solar house heating questions (was: solar greenhouse questions)
3 mar 2000
larry brasfield wrote:
>> when you live in a passive solar home, a certain amount of discomfort
>> is unavoidable--you're living inside a solar collector.
that's true for direct gain houses which store heat in floors and walls.
they need to change temperature in order to store heat: clear day swings
and cloudy day droops. smaller swings require more mass and insulation.
but a house can be passive without direct gain. is "passive" important?
>> even if temperature swings are controlled by an active venting system
>> and a good backup heating system (which will not be done perfectly),
>> the light levels can be harsh and unrelenting.
people find daylighting harsh with lots of south and few north windows.
and lots of windows lose lots of heat and feel cold at night, compared
to insulated house walls. and some solar houses lack privacy.
>i find that conclusion at odds with data published in
>"the passive solar energy book" by edward mazria.
that book now seems vague and cut-and-dried and obsolete, with
patterns and graphs and rules of thumb vs predictive arithmetic.
it's a codification of 70's mass and glass that was useful then,
but now completely misses a number of big points. for instance,
mazria goes on and on about trombe walls, and says they work
almost as well unvented as vented, but fails to mention that
they are miserably inefficient, either way, and they would be
cheaper and gather a lot more net heat with no thermal mass, and
they would lose a lot less heat at night and on cloudy days if
they had any insulation, ie if they were air heaters, more like
"solar siding." but serious solar houses still need heat storage.
>for the better designs, temperature swings are kept within
>a 5 degree f range during clear days.
a superinsulated house with no windows or thermal mass can have a zero
temperature swing if heated by the energy use of its occupants :-)
a 32' r40 cube only needs 24h(70f-40f)5x32^2/r40 = 92k btu/day in
january in seattle, a modest 810 kwh/month. with such an electric
bill, why add lots of mass and glass and temperature swings?
well, maybe it's fun, or you are crazy or electrically-frugal or
competitive, into solar house races, and want to build a sleek
high-performance living machine like an america's cup boat, or
one that is sooooo practical and cost-effective that it will be
reproduced thousands of times, just because that makes sense,
whether the owners are earthy-crunchy or not, or you'd like to
pretend that we've already run out of fossil fuels or clean air,
or you'd also like solar hot water for showers or thermal mass
for natural cooling--button the house up on a hot day, and
open it up at night.
>...to deal with what is sure to be excess solar gain at times,
>most of the biggest solar windows are equipped with automatic shades
>which also serve as nighttime insulation when heat is valuable.
but commercially-available automatic shades seem expensive and
heat-leaky and air-leaky, compared to well-insulated house walls.
bubblewalls may help. i'll get back to that.
>> we can build better, cheaper solar homes that are not passive or massive.
>> the drill:
>> deal with the building envelope, so the heat need is a minimum
>> -- super insulate, reduce glazing to viewing-only.
building codes have something to say about that, and curb appeal.
how many people would buy a 32' cube with no windows? (i might :-)
fluorescent lights, big-screen closed-circuit tvs for views, 10"
sip walls, and an air-air heat exchanger (or scandinavian-style
"breathing walls" with no insulation, in which natural stack-effect
driven outdoor air moves out of the walls and ceiling and into
the living space in an imperceptably slow draft which makes the
effective thermal resistance of those surfaces infinite...)
>> now that less is needed, some manual or automatic nighttime
>> window insulation (refer to nick for this) is a good option.
>once you have that, why not make the windows
>big enough to grab more solar gain, at least
>up to the limit of "harsh" lighting levels?
because it's hard to "get that." my favorite "movable insulation"
is an inexpensive low-thermal-mass sunspace with a couple of holes
in an insulated house wall to allow air circulation between a high-
thermal-mass living space and the sunspace during the day. at night,
the sunspace gets cold and loses little heat to the outdoors. unlike
a direct gain house, adding more (plastic film?) sunspace glazing
adds heat during the day, but doesn't increase the heat loss at night
or on cloudy days, when it is normally cool (although it can be
heated up quickly for occasional parties :-)
we might add thermal mass to that 32' cube, and reduce the wall
insulation, and change the south side to a 8'x32'x32' tall sunspace
with decks and stairways and hanging gardens and chairs and a few
windows into the living space.
the clear-day temperature swing is a) the daytime air-to-mass
temperature difference during solar "charging" (which decreases with
increasing mass surface and insulation) plus b) the temperature change
of the mass itself (which decreases with increasing thermal capacitance)
plus c) the nighttime air-to-mass temperature difference during the long
"discharge" (which is smaller and opposite to the daytime air-to-mass
temperature difference.) thick thermal masses (except water) also have
significant bulk thermal resistance, eg r1 for 5" of concrete, vs
r0.67 for a slowly-moving airfilm at the concrete surface.
