Passive Solar Produces

[Thomas Anderson]

Passive solar still amazes me... it was in the 40s last night outside and a high of 59 this afternoon, but the house ranged from 72 to 77 with no active heating systems engaged. When I built my house, I added several redundant heating systems. Based on my research, my passive solar design should provide most of my heating requirements, but I never fully trusted that it would perform according to theory. But so far, it has. In my zone 6a environment, with the given insolation and insulation, passive solar was predicted to supply all of the BTUs needed but for four months of the year -- Dec through March (mostly Jan & Feb) -- when some supplementary heating would be required. Accounting for varying weather from year to year, that prediction has been pretty good. We'll still light up a fire for the occasional Halloween snow storm or April freeze to help warm the bones, but on average, our wood and propane usage is negligible outside of those four months. But on a cold September night, it's still an amazing thing to behold a warm house heated by nothing but solar.

 

[woodgeek]

Nice. The shoulder seasons, while not cold can really stack up the annual BTU consumption.

Do you have any numbers/models for your house's BTU load as a function of outside temp? It sounds like your house is modern-style passive solar--lots of insulation and small glass (per floor area) rather than a small insulation, large glass area and dedicated thermal mass. It would be interesting to estimate your solar fraction.

 

[Thomas Anderson]

I do have a spreadsheet I used for estimating this stuff. I'll see if I can find it.

It's definitely not a glass house. The primary collection area is about 4 x 18 sqft of picture windows in the first floor solarium which charge the 4-5" thick insulated concrete slab underlying the solarium, kitchen, foyer, and vestibule. Most of the rest of the window area is also on the south side. The first floor is recessed into the hillside to the north and few windows are on the north side on the second floor. The long side of the house is oriented to magnetic south, which is slightly east of solar south here. That allows for more morning heat collection and prevents afternoon overheating. The east/west walls are a significantly smaller surface area to prevent summer heating. Overhangs block all insolation from June through August. But by December, the sunlight reaches all the way to the back of the kitchen.

The walls and 2nd floor ceilings contain 5.5 inches of closed-cell spray foam plus radiant barrier. And there's an additional foot or two of blown-in cellulose in the attic plus another sheet of radiant barrier. There's also 3-4" of stone veneer which acts as an external thermal mass and actually creates a little micro-climate which tends to stay warmer than most of the rest of the ambient environment. It's a very tight and well-insulated house, so not a lot of BTUs are lost.

 

[Thomas Anderson]

Here's the passive solar analysis spreadsheet I've created. It won't allow me to attach an ods or xls file, so I did a screenshot of it.

 

[woodgeek]

So, if I understand, the ss says your insolation on the front of a 20 m^2 window is 33 MMBTU/yr, discounting the small input during the cooling season. If you are actually getting 75% into the space and the area is 18 m^2, then that is ~24 MMBTU, which would be quite impressive.

The demand numbers seem kinda low....IIRC you burn some wood and 300 gal/yr of propane? Still your solar fraction could be 30-40%??

House looks large....what is your BTU/sqftHDD??

 

[Thomas Anderson]

Something like that... between 30 and 40 MMBTU/yr during heating season according to the PVWATTS calculations. I would say that probably 200 gal/yr of propane is burned during heating season. 100 gal for cooking and hot water during the off season. That's about 18-20 MMBTU/yr for heating. And I burn about a cord of wood, give or take, which is between 15 and 20 MMBTU/yr. So the solar fraction might be 43-57% assuming the PVWATTS insolation number for a 90 degree vertical collector at my latitude is 100% accurate and derated for shading as per the spreadsheet. Although it could actually be more than that, because some of the propane heat goes to heating my power shed and some of the wood heat goes up the chimney.

The demand (heat loss) is calculated at about 32.5 MMBTU/yr. That may indeed be quite a bit low, although I reduced my R-values by about 20% to try to account for window surface area, air leakage, ventilation, and doors being opened, but that's probably not enough. I really need a more complex calculation for heat loss which incorporates these other factors. The spreadsheet basically assumes none of that except for that 20% derating of R-value. Also, for the last few years, we were also heating some uninsulated spaces which have since been insulated this year. So in past years, the heat loss was greater than what is calculated on this spreadsheet. Hopefully, we're closer to this number this year.

