In John Michael Greer's entertaining post on the Energy Bulletin of April 6, 2011, Alternatives to Absurdity, he noted that passive solar opportunities in existing construction are more limited than in new. He went on to inform his readers that passive solar requires that "a good part of the south or south-east face of your house receives direct sunlight during at least a significant fraction of each winter, spring, or autumn day."
Around winter solstice is the crucial period to check thoroughly, for if the south wall is in sunlight for the critical hours between 9 a.m. and 3 p.m. one can be assured of sufficient sun in spring and fall as well. Old Sol will not soon alter his course through the heavens, and deciduous trees will spring into leaf and later let their leaves fall just when it will benefit the passive solar householder the most.
Where John Michael Greer's missive may have been overly enthusiastic, it seems from this humble writer's viewpoint, was in a bit of advice given to homeowners in the latter half of the essay. That is, I wish to add a note of caution, lest an army of intrepid Do-It-Yourselfers begins casting new Trombe walls in living rooms across America for passive solar heat storage.
Passive solar is a strategy to make the building itself into a heating appliance without the need for machines to move heat around. It involves both the collection of free heat from the sun, and the storage of that heat for gradual release overnight, at least. It is the storage function that is often neglected or poorly designed; without storage, there is little hope of energy savings.
In the halcyon era of the 1970's, research into alternative, appropriate, earth-friendly technology was going strong, and Edward Mazria's classic The Passive Solar Energy Book. was published by Rodale Press in 1979. It's a great place to start learning about passive solar.
A copy of this venerable handbook graces my bookshelves, as no doubt it does those of John Michael Greer. Yet since its publication, there has not been much evidence of serious study of the topic. A notable exception is the Passivhaus movement/standard of more recent European origin. Here in Canada there seems to be almost no understanding of the concept of passive solar space heating by the average individual.
Och! Canada! The True North Strong and Free! Comprising Alberta and some other extraction zones. Canadian
per-capita energy use exceeds even that of Americans. I still have a sense of shame so I won't reveal a figure in gigajoules or barrels of oil equivalents; you can discover it easily enough for yourself if you're curious.
Were we to look up a report on sustainability prepared by The Office of Energy Efficiency, Natural Resources Canada, (and available at www.canadianarchitect.com) with statistics for the decade 1990-1999, we would see that about 17% of Canada's energy use was accounted by residential buildings. Of this residential sub-total, space heating amounted to 59%, and hot water heating 22%. Space cooling was less than 1%. So we see that residential space heating in that decade was about 10% of the nation's energy use. Interestingly, the residential sector showed a very low increase in energy use over the decade, only 1.3%, less than one tenth the growth rate in commercial buildings, agriculture, industry, or transportation.
Still, saving a significant portion of this 10% of our societal energy use would certainly be moving in the right direction as 21st century civilization copes with a gradual decline in energy stocks. In the second part of this article, I'll suggest an inexpensive, low-tech heat storage improvement for a passive solar retrofit project that is an order of magnitude better than a Trombe wall. In this part, we'll cover some basics.
While there may have been a dearth of research on passive solar over the last few decades, construction products have seen significant improvements over the same period. These improvements were no doubt partly in response to more stringent building codes, and partly due to consumer preference. Nonetheless, a number of technical advances have come onto the market, contributing to that flattening of residential energy demand.
Windows have improved greatly in performance, both in thermal resistance and air-tightness. Low-emissivity films, argon and even krypton gas fills, improved thermal seals, tighter gaskets, and better-insulated frames are part of this picture. For specialized applications there are now spectrally-selective or multiple glazing options.
Looking at the wall envelope, air barriers have been a major advance, and the widespread use of foam sealants and newer, safer in-situ sprayed foam insulations have contributed to improved energy efficiency. Higher insulation standards for walls, floors, roofs, doors and windows have come with each code revision. On the mechanical side, high-efficiency furnaces are now the norm, ground source heat pumps for heating and cooling are common, and energy-recovery ventilation is standard in new homes and retrofits. Improved lighting and motors are readily available. Government rebate programs, software tools and information assist building owners in energy retrofits of existing stock as well as with design for efficiency gains in new buildings. Voluntary green building standards steadily gain in popularity with both developers and tenants, and broadly increase public awareness of solutions for sustainability in buildings.
None of this is breaking news, but it serves to point out that there has been some progress in building energy-efficiency since Mazria published in 1979, which makes the task of passive solar design just a bit less daunting, though undeniably it is even more urgent in 2011. As well, we now have the benefit of widespread computing power in the hands of designers, architects and engineers, indeed inventors of any sort, allowing more sophisticated modeling of building performance than ever before.
