I was waiting for someone to ask that question!
 

Daox, it seems appropriate that the first smart cookie to ask it was you. I have thought long and hard about that and have decided the best fan is no fan. I built the ducting to the living space with oversized dimensions and with as minimal a distance as possible just to cut down on air friction. There will probably be less air friction in the duct than there will be in the little filter at the bottom that the air goes through. That means that the only real effective resistance to bringing that air down to the living space will be due to overcoming the stack pressure.

Since this system is bringing warm outside (fresh?) air in winter it makes sense to just use it as makeup air for a bathroom fan. So that's what I plan to do. I think a an 80 or 130 CFM Panasonic bathroom fan should be plenty to overcome the stack pressure and bring that warm air in. The advantage of doing it this way is that you avoid most of the draftiness of only warm, (not hot), air being blown on you from a pwerful fan.

The trick is you have to make the house pretty air tight to do that. Otherwise the house will find some other way than the open duct to get that makep air. I've been spending a LOT of time making the envelope airtight.
I'm also planning on installing a Aldes makeup air vent for those times when I can't, or don't want to, use the attic as a makeup air source.

That sounds like it should work if your house is tight enough and your bath fans can pull enough air through. I have two unknown bath fans (recycled), but I'd like to add a 3rd to pull even more air through with my setup.

Hi Daox,
I may have sounded more confident than I was about the adequacy of a bath fan to bring down the air. What I've been thinking about recently is using a 130 cfm Panasonic fan in full ON mode rather than the SLOW mode. The beauty of those fans, besides being efficient, is that they are built to compensate for a somewhat tight house. That is, it will use up to double or triple the power in watts to keep the air moving at the rated CFM when things are tight. I'm hoping that at the full rated 130 cfm, and the ability to compensate for high equivalent wg air friction in the house, that one fan in my relatively small, open plan house will be sufficient.

That also brings up the whole issue of controlling the fan and the duty cycle of the fan. Since I really only want the fresh warm air to come in when the air is actually warm maybe it makes sense to use it on high for a limited amount of time and not at night when the air might be cold. It would depend if you have heat storage. I don't. A lot of the answers to the questions just depend on trial and error. If you plan on going ahead with outsulating your house envelope then it might be worth putting off some of the ripping and tearing until then. You'll then have a good baseline to incrementally increase the airflow as needed.

Update on the bath fan CFM requirements... After googling I found this:

"First, determine the height of the building (H) above grade. Second, arbitrarily select a neutral pressure plane height (Hnpp) above the grade of the building, usually one or two storeys above grade in a multi-level building. In a low building select grade as the neutralpressure plane (npp).
Third, select a given height above (or below) the neutral plane, such as the roof plane. Compute the pressure difference at that plane as follows:

"ΔPs = dig(H-Hnpp)(Ti-To)/To
where ΔPs=stack effect pressure difference (Pa)
and di =density of indoor air, kg/m3
and g=gravitational acceleration, 9.81 m/s2
and H= height of plane above (or below) Npp, m
and Hnpp=height of neutral pressure plane, m
and T=absolute temperature, Kelvin
and i = indoor, o = outdoor

"For a 60-m (197 ft.), 20-storey building with a neutral pressure plane at about the 4th storey or 12 m (40 ft.) from grade, in winter with an outdoor temperature of -20°C and an indoortemperature of +20°C, the stack effect at the top of the exterior wall (and roof plane) is:

"ΔPs=dig(H-Hnpp)(Ti-To)/To
=1.2x9.81x(60-12)x[(273+20)-(273-20)]/(273-20)
=89 Pa
"
I calculated the stack effect pressure I would need to overcome by inputing the information from my house. It is 13.5 feet to the top of the roof from the floor in the house, which I'm using as the neutral plane because the floor is where the duct will vent its air. I'm using 4.44 degrees C (40 F) inside air and 29.44 C roof line air (85 F). Those are worst case scenarios with 45 F degree differential temperatures and will cause the most stack pressures that might reasonable have to be overcome after having the house unheated for a time.
Using those inputs:
=1.2x9.81x(13.5-0)x[(273+4.44)-(273+29.44)]/(273+29.44)
=18 Pa

Blower door tests are done at 50 Pa. Since I'm going to want to get a blower door test done anyway all I need to do is get that result and convert by 18/50 = .36 My understanding is that a blower door test usually gives the CFM rate within the overall air changes per hour. So when I get that CFM rate I'll multiply it by .36 to get the overall CFM rate of a bathroom fan needed to overcome the 18 Pa stack pressure. I hope I'm getting this right as I'm learning as I go.

Anyone who sees an error let me know. I'm fallible.

Edit:

Yep, I'm fallible. Used feet instead of meters. It should be
=1.2x9.81x(4.154-0)x[(273+4.44)-(273+29.44)]/(273+29.44)
=5.5 Pa

And the correction factor for my house won't be .36 but will be
= .11

Well, its seems I was looking through rose colored glasses when I estimated that the temp rise in summer of 40-45 degrees would continue into winter. My reasoning was that the lower inclination of the sun was already included in the lower air temperatures in winter so that there would be no reason to double count that in the temperature differential between outside and attic temperatures. So there is a lower temp rise than what is already included in the lower air temperature in winter (technically Fall).

Everyday I've watched the temp rise between outside and attic get lower. I still haven't closed the roof vent because I wanted to see how things change without a lot of variables. Here are some temp rises I've recorded:

Oct 8 - 34 degrees rise, attic high temp 110 degrees
Oct 28 - 31 degrees rise, attic high temp 98.