Based on the demand calculated on the spreadsheet (which will hopefully be closer to reality going forward), the BTU/sqft/HDD is about 1.6. Based on the BTUs supplied in past years, including solar+propane+wood, it's about 3. That only includes living space, and not the garage, which is maintained above freezing. If you exclude propane and wood losses as mentioned above (let's say 20%), it's about 2.8 BTU/sqft/HDD. If you include the garage, it's closer to 2.2.

That doesn't include any passive annual heat storage gained from/lost to the ground or any contribution from electrical devices or human occupants either. Also 20 m^2 is a rough calculation of window surface area on the south-facing side and doesn't include impingement from window frame elements and also excludes east and west facing windows, which may contribute to solar gain. It also doesn't include solar gain from the stone face of the house, which can easily heat up to 80 degrees in the winter on a sunny day. So, this is a fuzzy estimation at best. But the purpose is to give a general idea of solar heat gain versus heat loss with a wide margin of error. That margin was designed to be filled in with redundant heating systems, including solar thermal (eventually, when I get it fully hooked up), propane, and wood.

So far, the margin of error is much less than I expected it to be. My BIS Ultima wood stove could supply 1.3 MMBTU/day by itself if needed, and during the design of the house, I was afraid it might be. Thankfully, we only need to fire it for a few hours during the coldest days, and then we often exceed the 65 degrees this spreadsheet assumes for the inside temp. I'll check this year to see when the first day of propane heating starts and adjust the heat loss calculation to better coincide with that.

 

[jebatty]

Your experience, which isn't unique but also isn't well appreciated, makes me wonder why anyone who builds a new house would not site and build for passive solar. Additional cost, if any, is nominal. Energy for heat and natural light are free every day and are substantial, as you have learned.

Our single story house, with plenty of attic insulation, and with a mix of 2x4 and 2x6 fiberglass insulated walls, 1500 sq ft, faces SW overlooking a lake with about 24' of 5' windows installed for the view. We also have 4' eaves. Besides the view, the eaves along with the high summer sun provide complete shading pretty much from May to August and keep out the heat when it is not needed. And then beginning in September through April the low sun angle, plus reflection off the lake, gives us passive solar which will heat the house during sunny days, even when outside temps drop to -20F and lower. Our other heat source is a small wood stove in the living room. The house originally had electric baseboard, which we keep for backup, and based on wattage heat loss is not greater than 40,000 btuh.

Just like you said, when the sun shines inside house temp rises to the low 70's and no other heat is needed. We don't have a concrete heat dump like you, but all that solar is absorbed into the walls, wood floors, carpet, furniture, etc. just the same. Our total electric bill for heating is less than $200/year, and stove wood usage typically is 4-5 cords of aspen (about 3 cords of oak equivalent). Passive solar is a no-brainer.

 

[woodgeek]

I generally trust usage numbers more than I trust area/r-value estimates. But your usage numbers are impressive...say HHI=3 and a ~50% solar fraction. You might want to add some SHGC numbers to your passive solar sheet...I'm no expert but I thought that even 'solar windows' will reflect and absorb a good part of the solar spectrum. But worst case I would guess that would pull you down to 35-40% solar fraction. Did you design the house yourself, or did you have some models to work from??

This HHI=2-3 range is what the BSC folks call a 'pretty good house', not as a prejorative, but as a mainstream-able alternative to the passivhouse, which of course comes in around HHI=1. They argue that it is easier and cheaper to get to 'net-zero energy' with HHI=2-3 and a big PV array, than a passivhaus and a smaller array. So they would say you are in a sweet spot. Those guys would build a similar house, with energy eff appliances (like you have for off-grid), then heat it with minisplits and a rather modest amount of elec, and then get to 'net-zero' residential energy with a slightly oversized grid-tie PV array.