However, the DIY retrofitter may prefer to stick with a simpler, low tech approach to design, like The Passive Solar Energy Book. Let's imagine the project is a bungalow, 25' by 40', of 1000 square feet (sf) of footprint area. Let's say it's sited with good solar exposure on the south and with the long axis due east-west (see Mazria's Pattern #1, Building Location, pages 72-77, and Pattern #2, Building Shape and Orientation, pages 78-84. Check! Check!) Let's assume we're in Columbus, Ohio, on the 40th parallel, which had an average January temperature of 32.1 F back in 1979, per Appendix 4, Average Daily Temperatures (F) in North America.
Pattern #13, Sizing the Wall, pages 152-157 and #14, Wall Details, pages 158-171 suggest that the Trombe wall be about 500 sf in face area and at least 12" thick. Just to keep ourselves down to earth, 500 cubic feet of dense concrete weighs in at about 36 tons. Check? And the south wall of the house is only about 320 square feet. Doesn't check! Maybe we can make it shorter and thicker, though. An 18" thick Trombe wall with 320 sf of face area is only 4% less mass. Check. Write a memo to oneself about the 36 tons. Will an additional foundation wall be necessary to support the weight?
Just a minute, that photo on page 152 shows a guy in a 3 foot space between the Trombe wall and the window wall! That's 4 1/2 feet, including the concrete, times the 40 foot length of the house or about 180 square feet of floor area to give up out of an already snug 1000 sf., over one sixth of the floor space! And the whole south wall is windows!
What would that cost? Aren't the north and south walls supporting the roof? Some hefty lintels will be needed to get all those windows installed and still hold the roof up. This is getting complicated. The municipal building department will definitely require a building permit application complete with engineering drawings..
On page 110 of The Passive Solar Energy Book, in reference to thermal storage walls, the author states that "This system is easily added to the south wall of a space with a clear southern exposure." (Italics in original text.) Was he just a bit optimistic? The general comment doesn't seem to fit our specific example.
A few lines above on the same page we read that "The glass functions as a collecting surface only, and admits no natural light into a space." Check??
Before our DIY urges get completely out of control maybe what we need is a
REALITY CHECK!
Let's brush up, then, on our fundamentals of heat transfer, just to make sure of what we're doing, before we commit to that Trombe wall.
The project envisions winterizing the house to achieve the following thermal resistance values: R-20 walls, R-50 ceiling, (assume the R-12 slab-on-grade floor is ok as is) and new R-3 windows and doors. Caulking, sealing, and an air barrier under new siding are expected to improve air-tightness to one air change per hour (ACH of 1.0). This important factor assumes a design wind speed of 15 mph at the building exterior.
Let's assume 40 sf of window on the north wall, 40 sf of door and window, (each 20 sf) on the east, 120 sf of window and patio door on the south, and no windows facing west. Walls are 8' high.
Dear reader, it's a good exercise to draw a little 2 point perspective sketch at this point, from the south-east would be good, and to give ourselves a roof with a deep overhang like a Frank Lloyd Wright Prairie house, for summer shading of the big south windows. For those who prefer to do anything but math, it's also a good opportunity to go find some coloured pencils to render the sketch, because now it's time to do some arithmetic and a bit of algebra.
Equation 1. The basic equation of conductive heat transfer is Q=∆T•A•U•t
Q is the quantity of heat, in our case expressed in those familiar British Thermal Units, or BTU's; ∆T, or Delta T, or (T1-T2), is the difference in temperature in our case, from indoors to outdoors, in degrees Fahrenheit; A is area, in square feet, of the given assembly (wall, or ceiling, or window, etc.); U is conductance, (equal to the inverse of the sum of the resistances, the R-values). U is expressed in BTU's per hour-square foot-degree Fahrenheit, and t is time in hours.
For our purposes, the R-values are assumed to be totals, and since U=1/R, so we can write Equation 1 as:
Equation 2. Q=∆T•A•t/R
First question: What is the conductive heat loss through the west wall for a typical 1979 January day (24 hours) in Columbus, Ohio, assuming the interior of the house is maintained at 69.6°F?
Answer: Q = (69.6-32.1) °F x (8 x 25) sf x 24 hours x BTU/20 h-sf-°F
= 37.5x200x24/20= 9000 BTU
Second question: What is the heat loss through the floor for the same period, assuming an earth temperature of 54.6°F?
If you got 30,000 BTU, great. Now do a spreadsheet on your computer to calculate conductive heat losses for the house for the same sunny January day in Columbus, Ohio.
