[Greenbuilding] Swedish solar-heated village

Nick Pine nick_pine at verizon.net
Tue Nov 17 14:06:51 CST 2015


Kimmo writes:

>I’m an entrepreneur that is doing research the possibility to use some 
>architecture ideas to heat houses in Sweden similar
to what Soldiers Grove did back in 1979. We will of-course try to modernise 
the design but the basic concepts are the same
with the “solar attic”.

Soldier’s Grove attics required moving warm air down to the lower part of 
the building using fans or blowers and a motorized damper, and it’s hard to 
store solar heat from warm air. I figure cloudy days are like coin flips, so 
a building that can store enough heat for 1 cloudy day can be at most 50% 
solar heated, with a possible max 1-2^-N solar heating fraction if it can 
store enough solar heat for N cloudy days in a row, eg 1-2^-5 = 0.97 with 5 
days of storage. Most of the SG buildings were only 50% solar heated. Why 
stop there?

It seems to me that collecting enough heat to warm a building on an average 
day would be simpler with some passive solar heaters built into the south 
wall, eg 
http://www.builditsolar.com/Projects/SpaceHeating/solar_barn_project.htm

Where I live near Philadelphia, PA, 1000 Btu/ft^2 of sun falls on a south 
wall on an average 30 F January day, so a 4 foot x 8 foot vertical south air 
heater with US R2 twinwall polycarbonate glazing with 80% solar transmission 
would gain 0.8x32ft^2x1000 = 25.6K Btu/day. With a 70 F building and a T (F) 
exit air heater temp and a (70+T)/2 average air temp inside the heater and a 
6-hour solar collection day, the heater would lose about 
6h((70+T)/2-30)32ft^2/R2 = 48T+480 Btu/day. With a constant C cfm airflow, 
the collector would provide 6C(T-70) Btu/day of heat to the building... 2 1 
ft^2 vents with one-way plastic film flappers and an H = 8 foot height 
difference would make C = 16.6x1ft^2sqrt(8'(T+70)/2-70)) =  33.2sqrt(T-70) 
cfm, and 25.6K = 48T + 480 + 200(T-70)^1.5 makes 543 = T + 4.15(T-70)^1.5, 
ie T = 70+((543-T)/4.15)^(2/3). Plugging in T = 100 on the right makes T = 
92.4 on the left, then 92.7, then 92.7, with C = 158 cfm and a 
6x158(92.7-70) = 21.5K Btu/day heat gain for the building, which might have 
just enough thermal mass and insulation and airtightness to cool from 70 to 
60 F by dawn... 60 = 30+(70-30)e^(-18h/RC) makes RC 
= -18h/ln((60-30)/(70-30)) = 63 hours.

And given the present low cost of PVs and inverters, we might heat the 
building with Mitsubishi or Fujitsu mini-split heat pumps on cloudy days 
(they work with an outdoor air temp down to minus 13 F), powered by PVs, 
using the electrical grid for storage instead of a 5000 gallon hot water 
tank. How many peak watts of PV would be required for space heating alone, 
with a COP of 3? A simulation using hourly EPW weather data could help 
estimate this. 
http://apps1.eere.energy.gov/buildings/energyplus/cfm/weather_data3.cfm/region=6_europe_wmo_region_6/country=SWE/cname=Sweden

Then again, a solar village could have passive solar air heaters on each 
house and a large common underground heat storage tank with a geodesic 
transparent roof and a simple drainback hydronic collector on top of a 
floating insulated cover under the roof.

Nick 






More information about the Greenbuilding mailing list