re: today's high, in the solar attic
27 oct 1995
nick pine wrote:
>... closer to 478 f, using ohm's law for heatflow:
>300 = (ta-30)/r0.67.
jeez, doesn't anyone look at the numbers in these postings?! it's been
almost a week now, and nobody's noticed the simple and glaring math error
above. nobody gets any points for finding this mistake. the maximum attic
air temperature above should be t = 0.67 x 300 + 30 = 230 f, not 478 f,
which is 300/0.67 + 30. one of the nice things about usenet is how people
so kindly offer corrections, when one makes a mistake, but that useful
process does not seem to work well when we use numbers. how depressing.
our solar closet house model is now overheating, because of some lack of
forsight on my part. yesterday when it was 70 f outside, the house was 104 f,
the 1,200 pounds of water in the solar closet was 81 f (still warming up
very slowly :-) and the sunspace was 154 f. so today i'm adding a small
cooling fan with a thermostat, inter alia, to bring in some cold winter air
during the day. this could be done with less energy, using the stack effect
and winter winds. we will be measuring and recording windspeeds shortly.
yesterday, the average power used in this 2' x 4' x 8' "house" was about
20 watts, counting power for the three fans and the modem and the data logger
itself, which has been taking readings every 2 minutes for the last 24 hours.
the data logger also measures its own power. the maximum solar intensity was
about 900 watts/m^2. the pyroheliometer sensitivity was insufficient to
record starlight, and we observed no significant solar radiation at night :-)
if we have some thermal mass up near the ceiling of the house, we can also do
night ventilation for summer cooling--low-energy air conditioning: keep the
house buttoned up during the day, then ventilate it at night, when the outside
air is cool, to cool the thermal mass of the house, which will very gradually
warm up over the next day. the 1994 van nostrand (?) book by architect/engineer
professor baruch givoni, _passive and low-energy cooling of buildings_,
indicates that this technique works well in many parts of the world.
steve baer estimates that there are very few days in philadelphia, even in
august, in which the night temperature fails to get down to 74 f. we can test
that weather hypothesis, and find a cumulative distribution of summer night
minimum temperatures, and experiment with this simple form of cooling, now
that we have this cooling fan. steve thinks this will require about 1 cfm/ft^2
of house, which is a lot of ventilation. we want to connect the fans to the
data trap soon, so we can control them in more interesting programmable ways.
perhaps some gentle reader can do a small theoretical investigation of this,
and post the results. how much thermal mass do we need in the house, eg some
containers of water near the top of the house, and how much surface area does
the thermal mass have to have, and how long does our 6.8 watt, 55 cfm fan have
to run each night, if the daily night min is 74 f, and the daily max is 94 f,
in order to keep the house at 80 f max, given that it has 72 ft^2 of r20 walls?
what will the daily temperature swing be?
by the way, contributions to this project are still welcome. if someone sends
us some money, we might give them the phone number and password for the house,
so they can call up the data trap and observe how things have been going,
using a modem with simple help menus. perhaps it can be accessed via a web
page. if someone sent us more money, they could do their own experiments
remotely, eg write some new real-time fan control algorithms and measure
the results. we could even put in a trombe wall, if somebody wanted to measure
its performance, or fill up the solar closet with cement blocks instead of
sealed containers of water, or put some fin tube pipe near the ceiling and
an insulated container of water on the roof for a solar closet water heating
experiment, using a warm water convective loop.
this would be a little bit like renting a part of the space shuttle, but much
less expensive, and administratively simpler :-) it is more like what howard
reichmuth, pe, is now doing by modem from hood river, oregon. howard also has
a data trap and a modem and a solar heating experiment underway somewhere in
hawaii. he probably has a project schedule that requires a site visit in early
february. i'll likely be going to lewiston, maine, instead :-)
so please feel free to send money, even tiny amounts, to:
nicholson l. pine system design and consulting
pine associates, ltd. (610) 489-0545
821 collegeville road fax: (610) 489-7057
collegeville, pa 19426 email: email@example.com
we will put it to good use.