Done already? Did you get a total of 145,800 BTU? Great! I hope the neighbors like the new color scheme as well. Now these figures represent the losses by thermal conduction (radiation and convection losses are included). They are good approximations for our purpose, are fairly simple to calculate, and indeed are the norm in the building and home heating industries.
HEAT LOSS/SOLAR GAIN CALCULATION, IMPERIAL VALUES
|
HEAT GAIN SOUTH GLAZING
| ||||||||
40 N LATITUDE, COLOMBUS, OHIO
| |||||||||
JANUARY AVERAGE TEMPERATURE, F
|
32.1
| ||||||||
CLEAR DAY INSOLATION
|
INDOOR TEMPERATURE F
|
69.6
| |||||||
DAYTIME CALCULATION
|
24
|
HOUR
|
DELTA T
|
37.5
|
SLAB DELTA T
|
15
| |||
COMPONENT
|
AREA
|
R-VALUE
|
ONE DAY TOTAL BTU
| ||||||
WALLS
|
840
|
20
|
24
|
37.5
|
-37800
| ||||
ROOF
|
1000
|
50
|
24
|
37.5
|
-18000
| ||||
WINDOWS
|
200
|
3
|
24
|
37.5
|
-60000
| ||||
SLAB/FOUNDATION
|
1000
|
12
|
24
|
15
|
-30000
| ||||
SUBTOTAL
|
-145800
| ||||||||
There are two other important ways that buildings lose heat, one of which is by infiltration. This is the heat in the air that goes out the door, through cracks, and so on as cold drafts creep in.
When designing a heating system one estimates the volume of air lost over a period and multiply by ∆T (where the outside temperature is an extreme design temperature) and the heat capacity of dry air. Dry air because it's winter air, if it's heating loads we're calculating. It's only an approximation, as it assumes a steady 15 mph breeze underlying our ACH value. The formula is:
Equation 3. Q=∆T•V.•ACH•Hcap.t
where V is building volume in cubic feet, and Hcap of dry air is 0.018 BTU per cubic foot-degree F.
And this works out to 129,600 BTU/ day if we use our average January temperature of 32.1°F. It's nearly as large as the conductive loss. ACH, the air change rate, is difficult to estimate, and is very dependent on the quality of workmanship in caulking, sealing and air and vapor barrier installation. However, it can be measured after construction, by means of a door blower test. It is one factor that the builder has some control over, while the occupants have some control over ∆T, as does the gas company if you don't pay the bill.
While an ACH as low as 0.5 can be achieved by scrupulous attention to design and detail execution (as in Passivhaus construction), in an indoor environment ventilation is also necessary - fresh air from outside - and that air change per hour is welcome, but not always reliable, recalling that assumption of a steady 15 mph breeze.
Accordingly, many building codes also require an ERV (energy recovery ventilator) which will salvage about 2/3 of the heat in the exhaust stream, and can be easily retrofitted in any house. So if the code ventilation rate is 0.5 air changes per hour we can expect a design loss of 21,600 BTU per winter day by ventilation. So our total heat losses could be estimated at:
Conductive losses 145,800 (average January day temperature)
Infiltration losses 129,600 (design, average January day temperature)
Ventilation losses 21,600 (code, design, average January day temperature)
Total 297,000 BTU/day (average January day temperature)
We see in this summary that the total of infiltration & ventilation losses (using codified design parameters, average temperature and building spec's) exceeds the conductive loss (which used only average temperature and building spec's). Let's consider this a little more carefully, backing out the design assumptions. For instance, the average wind speed in January in Columbus is recorded as only 9.8 mph, not 15 mph. Pressure on the building is proportional to the square of wind speed, and infiltration is proportional to pressure, so in fact our infiltration at this average speed would only be about (9.8 x 9.8)/(15 x 15), or 43 % of the estimate.
A second consideration: why would anyone interested in saving heat run the ventilation continuously if there's so much infiltration air coming in any way? We only need 0.5 ACH for decent air quality, and we're getting 0.43 ACH from infiltration. Let's realistically just allow half the earlier infiltration figure which then allows for some ventilation when it's needed. This seems reasonable. So revised, realistic figures are:
Conductive losses 145,800 (average January day temperature)
Infiltration/Ventilation losses, say 64,800 (average January day temperature)
Total 210,600 BTU/day (average January day temperature)
The last little tweak to add, before we discuss passive solar in earnest, is an accounting for heat gains from occupancy. Let's say 3 people live here, they are each inside the house about 12 hours out of 24 during January, and also, the household uses only 12 kWh of electricity per day.