Anyway, I respect that you have made another choice, off-grid for resilience rather than net-zero for carbon footprint minimzation. Still, your carbon footprint for residential energy is likely smaller than 98% of existing N American homes, and your HERS score would prob be ~15-20/100, or 15-20% of the site energy usage of a typical new house??

 

[Seasoned Oak]

Passive solar still amazes me.. But on a cold September night, it's still an amazing thing to behold a warm house heated by nothing but solar.

I accomplished about the same thing enclosing a south facing porch with about 150SF of windows. THe house is about 100 years old and only about 65% insulated.
IT has to get into the 30s at night and under 55 during the day for it to drop below 70 in the house. Of course overcast days and rainy days dont produce much heat.
I bought the windows 6@ 3ft by 8ft at a lumber auction so i had only about $500 in the whole job.
Saved that back the first winter.

 

[Thomas Anderson]

Indeed it makes no sense that homes today are built without this in mind. The cost is negligible. The biggest decision is orientation, which at most might cause the driveway to be a tiny bit longer. But I continue to see new homes going up with the primary window area facing the street instead of the sun. There are better ways to get curb appeal! The cost of that choice over the life of the house is gargantuan.

I read many accounts of passive solar homes before designing mine. Like I said, I was a little skeptical that it could be that effective and was fully prepared with low hopes. I just couldn't believe you could get so much for so little, or why wasn't everyone doing it? But the principles seemed sound and I gave it the benefit of the doubt, and I've been pleasantly surprised that the accounts were not embellished.

I designed my house from scatch to incorporate passive solar and high efficiency while avoiding the modern style elements and opting instead for a neo-gothic or romanesque style. The orientation, shape, window location, and materials were chosen with passive solar in mind. The only thing I would do differently is build larger overhangs. While mine work well to prevent overheating in mid-summer, we get a bit more sun than desired in late August, early September. But that's manageable with minimal air conditioning on sunny days and bringing in cool air at night. I'm also planning to erect arbors over some windows and use automated shades on others. The shading calculations on my spreadsheet were added long after framing. I wish I had them from the beginning.

I think the PVWATTS program incorporates reflections to give net insolation, but I could be mistaken. Either way, reflections would have decreasing impact the closer the incident sunlight gets to perpendicular. That is, they wouldn't affect deep winter numbers as much as the other seasons, so they would tend to work in concert with passive solar.

 

[woodgeek]

What windows did you pick to get high SHGC and low U-values....

 

[Thomas Anderson]

What windows did you pick to get high SHGC and low U-values....

I'm using mostly Andersen windows, but I didn't choose those with the absolute best performance because window performance is fairly irrelevant in the grand scheme of things. Whether they are R-1 or R-2 or R-4 (U-1, U-0.5, U-0.25) doesn't make a great deal of difference, although the difference in costs can be significant. The main point being that windows will always be a great source of heat loss, no matter if they are high performance or not. I chose mostly mid-grade low-e wood-framed windows and doors between U-0.33 and U-0.6, and I installed them with proper foaming and caulking to prevent air leaks. More important is window placement and window coverings. Most of my windows are on the south face of the house. The whole north face (and most of the east and west) of the first floor is below grade. On the second floor, north, east, and west facing windows are minimized. That provides both optimal winter heat gain as well as summer shading. I don't have all of my optimal window coverings in place yet, but my intention is to put automated blinds or curtains on the main picture windows and insulated curtains on all of the other windows. This way we can have significantly better insulation and radiant barriers over those surfaces and open them selectively for insolation or daylighting or view.

 

[jebatty]

... because window performance is fairly irrelevant in the grand scheme of things. Whether they are R-1 or R-2 or R-4 (U-1, U-0.5, U-0.25) doesn't make a great deal of difference, although the difference in costs can be significant.