I was starting to get concerned because on poor solar days it was dropping into a temp rise of the low 20s. I needed to do something before things deteriorated further, which I did. More tomorrow on how I improved things while keeping the roof vent open for comparison purposes.

http://ecorenovator.org/forum/attach...p-img_0115-jpg

http://ecorenovator.org/forum/attach...p-img_0142-jpg

Getting back to the problem... attic solar heat has a quality problem, not a quantity problem. That is, there is a huge quantity of roof surface space that the sun radiates onto. There is a lot of solar energy impinging on any roof. But it also has lower temperature than a dedicated solar panel because for various reasons it is not as efficient at capturing energy as a dedicated solar panel. This means that for a given surface area that the sun radiates onto the internal temperature of the air in an attic will be much lower than in a solar hot air panel.

I was realizing that there was a significant amount of heat leaking out on my hip roof that wasn't from leaks in the radiant barrier (besides the open roof vent). I could tell because everytime I went up there midday I'd be sweating, even after the radiant barrier was up. It was a lot less than before the radiant barrier, but it was still about 85 degrees up there even when it was signifiantly colder outside.

The problem isn't equal in all areas so I decided to just fix it at the worst areas. Those are the areas where the heat concentration is the highest.
In a gable roof the greatest heat concentration is right near the ridge. In my hip roof there is a big concentration there, but also along the hip rafters where the regular rafters meet it. One has to allow the air to "slip" past each junction point by dropping the radiant barrier a bit at that point. Every regular rafter that meets the hip rafter must channel hot air from every preceding regular rafter that intersects with the hip rafter. So by the time you get to the highest and last regular rafter junction point all the hot air from the previous lower rafter channels is meeting in that narrow junction point.

Here's a picture I drew that might give anyone a better idea. The areas in the broken lines are the areas of greatest heat concentration. It's a seat of the pants estimate and not a quantitative picture of heat concentration.:p

http://ecorenovator.org/forum/attach...1&d=1384668392

So the obvious answer to what I just said is add more insulation at those heat stress points. That's what I did and I used duct board rather than foam board, mostly because it was my preference, even though it is more expensive. Total cost was about 350 bucks for the duct board but its been worth it. The temperature near the roof but below the radiant barrier does not get over about 3 degrees of the rest of the open uninsulated part of the house. My house is under major construction as later photos will show but the one and only advantage is that it pretty much eliminates heated or conditioned air from the living space (except my bedroom) leaking into areas that I don't want and affecting the temperature results. Its basically one big unconditioned space except right above the radiant barrier at the roof.

It seems to have worked. The dramatic spiral in lower heat rises as we approach the winter solstice has been arrested. I'm getting consistant heat rises at or near 30 degrees above outside ambient. I think that's pretty good for having a foot diameter vent wide open at the top of the roof!

http://ecorenovator.org/forum/attach...1&d=1384669773

That 60 degree temperature reading is in that large unconditioned space. So you can see that even at mid-afternoon while the outside temperature was 63, bedroom temp around 70ish, and roof temperature was 93 degrees, there is surprisingly little leakage of heat from the roof to the unconditioned space. There is no ceiling drywall in the house yet.

Nice write up.

Thanks Iamgeo. Even with the limited amount of success I've had here I'm still not convinced that an attic solar system will do well in all climates represented on this forum. I think it will probably work and save considerable amounts of energy in places like California, Texas, and most states south of the Mason-Dixon line. But the Pacific Northwest and Alaska... forget about it. In the former there is just too much cloud cover and the latter the sun is just too low in the sky. I think in selective areas north of the Mason-Dixon line it might eventually be a useful way to help heat your house. I think the jury is out. I think you would have to forget about using a direct heating system from the attic in most of those places and just use it as a tie in to a basement, crawlspace, or as a fresh air input to a forced air heating system. :thumbup:

Here's some pictures of the insulation installation. The ductboards are 2' x 4' in dimension and reflective on the side opposite being seen. They are equivalent of about R-8 (2 inches of fiberglass insulation). I first had to attach 2x2s as extensions to the hip rafters. That allowed an air space between the reflective surface of the ductboard and the foil above it. You can't have a reflective surface touching the surface above or it will directly conduct heat and bypass the radiant insulation properties of the reflective surface. The edges where they touch are insignificant area of the total surface. It can be seen here.
http://ecorenovator.org/forum/attach...1&d=1384765338
http://ecorenovator.org/forum/attach...1&d=1384742431

I really couldn't use insulation batts because they flop around too much. But even the duct boards can't be screwed directly, thus the wood strips sandwiching the ductboard against the 2x2s and giving some security.
http://ecorenovator.org/forum/attach...1&d=1384742431

I then covered all the insulation with foil barrier just to look better. The center section with the backflow damper will remain removable and accessible because behind it is the temperature sensors and damper motor. So I just wrapped the foil around the ductboard next to that section so there won't be damage to the adjoining fiberglass when I have to remove the center section.

http://ecorenovator.org/forum/attach...1&d=1384742431

http://ecorenovator.org/forum/attach...1&d=1384742431


Yeah, I know this is all way overkill and taking too long. The only reason I'm even paying so much attention to detail is because, to the best of my knowledge, no one has ever tried using an open system -no air recirculation- and still get high enough temperatures to use that attic air directly. If it works then it will be a first. That's not to say someone somewhere hasn't done it before successfully. Just that I haven't seen it documented anywhere.
http://ecorenovator.org/forum/attach...1&d=1384744577

Here's the proof of the pudding. Today I got the temp rise back up to 34 degrees with a 98 degree high at the roof. When I close the roof vent I am hoping and expecting to keep that high temp even as the days get colder and shorter. In my climate a 34 degree temperature rise is usable to heat the house. I can't say for other climates.