Firstly, the average person's metabolism puts out about 400 BTUH, so that's (3 x 12 x 400) 14,400 BTU per day in this scenario. Secondly, 1 kWh equates to pretty nearly 3412 BTU, and let's assume 90% of this lovely electricity degrades to sensible heat, so rounding off a bit, that's (12 x 3412 x 0.9) 36,850 BTU per day. These two figures will save the furnace 51,250 BTU of work, as our estimated heat loss drops to:
159,350 BTU per day, roughly 6,600 BTU per hour (BTUH).
We are just about ready now to think about passive solar. Referring once more to our passive solar handbook, we see Mr. Mazria provided an Appendix 6, Clear Day Solar Heat gain through Vertical Double Glazing at Various Orientations (in BTU/sq ft). Herein we find, for a clear January day at 40 North latitude, figures of 120 for north windows, 474 for east windows, and 1506 for south windows. There is a note below the appendix title advising that the values should be reduced by 6% to account for absorption losses. If our windows are not double clear, but have a low emissivity film included, (which is how we got that whole window R-value up to 3), we should adjust our figures downward again. This involves a unitless, but self-descriptive number ranging from 0 to 1, a factor called the Solar Heat Gain Coefficient, or SHGC, which typically can vary quite a lot, depending on the low-e film. We want it a bit to the high side, say 0.69, because we want to use that solar gain. Double clear glass will be higher at around 0.75 SHGC (but with a lower R-value). So tossing these all together (0.69 x 0.94/0.75) we could safely use about 86% of the values in Appendix 6.
And since we're fussy about accuracy for reasons which may become clearer as we proceed, let's not forget the width of the window frames themselves. The frames could amount to about 20% of the area, even for large windows, and the glass 80%. So if the east door is mostly door, we'll ignore it, and calculate our solar gains!
North 40 sf x 0.8 x 0.86 x 120 = 3303
East 20 sf x 0.8 x 0.86 x 474 = 6522
South 120 sf x 0.8 x 0.86 x 1506 = 124,335
Total 134,160 BTU (about 85% of our daily heat loss)
Great! The heating bills will go down by 85%. Right? Well, let's see.
Probably the house was already getting some free solar heat anyway. Maybe the windows weren't mostly on the south previously, but probably some were. In any event, now there is 134,160 BTU of solar heat gain, and it is almost all coming over a period of about six hours, from 9 am until 3 pm. The house needs about 6,600 BTUH, as we showed above, or 39,600 BTU in these six hours. So almost a third of the solar heat maintains the house temperature, giving the furnace a break for six hours. But there's 94,500 BTU that needs a place to go.
That's when we look around for some thermal mass. We're in luck, since we have that insulated floor of 3" concrete slab on 3"of foam insulation. That's about 250 cubic feet of concrete. And it's stained a nice rich dark brown, say, so that the vast majority of the incident photons happily turn into thermal energy (are absorbed). In the walls and ceiling, there's about another 60 cubic feet of drywall core with similar heat capacity.
We can solve Equation 4 for the concrete and gypsum mass and the 94,500 BTU to try to find out how much it will warm up, assuming all that excess heat can be spread around evenly and decides to take a nap in the slab! Concrete has a heat capacity of about 22.5 BTU/cf-°F.
Equation 4. Q=∆T•V•Hcap
Solving for ∆T, 94,500/((250+60) x 22.5) = 13.5 °F. The reality, of course, is that this won't happen near as ideally as our pretty equation suggests. Some other factors are in play.
Firstly, it's unlikely that the floor layout is a series of open U shaped rooms opening to the south, like one of Mike Reynolds' Earthships. So maybe the sunshine doesn't wash the entire slab perfectly evenly. And not all those photons that do hit the slab will decide to stay, as a few will bounce off and find some other resting place and a very few will bounce once or twice and escape the house entirely as visible or ultra-violet light and head out for parts unknown. Some will bang into something that is not a slab, which something might then decide to get up and go outside to cool off. And some of those that do become thermal energy in the slab might just get bored and decide to get up and mingle as infra-red radiation, which our low-e glass will mostly reflect back into the house. With all this pandemonium going on, everything in the house will warm up, including the air.
At about this point we might be tempted to call in a thermal physicist to get out the Boltzmann distribution curves and some dynamic modeling and arrive at some answers to 2 or 3 significant digits.
