These are initial comments from Nick on the first Deep Mesh Test
Very interesting results...
The concept description might add that we are trying to make a high temp air heater that uses "dynamic insulation," as in a Scandinavian breathing wall or a deer who can't smell you if you are downwind. Insulation with enough air flowing towards it cannot lose much heat upwind by convection... This is more than saying that the mesh keeps cool room air near the glazing or helps assure uniform flow with fewer hot spots over the collector surface.
But that depends on the amount of airflow and the amount of insulation, ie the static R-value. It's a matter of degree. I posted an equation for the dynamic metric U-value. Maybe 3 screens do not have enough static R-value to make a high convective R-value (eg US R10) at a 2 cfm/ft 2 airflow. At some point, we have to ask "where will all the solar heat go, as we reduce the airflow?" If the box is perfectly insulated, lower airflows will make higher temps behind the mesh, and the mesh will need to have a higher static R-value to prevent these higher air temps from heating the air to the south of the mesh by convection.
We might try more layers of screen or polyester felt instead of or behind window screen...
Lower airflows will also make warmer front surface mesh temps with more radiative heat transfer to the glazing. We might measure that surface mesh temp. With a poly film cover? Can we measure the air temp between the mesh and the glazing at various heights?
Why is the upper glazing so hot at low airflows? Maybe the upper part needs more layers of screen. Even if all the air between the mesh and the glazing is room temp air, the air behind the mesh will increase in temperature as it rises to 150 F?, since the finite collector airstream will be collecting more and more heat from behind the mesh.
It looks like the lower parts of the glazing are cool at low flows, so it seems to me that the increase in glazing temp is more related to a lack of enough dynamic R-value in the upper part of the collector than to a lot of radiative heat loss from the front surface of the mesh. I would think that radiative loss would not depend on height.
I want to explore that dynamic U-value equation more and try to discover how the static mesh R-value needs to increase with height, in order to reduce the convective heat loss until it is constant with height, in spite of the fact that the air temp behind the mesh increases with height. Maybe we need to supply room air between the mesh and the glazing at several heights instead of putting it all into the bottom.
Dynamic U-value Ud = VRhoaCa/(e^(VRhoaCaRs)-1) W/m 2-K, where
V is the air velocity in meters per second,
Rhoa is air density, 1.2 kg/m 3,
Ca is air's specific heat, 1000 J/(kg-K), and
Rs is the wall's static thermal resistance in m 2-K/W.
Using V = 1/3600 (1 meter per HOUR , and Rs = 5.7 m^2K/W (US R32),
Ud = 0.058 W/m 2, like a US R98 wall. V = 10 meters per hour makes
Ud = 1.7x10 -8 W/m^2K, ie a US R-value of 334 million
V = 2 fpm is about 0.01 m/s, which makes Ud = 12.2/(e^12.2Rs-1) W/m^2K.
US R10 is metric R1.76 or metric U0.568, and 0.568 = 12.2/(e^12.2Rs-1)
makes Rs = 0.26 metric or US R1.45, which sounds like more than
the static resistance of 3 layers of window screen.
Try felt, or 10 layers of window screen?
It seems to me that the 10 layers of window screen would have less rad loss to the glazing,
but if we can see some light through the felt it might work fine.
Unless it melts.
Greenhouse shadecloth might also melt. It shrinks about 30% in boiling water.
I'll explore further.
January 19, 2012