Well, that is not to say you should go and blow it all on stupid stuff. What we are saying is that you can really save a lot of energy and money if your home has a tighter thermal envelope.
This is why the Energy Efficiency section in newer editions of the International Residential Code (IRC) are gradually getting more stringent. This made me wonder how tight existing single family homes are. According to a 1998 study “Air-Tightnedd of U.S. Dwellings” published by the Lawrence Berkeley Laboratory http://epb.lbl.gov/publications/lbl-35700.pdf, the average SFH in the U.S. has a 1.72 NL(Normalized Leakage) or 29.7 ACH50 (Air Change per Hour at 50 Pascal). I won’t bore you with the relationship between the two but basically the ratio is 1:17.5 with a 13% standard deviation. 29.7 ACH50 is quite leaky. The study also showed that there are a few significant factors that causes some homes to be tighter than others.
Number of Stories: multi-story homes (31.5 ACH50) were about 11% leakiers than single-story homes (28 ACH50) . This makes sense, more stories means more joints, penetrations to seal and more places for failure.
Floor/Basement Type: homes with crawl spaces and unconditioned basements (30.6 Ach50) are 5% leakier than homes with slab-on-grade or fully conditioned basements (28.7 Ach50). This means a lot of heat loss occurs through floor leakages and highlights the importance of insulating slabs, foundation walls and air sealing all the critical joints in the foundation.
Age of construction: Homes built prior to 1980 were about twice as leaky as ones built after. The data also shows that from 1980 t0 1998, there was very little improvement in air-tightness. This maybe the reason for the increasingly stringent air-tightness standards in recent editions of the IRC.
Duct System: Homes without a duct system (15.75 ACH50) were reported to be leakier than the homes with one (12.25 ACH50). Another interesting result is HVAC duct systems account for just under 30% of the overall air leakages of a home. This explains the need to build air-tight duct systems in homes.
Retrofits: Homes that undergo energy retrofits on average were 25% tighter than prior to retrofit. This shows that retrofits do work, but I think 25% is probably too low a number. If you are gonna do it, you might as well go deep!
This is an old study but it explains why IRC seeks to reduce air-leakages of new constructions. The 2009 IRC N1102.4 which is just beginning to be enforced in Virginia provides builders with two options: Either to comply with a checklist of visual inspections by an inspector or show a blower door test result of 7ACH50. Here are two interesting articles analyzing this new code provision.
As Martin Holladay pointed out in his analysis, the code writers made a mistake by not explicitly specifying “who” is to perform this blower door test. This creates a significant loophole which I hope jurisdictions will close before they adopt them. In any case, at 7 ACH50, this air-tightness standard is still considered leaky but it will probably force builders to start thinking about the issue.
Fast forward three more years, hopefully by then the already published 2012 edition will start to be adopted. The 2012 edition eliminates the visual inspection option, instead requires 5 ACH50 in Zone 1 and 2 (really deep south), and requires 3 ACH50 in Zone 3 through 8 (the rest of the U.S.). This standard will force builders to come up with an air-tightness strategy. However, this new edition still leaves the “who” in doubt. It states, “Where required by the building official, testing shall be conducted by an approved third party.” This is unfortunate, not only does it leave the question of who is to perform the blower door test to individual jurisdictions, if the state code is silent on the issue, it actually leaves it to the “building official”, i.e. the plan reviewer, field inspector, etc. to decide whether to require the test be done by an “approved third party”. I can see various compliance problems ranging from uneven enforcement to corruption issues.
All this legal analysis aside (I set that aside four years ago…), how does one actually tighten a building envelope?
We are by no means experts in this field (in fact, we are newbies) but we do have expert/supplier friends at Four Seven Five Performance Building Supply and Small Plant Workshop and all the Passive House designers who give us advice. Here’s what we’ve done so far to try to reach our goal of 0.6 ACH50. If you have a suggestion, feel free to comment.
This is a vapor-open tape sold by 475. The adhesion is very good but because SIPs edges and corners can get a little rough, it’s very important to clean up the OSB a little bit before applying the tape.
Ideally, we would have taped these joints above and below the plate too. But because the joists were already there, taping it would have been very labor intensive. Here we opted for a good sealing caulk. The lesson here was to do air-sealing as we build the structure instead of air-sealing after the build. But then I guess we will need to ensure the work is not exposed to rain.
Here, the the joint between the SIPs and plate, we caulked. Between the plate the Superiorwall is a larger gap, which we caulked and sprayed foam. For the larger cavities, we used high expansion foam.
For the windows and doors, we installed them using Tremco ExoAir Trio (you can get it from Small Plant Workshop). I think it is a good product for installing windows. I hope it will eliminate having to come back with a can of Great Stuff to go over the gaps (Only the blower door test/Thermal imaging will give us the answer). It’s also important to caulk the corners prior to installation as an insurance policy. I have my reservations about it when it comes to installing doors, especially at the threshold. Because it is expansion foam, it can potentially push up the threshold if the opening is large, like in a French door, causing the doors to not operate properly. Additionally, don’t trust it be weather tight as the literature claims. When we noticed our French door threshold bulging, we pulled the trio out from under the threshold and the Trio was soaking wet. Basically, we don’t recommend it where water can potentially accumulate.
For the exhaust and supply points of the ERV, first, we made 10″ diameter round holes for the 8″ ducts. We sleeved the ducts through a Roflex 200 Gasket, using Tesco No.1 tape to tape the connection and taped the gasket to the SIP panel. Finally, we made sure the duct is positioned in the center of the penetration (not touching the sides), then sprayed foam from the outside. This is my favorite seal! I must be a real geek to have one…
For all plumbing penetrations, we made sure the penetrations are larger than the pipes, position the pipes in the center of penetration and sealed around them. We will most likely trim off the foam, caulk and then tape around the pipes to ensure the seal holds over time.
The MiniSplit line set is basically sealed the same way as the plumbing pipes except with the insulated coolant line. We cut back the black insulation for the portion that is in the wall and sprayed around the line set.
This is less of an Air-tightness but more of a insulation issue. Our SIPs (except for the roof panels) came with electrical and plumbing chases. These chases cut into the insulation value of the walls. We weren’t sure at the time we ordered the panels whether we would need to use them. In the end, we did not use even one of them. We basically drilled holes along the chases every foot and filled each hole with 2-part foam. The work is not difficult but unnecessary if we had ordered roof panels for the entire house. So, we learned our lesson. Also, don’t believe it if anyone tells you the chases have a negligible impact on the overall performance, we emptied at least 6 tanks of foam into the walls.
Oh, if you survived reading this post, check out our video, it’s a lot more fun