a fine-fiber
>blanket form of material. try black polyester dressmaker's felt, or at least
>four layers of nick's favorite, shade cloth.
how would adding more screens or blanket material decrease the temperature
of the wire or fabric facing the glass, if the screens or blankets don't
_conduct_ much heat from south to north? the sun hits the south layer first,
and that's where most of the reradiation comes from. an air-cooled wire with
nothing behind it would be cooler than a wire with more hot mesh behind it.
btw, plastic shadecloth shrinks about 30% when boiled. i wouldn't use it in
a collector if it could reach 212 f.
>an ideal air-cooled absorber should occlude (and your subjective visual
>judgement is good enough here) 80-90% of incident sunlight on the first
>pass, before any reflection off the back wall of the cavity. use of a
>fiber-blanket-type absorber will lower the absorber temperature, raise the
>exiting air temperature, and increase total heat removal.
have you tried this without a fan?
>additionally, line the backwall of the cavity with foil to reflect
>un-intercepted sunlight, and secondary radiation off the absorber, back
>toward the absorber. do this by using foil-faced foamboard (polyisocyanurate
>foam) as a backwall insulator.
sounds good for a freestanding box with cold outdoor air behind the back
wall, but not good for a collector with a warm house as the back wall.
>right now you are sending air in unimpeded "burbles" that free-fall upward
>through the screen, moving haphazardly from front to back and from back to
>front, in a rather random and oscillating manner, thus cooling some areas of
>the absorber well, while leaving other areas too warm. this uneven burbling
>flow can be detected by a thermal imaging camera, ind manifests as a
>shimmering "marble-cake-like" figuring that wanders across the face of the
>absorber. this is addressed in a fiber-blanket-type absorber, by a slightly
>increased flow resistance of the absorber.
>a slight front-to-back pressure differential encourages an even distribution
>of intake air across the face of the absorber, thus cooling it more evenly.
does that require 10 screens or a thick blanket?
>the resulting temperature evenness across the face of the absorber will
>cooperate with the reduction in temperature from better cooling to
>dramatically decrease the re-radiation losses of the absorber. you will get
>a slightly reduced flow, but a greatly increased temperature and total heat
>output. incidentally, running a small fan can nearly double the heat output.
what's the optimal number of screens, with no fan?