I know attic ventilation is a really contentious topic but I’m particularly confused on how it applies to older homes.
I see many, many eaveless homes without any ventilation whatsoever–many have been insulated for decades and I have yet to see any water damage from condensation–only from water penetration. This is also in homes with eaves and homes where the attic is finished/unfinished/occupied/unoccupied/insulated/non-insulated/etc. The eaves are often very tightly built and well-sealed–at least as good as good new construction.
Many have whirlibird ventilators installed when a new roof is applied with NO provision for inlet air. Is this as utterly stupid as I presume? Won’t it just lead to depressurization, loss of conditioned air and and even backdrafting of cumbustion equipment?
I’m in SE MO–an extremely mixed continental climate. Most common condition regardless of outside temp is general high humidity.
What I have seen in new homes though is really strange. We had a unusually long period of constant below freezing weather two seasons ago. I noticed many new roofs–reasonably tight new construction, full ventilation–where snow and ice were melting between the rafters/trusses. There were literally 4″ or so wide “stripes” over the rafters that persisted for many days. This was usually in large attic spaces and the homes have LOTS of can lights. I haven’t been able to inspect any of these, but I almost believe that warm, moist conditioned air may have been condensing on the coldest thing in the attic–the rafters/trusses. Is this a valid concern?
Replies
Because older homes are not sealed air tight, the attic can still draw air through attic hatches, light fixtures cracked plaster etc. This of course draws warm air into the attic space which results in snow melting sooner on the roofs of older homes. Newer homes are often sealed up tighter than a ziplock baggie which results in no air transfer from the interior living space into the attic cavity. Top that with R36 to R40 insulation and you have a roof system thermally separate from the living space. Condensation is not a concern because the space is venilated to such a degee that moisture is not trapped in the attic cavity.
sometimes board sometimes knot
You mentioned seeing "striping" on roofs. For clarification, was the melting occuring over the rafters/trusses or over the bays? If the melt occured over the rafters, then you'd see thin stripes of the roof at 16" or 24" centers, etc. If the melt occured over the bays, then there'd be little strips of snow/ice remaining over the rafters at 16" or 24" centers, and the roof would be exposed everywhere else.
Thought I said it before, but the unmelted stripes were directly over the rafters/trusses.
What I have seen in new homes though is really strange. We had a unusually long period of constant below freezing weather two seasons ago. I noticed many new roofs--reasonably tight new construction, full ventilation--where snow and ice were melting between the rafters/trusses. There were literally 4" or so wide "stripes" over the rafters that persisted for many days. This was usually in large attic spaces and the homes have LOTS of can lights. I haven't been able to inspect any of these, but I almost believe that warm, moist conditioned air may have been condensing on the coldest thing in the attic--the rafters/trusses. Is this a valid concern?
You did say it before -- I just wanted to verify it.
Here's my analysis of what you describe: Snow builds up on the roof. Then, the warm air in the attic starts to melt the snow from underneath. When you think about it, the thinnest, least insulated part of the roof is the sheathing itself. Therefore, this is the area where most heat transfer will occur, and we see this evidenced by the fact that the snow is melting between the rafters.
The rafters are not really "the coldest" part of the attic. Rather, they are acting as extra insulation and preventing the warm air of the attic from reaching (and warming) the snow load on the roof.
I don't think you have to worry about ice building up on the rafters.
Regarding attic ventilation, you're right that adding whirlybird ventilators with no air inlet isn't a particularly brilliant idea. Air in = air out, so that air going out has to come from somewhere -- in this case it's being sucked out of the heated living area of the house (although a portion could be coming in through leaks around the eaves, etc.). In the case of regular roof jack ventilators, I'd suspect that some air flow simply comes in ventilators on the windward side of the house, and goes out ventilators on the leeward side.
I also agree with your observation that hundreds of old houses have no attic ventilators whatsoever, and do not seem to suffer from condensation issues. However, I believe I have, in fact, seen some proof of condensation. One client of mine has a 1907 house with a covered porch on the south side. The hip roof is in good condition, and has no ventilators or air inlets, etc. After having the house repainted, some dark, mildewy-looking stains started appearing on the box beams of the porch. My theory is that the air trapped in the porch attic gets very hot due to the southern exposure, and then plunges below the dew point as it cools at night. However, I haven't bothered to gain access to the attic and determine through direct observation whether there is, in fact, any condensate.
Also, I have always wondered why beaded soffits on old houses get so beat up. That is, the paint is always peeling off and is much rougher than any other place of the house. Could this also be due to condensation issues? Could water be condensing in the unventilated soffits and pooling on the blind side of the beadboard?
Does anyone else have any theories on this?
Ragnar
"Everything looks fine when the holes are covered with a blanket of insulation, but when melting snow shows the rafter pattern on the shingles, it becomes obvious that the thermal boundary is really the roof deck. The 12 in. of insulation in the ceiling is yielding an R-value of close to 0."
I found this in the "energy efficiency" section here. It seems to describe what I'm seeing.
I've definitely noticed what you're talking about on beadboard eaves, but the culprits I usually see don't involve condensation.
1) Leaks higher on the roof--dormers the worst, followed by plumbing vents and plain bad roofs. Small amounts of water follow the steep rafters/sheathing to the eaves, not always showing even on finished surfaces inside.
2) Clogged gutters leading to water backup under the roofing.
3) Excess paint/poor prep. Being down facing, the eave doesn't get the cleaning action of rain, but it certainly gets more than its share of dirt from the wind. I don't think anyone enjoys painting this surface--much less cleaning/prepping it properly. The surface itself (beadboard) is VERY easy to "overpaint" and extremely difficult to scrape/strip. Since it's not exposed to the sun, it doesn't chalk like normal and since it's often painted with higher gloss aklyd paint, the lack of a good TSP scrubbing leads to paint failure between the layers. Even without water damage, I think the difficulty of application/prep and unique location lead to excessively thick paint film and eventual failure.
My general adherence to historic authenticity often goes by the wayside with beadboard eaves. After repairing leaks; ensuring there is a metal drip into the gutter; repairing the fascia (and covering it with aluminum); repairing gross problems with the beadboard; I usually install 1/4 exterior plywood over the beadboard. The cost of stripping/replacing it is more than most anyone around here can bear and the plywood holds paint well and is easier (my opinion) to prepare for subsequent paint jobs.
I've switched exclusevely to latex topcoats and find myself using alkyd primers less and less. The most insidius problem I'm had with alkyd primers is mildew forming in the prime layer and working its way OUT to the finish. This happened after a 6"+ summer rain with over a week of weather with dewpoints near/even above 80! The air literally dripped every evening as the sun went down. Similar things using latex prime & top coat didn't seem to suffer.