Alternatively, we could look around for empirical examples. I lived for a couple of years in a house somewhat like our example, that I had designed, except that it had a second floor, and was closer to 45 North latitude. It wasn't completely passive, in that the slab had radiant hydronic heating tubes embedded in it. We kept the air temperature, i.e. the thermostat, set at about 18° C, just under 65° F, in winter. But that slab was somewhat warmer as it was a giant radiator heating the ground floor level.
On a sunny winter day, the air temperature would go up to 25 °C, or 77 °F before noon, on the lower level, and slowly drop back down in the evening. So it looks like our air ∆T in Equation 2 will increase by about 12°F or one third, meaning our heat loss of 6,6000 BTUH will likely become more like 8,800 BTUH for most of the six hours, blowing away around 13,200 BTU.
As the house cools down to the 69.6°F thermostat set point, our ∆T slowly declines back to 37.5 °F from 49.5°F. We still have 81,000 BTU to account for, which now flows out of the mass of the slab and drywall. At an average of 7,700 BTU hourly heat loss, this cooling down process takes over 10 hours. We won't need the furnace in the evening, for a total of about 16 hours of free heat, from 9 am to 1 am. This is 66% of the day.
We've accomplished our passive solar project simply by putting most of our new efficient windows on the south wall. The winterizing work was crucial, as was the pre-existing insulated slab-on-grade floor. The Trombe wall
hasn't come into the picture. This is a good point in our discussion to add the solar data to our spreadsheet, and define day and night as separate calculations, for some future tweaking, we'll show a BTU/hour for each component for comparison purposes.
CHART 1: MINIMAL PASSIVE SOLAR PROJECT
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Note that the above spreadsheet does not show the 24,200 BTU lost due to overheating as just discussed.
For those who are skeptical about this very simple passive solar project, please change a few assumptions and check things out. For instance, on your spreadsheet, put the south windows on the north, and the north windows on the west or east (no south windows), and do the calculations again. The results are more accurate, as there is no additional loss due to overheating.
CHART 2: PUTTING THE WINDOWS ON THE NORTH SIDE
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
On a clear day we will collect only 29,500 BTU, mostly through the east windows, less than a quarter of our previous total. Thermal mass doesn't much matter, because there is no net daytime solar gain to store. Comparing the bottom lines, we see that our heating load (and therefore cost!) on a sunny January day in Columbus, Ohio, will be 160% higher if we get it backwards!
We'll talk about that substitute for the Trombe wall next time, because there are probably quite a few folks without an insulated slab-on-grade who are still interested in an elegant, economical and simple means of capturing a good deal of free winter heat from Old Sol, saving their hard-earned cash, and helping to save the climate, civilization, and the resources their grand children might very well appreciate and put to good use. Also, we'll discuss a few more windings on the road as we renew our passive solar quest. We'll look at average January performance, the value of insulating night blinds, and the outline of a technique for virtually eliminating overheating while making an order of magnitude leap in room temperature thermal storage.
But just for a moment, before we sign off, let's return to that character, mentioned in passing, who never really came on stage. The thermal physicist or engineer.
Picture this chap, (we assume a male character, for literary convenience only) around the year 1980, having spent a bit of time delving into the solar heating problem with little in the way of socially useful results, given that the political overlords of the research project dictated Business As Usual parameters - the building industry could not be expected to change, unthinkable, really, - and increased tax revenues were prerequisite to any application of useful discoveries that might be made - a project designed to fail, in short, though politically necessary as some citizens were demanding that the government take a serious look into energy efficiency.
In the end, a nonsensical demonstration experiment was fabricated to illustrate how pointless passive solar was, anyway; and so the funds having been appropriated, disbursed, and accounted for in a proper, bureaucratic manner, all manner of meetings were held, and the acceptably correct conclusions reached after a few drafts were circulated across the requisite number of desks, translations into the other Official Language drafted, circulated, etc., and final bilingual reports were submitted complete with proper references, charts, diagrams and formulae, for final approval from on high; a bland cover designed, a professional printing job got up in a rush, copies distributed, to be read by sub-functionaries, filed, then archived, and, in the fullness of time, completely forgotten.
Leaving the passive solar field empty, save for a few courageous souls, such as the Garbage Warrior, not quite so burdened by bureaucracy, yet fuelled by passion and funded by a few principled clients as maverick as him.
Back in1980, our scientist relaxes after winding up the project, wondering about the Next Big Thing and who might be funding it, what they're hoping for and when funding application deadlines fall due, with a well-deserved glass of good Scotch and ice in fevered fist. I ask you, dear reader, to consider, not what our scientist may do next, but rather, what is the ice doing at that moment?
Thanks for this Robert, very informative. I'm off to enjoy part 2!
ReplyDeleteShane