This may be technically true, but ... beginning in 1992 and ending in about 2002 we replaced all the windows in our house with a Hurd window rated R8 (probably center of glass). Specs showed net solar gain even on the north side of a house, coatings to block infrared both directions (keeps outside summer heat out and inside winter heat in); 2 panes of glass with 2 panes of plastic film between; wood frames, alum clad exterior ... and much more expensive than the best Anderson had to offer at the time.

These windows have performed flawlessly and with the passive solar gain mentioned in my post above. Plus, no condensation (even with the windows over the kitchen sink and dish washing), no cold air wash, warm to the touch on the inside during winter, no maintenance. My wife can sit right next to the 14' x 5' expanse of glass in our living room during the coldest winter day, -35F, howling winds, blowing snow, and no draft off the windows at all. Comfort performance can be as important as technical performance, because some of us know too well that a cold wife isn't good for a blissful existence.

The lack of condensation has meant that the interior wood frames, which I stained and varnished after install, look about as good today as when first done; and the windows block nearly all UV, meaning no fading of furniture next to the windows.

Window performance in all aspects can be very important. We live in the woods off a lake, and total window perimeter in our house is about 70' for house perimeter of 160', with full views of our natural surroundings in all rooms of our house -- in fact we even have no window coverings, shades, blinds, etc. And energy performance must be pretty good because the whole house is heated with a small wood stove in the living room.

 

[woodgeek]

Yeah. I worked out that an uncoated single pane window, SHGC=0.8, R=U=1 would just about break even on net winter BTUs, losing as much as it collected, with my HDD and insolation. That would suggest that a low-e window, with SHGC of 0.5 and U=0.5 would also break even over the season, while providing significant gain in the shoulder seasons. (of course, window insulation would mitigate losses). So, it seems that best practice is to get low-e windows specifically engineered for solar gain, with SHGC>0.7 and U-values at 0.3 or lower. With higher U values insulated blinds are a must, with lower-U values, there isn't much point.

jebatty...do you know your SHGC?

 

[Thomas Anderson]

It sounds like windows make up a very large proportion of your wall area, so it would make sense that window efficiency would be important for you. My windows only make up approximately 7% of my wall area, and my walls are R-40+. Making 7% of my wall surface area R-4 vs R-2 won't make much difference overall, especially if I put R-7+ curtains over them. But if you have 30-40% window area, window efficiency makes a lot of sense.

I searched for Hurd performance data and the best window I could find which has actually been tested (versus advertising claims) has a U-factor of 0.21, which is equivalent to about R-4.76. Do you have verified specifications for the R-8 figure?

Solar gain in the winter through properly oriented mid-grade windows is many times greater than any losses. Although, having a thermal mass to absorb the gain might be important in realizing it. Our floors radiate the warmth for many hours after the sun sets. Let's do a little math:

In my climate in January, PVWATTS indicates insolation at 869 btu/sqft/day (oriented SSE).

Single pane window, no low-e coating (R-1)...
Loss = 1 sqft * 41 / 1 (sqft**hr/btu) = 41 btu/hr = 984 btu/day
(net loss without window coverings at night)

Double pane window, no low-e coating (R-2)...
Loss = 1 sqft * 41 / 2 (sqft**hr/btu) = 20.5 btu/hr = 492 btu/day
(net heat gain of 77%)

Double pane window, with low-e coating (R-3)...
Loss = 1 sqft * 41 / 3 (sqft**hr/btu) = 13.67 btu/hr = 328 btu/day
(net heat gain of 165%)

Hurd triple-pane, very good window (R-4.76)...
Loss = 1 sqft * 41 / 4.76 (sqft**hr/btu) = 8.61 btu/hr = 207 btu/day
(net heat gain of 320%)

Hurd super window (R-8)...
Loss = 1 sqft * 41 / 8 (sqft**hr/btu) = 5.125 btu/hr = 123 btu/day
(net heat gain of 607%)

Single pane window, no low-e coating (R-1) + R-7 curtains at night...
Loss = (1 sqft * 32 / 1 (sqft**hr/btu) = 32 btu/hr) * 8 hr
+(1 sqft * 50 / 8 (sqft**hr/btu) = 6.25 btu/hr) * 16 hr = 356 btu/day
(net heat gain of 144%)