if that 32' cube has $4/ft^2 6" r25 sip exterior walls and ceiling
and a living space envelope conductance of 4532ft^2/r25 = 174 btu/h-f,
it needs 24h(70f-40f)174 = 125k btu on an average january day in seattle
(vs about 500k that falls on the south wall.) with no internal energy
use, it needs to store about 125k/6h = 21k btu/h per day of solar heat.
two 24' n-s and 32'x 24' tall e-w interior solid concrete block bearing
walls would add about 22.4k btu/f of capacitance with 2,688 ft^2 of
surface, and 4,352 ft^2 of 1/2" drywall would add another 2k btu/f,
lots of surface but not much capacitance, with a temperature rise of
about 21k/2k = 10 f per hour, by itself.
the average day living-space air temperature swing would be about
21kbtu/hxr0.67/2,688ft^2 = 5.2 f for "charging" plus 125kbtu/24kbtu/f
= 5.2 f for the mass itself plus 125k/18hxr0.67/2688ft^2 = 1.7 f for
discharging (over 18 hours.) that's 12.1 f altogether, eg 72.1 at
dusk and 60 at dawn. we could reduce this with lots of high-thermal-
mass furnishings, eg 55 gallon accent drums or banco (adobe) sofas.
or we might make a thick south wall with hollow concrete blocks with
dark exterior foam (dri-vit or insul-aid) insulation, and circulate
warm sunspace air inside that wall during the day, and house air at
night, using thermostats with 2-watt honeywell damper motors or $12
automatic foundation vents with bimetallic springs that open louvers
as house air cools.
>> use a well-controlled low-cost (active, not passive) solar
>> thermal system to produce some daytime heat.
maybe a concentrating solar attic with a water trough
along the north wall and a big hot rainwater tank
under a hydronic floor...
>> at this point, expensive thermal mass storage (costing time,
>> energy, space, and money) ceases to be cost-effective.
concrete can cost more than water in dollars and embodied energy.
it stores less than half as much heat by volume than water.
it often costs more to remove than install.
>at the present price of lumber, concrete walls
>are not an extravagance even when their thermal
>mass is unneeded. witness the growing market
>in insulated foam concrete forming systems that
>insulate both sides of the wall and make the
>thermal mass unavailable. so arranging that a
>concrete wall functions also as a thermal mass
>is bound to be cost effective if the wall is
>valued for its other attributes the way current
>icf customers value their walls.
yes, but there may be better ways...
>bill quoted another source:
>> > the loudest gun blow definitely tossed the whole concept of mass
>> > out of the (south facing) window. mass in this part of the world
>> > means constant robbing of the air and the feet of any heat retained
>> > during precious minutes of sun light during the winter. mass is bad!
this sounds like direct gain again, psic-style sunspaces with lots
of mass and glass that stay lukewarm at best over 24 hours in cloudy
climates, vs low-thermal-mass sunspaces that are warm during the day
and cold at night. god bless the brick institute of america for their
support of the passive solar industry council and their guidelines
for passive solar house design...
although all the members of psic, especially the technical committee,
contributed to the financial and technical support of the guidelines,
several contributed far beyond the call of duty. stephen szoke, director
of national accounts, national concrete masonry association, chairman of
psic's board of directors during the development of the guildlines; and
james tann, brick institute of america, region 4, chairman of psic's
technical committee during the development of these guidelines...
gave unstintingly of their time, their expertise, and their enthusiasm.
>i think this statement evinces a misunderstanding of how
>thermal mass works in a correctly designed solar building.
i suspect those guys had a good understanding of how thermal mass works
in a correctly designed solar building :-) proper psic houses are
often less than 30% solar-heated. does psic also get fossil-fuel
>the mass absorbs heat that would otherwise go toward elevating
>inside temperature above the comfort zone, when solar gain is being
>taken. at other times, it keeps the temperature from falling
>as fast as it otherwise would.
well sure, but there are better ways. for instance, we might store heat
in some 100 f thermal mass that cools to 80 f over a few cloudy days.
with a larger temp swing, the same mass stores more heat. with a higher
temp, it might keep the house at a constant 70 f for a few cloudy days.
>as for robbing the feet of heat, comfortable floors are kept
>at temperatures at or near where a thermal mass does its work.
>i suppose that using a thick concrete floor as the thermal mass
>could lead at times to a colder floor than would be considered
>comfortable, but that is not the usual form that thermal mass takes.
maybe it should, with the living space on the second floor, and
the low-thermal-mass sunspace on the first floor, eg a walk-out
basement or glazed garage with a concrete ceiling with lots of
stalactites for good air-mass heat transfer, or lots of pvc pipes
full of water tucked up under the rafters.
>> > south glazing is ok, but needs to be done in moderation,
>> > since it also robs heat most of the time.
unless it's attached to a low-thermal-mass sunspace :-)