Double pane window, with low-e coating (R-3) + R-7 curtains at night...
Loss = (1 sqft * 32 / 2 (sqft**hr/btu) = 16 btu/hr) * 8 hr
+(1 sqft * 50 / 10 (sqft**hr/btu) = 5 btu/hr) * 16 hr = 208 btu/day
(net heat gain of 318%)

So while there are clearly differences between windows, there's a diminishing return for higher efficiency and so it may not be cost-effective in many instances to get the best possible windows. I think my and jebatty's circumstances are probably on the opposite ends of the spectrum given our relative proportion of window surface on our houses. I found it more effective for me to spend my money elsewhere than on the top-of-the-line windows. I'm able to get considerable winter heat gain with what I have.

 

[woodgeek]

We agree completely about the losses....but many/most low-e coatings also reject incoming energy (much of it near-IR). Indeed, that is a selling point in cooling dominated climates. A common figure for a low-e window (not designed for passive solar) might be a transmittance of only 30-40% averaged over the solar spectrum. That can make an R-2, low-e window a net loser (w/o a curtain, in January).

Bottom line: one must be careful shopping for windows for passive solar.

 

[jharkin]

I just came by this. Great results. Passive solar is great and this is another one of those "whats old is new again" things. Our ancestors knew that when you build a house, siting it for southern exposure is key to help with winter heating. Along with blinds and placing windows for cross ventilation its one of those things we lost along the way.

My place is 200 years old, even with added insulation is not even up to modern minimum code.... But... I have near perfect southern exposure with no shading. Last week with daytime highs around 70 and overnight lows dipping close to 40 I never once had the inside temperature drop below 68. The only added heat was people and appliances. We rarely turn the heat on before Columbus day and often don't need overnight fires till mid-December.

 

[Thomas Anderson]

We agree completely about the losses....but many/most low-e coatings also reject incoming energy (much of it near-IR). Indeed, that is a selling point in cooling dominated climates. A common figure for a low-e window (not designed for passive solar) might be a transmittance of only 30-40% averaged over the solar spectrum. That can make an R-2, low-e window a net loser (w/o a curtain, in January).

Bottom line: one must be careful shopping for windows for passive solar.

I wonder if PVWATTS takes into account all wavelengths or only those which contribute to photovoltaic. My main picture window collectors are not low-e. They're just double pane, R-2, U-0.5. I really want to get automated blinds or curtains on them to cut back on nighttime loss but still get morning sunlight. However, we have not noticed huge losses through them as-is. As I've described above, our passive solar component is quite high. I think all of the rest of my windows are low-e. The four picture windows I didn't choose to use low-e for exactly the reason you described, blocking of some wavelengths. However, I'm not convinced low-e blocks too much in regard to passive solar because it's mostly the shorter wavelengths which absorb into interior surfaces you want to capture. Eventually, I plan to run some tests on them and on low-e glass to ascertain the difference in solar gain. If it doesn't block much, I'm going to put low-e film on them to help prevent losses. There's a combination anti-shatter/anti-burglar and low-e film that I've been looking at.

I'm also going to dig a reflecting pond in the front of the house to increase insolation. Along the bottom of the pond, I'll run earth tubes for tempering ventilation replacement air.

 

[woodgeek]

Yar. There is not a physical limit to low-e. To keep heat in you want to block long-wave IR, while to let solar in you need to pass near IR and visible. Perfectly doable optically.

But many low-e windows are sold to reject solar heat, and thus near IR to reduce AC. Presumably Anderson could give you the SHGC spec you want.

 

[jebatty]

jebatty...do you know your SHGC?

The windows I have are not made anymore, and I don't recall that SHGC was provided at the time I bought them. BTW, the windows are Hurd Insol-8. I learned that problems were experienced with some of these windows, but as mentioned, ours are 10-20 years old and no problems. I did find this info on a search:

http://www.repp.org/discussion/greenbuilding/200203/msg00191.html

The best performing (in our cold cloudy upstate NY climate) wood-framed window I could find was from Hurd (<http://www.hurd.com>). Their Insol-8 glazing system uses 2 sheets of lightweight Heat Mirror film between the 2 panes of glass, giving 3 argon-filled spaces (effectively a quad-pane window). I used the free RESFEN software to compare performance in a model similar to our house plan. The Hurd Insol-8 casements have u values around 0.18, and SHGC of 0.38 (whole unit R5.5).



I also found this: SC 0.42 and whole unit R-4.5 for the Insol-8
http://www.wec.ufl.edu/extension/gc/madera/documents/EH296.pdf

=====

Whatever the case, we are very satisfied with the performance of these windows.

 

[Seasoned Oak]

Yar. There is not a physical limit to low-e. To keep heat in you want to block long-wave IR, while to let solar in you need to pass near IR and visible. Perfectly doable optically.

But many low-e windows are sold to reject solar heat, and thus near IR to reduce AC. Presumably Anderson could give you the SHGC spec you want.

I wondered about the solar gain capability of my windows since i bought them at an auction, but they must be the right ones as the solar gain is tremendous.

 

[woodgeek]

Whatever the case, we are very satisfied with the performance of these windows.

For your application, climate and window treatment, sounds like a glass curtain wall effect....seems like you picked very well. Maximum R-value (since you weren't going to do massive insulated blinds). Even if the SHGC = 0.4, that is prob ideal too since more than that might lead to overheating on warmer days in the winter.

 

[Seasoned Oak]

Im lookin for some of those insulation blinds if anyone has any links?

 

[Thomas Anderson]

The windows I have are not made anymore, and I don't recall that SHGC was provided at the time I bought them. BTW, the windows are Hurd Insol-8. I learned that problems were experienced with some of these windows, but as mentioned, ours are 10-20 years old and no problems. I did find this info on a search:

http://www.repp.org/discussion/greenbuilding/200203/msg00191.html

The best performing (in our cold cloudy upstate NY climate) wood-framed window I could find was from Hurd (<http://www.hurd.com>). Their Insol-8 glazing system uses 2 sheets of lightweight Heat Mirror film between the 2 panes of glass, giving 3 argon-filled spaces (effectively a quad-pane window). I used the free RESFEN software to compare performance in a model similar to our house plan. The Hurd Insol-8 casements have u values around 0.18, and SHGC of 0.38 (whole unit R5.5).



I also found this: SC 0.42 and whole unit R-4.5 for the Insol-8
http://www.wec.ufl.edu/extension/gc/madera/documents/EH296.pdf

=====

Whatever the case, we are very satisfied with the performance of these windows.

That sounds reasonable -- R-4.5 to R-5.5. So they compare similarly to my double-pane low-e windows if I use insulated curtains or blinds at night. Most of my windows are 2'x4', so that's easier for me to do than for your "14' x 5' expanse of glass". Definitely different designs warrant different strategies for window efficiency.

 

[Thomas Anderson]

For your application, climate and window treatment, sounds like a glass curtain wall effect....seems like you picked very well. Maximum R-value (since you weren't going to do massive insulated blinds). Even if the SHGC = 0.4, that is prob ideal too since more than that might lead to overheating on warmer days in the winter.

Overheating was a concern of mine and one of the reasons I tried to model it in a spreadsheet. Orientation helps tremendously in this regard. We do get some overheating in late spring and especially early autumn when the sun is still relatively low on the horizon but daytime temps are also higher; but at those times, the nighttime temperatures are sufficient to expel excess heat from the house during all but the middle of the day. But if the sun is shining enough to heat it up, it usually gives us enough electricity to run some air conditioning. And this is another reason I'm looking into installing automated curtains or blinds... they can be programmed to respond to outdoor conditions including temperature, insolation, and wind speed, plus indoor temperature, in order to balance solar gain and heat loss.