For those conducting energy audits and developing retrofit work scopes on low-rise residential, please share your experiences in this thread. Successful methods, pitfalls and complete disasters are all fair game for sharing with the group.
Since there are distinct difference between high-rise and low-rise strategies, let's keep our discussion here only on low-rise. High-rise stories go here.
Do you want to QUANTIFY" flow, or just assess it?? If you need actual flow numbers, well, that's really tough. The only way I know to do that is with multiple fans and guarded blower door testing. (Sounds like that's just what you are doing.). That only makes sense if you need a number really badly. And in anything other than townhomes, it gets pretty complicated pretty fast.
But you CAN set up a blower door in one unit with an Energy Conservatory APT unit, throw tubes under the doors of adjacent apartments, and do a multipoint test. That will at least show you the magnitude of cross-talk. (When the tested unit is at -50 Pa, Unit B is pulled to -6 Pa, Unit C is pulled to -8 Pa, and so on.) Get in touch with me off line and we can get into details...
I have had several discussion with several different knowledgeable people about this. I am just not sure that it's true that you can test your way around a middle-of-building unit with just two fans; at least not in wood-frame buildings. I am just not confident that there would be NO communication across corners and through floor assemblies.
Think about the center unit in an 18-unit building with three stories and three apartments on each side of a double-loaded hall. I don't know of anyone that has tested the center unit with a blower door simultaneously in every single unit on one side of the building AND with a fan de-pressurizing the hall. And then gone back with just two blower doors and tested their way around the center unit, one surface at a time. I think someone would have to do that at least a few times, and confirm that you get equivalent results, to be confident that you were on solid ground.
But I do know (because I've done it) that you CAN put a blower door in one window of our hypothetical center unit, and see the pressure change in units that are "in contact" with the target unit only at the corners. That tells me that the communication and leak paths are far more complicated than is obvious. It leads me to suspect that only simultaneous testing of every surface will get accurate results.
But then again, the other question is -- how badly do you need to know? In new construction, it's generally accepted that you just get each unit very, very tight. Don't worry about where the air comes from, just stop it. In existing units, never promise that you'll get the unit tight enough to stop that cigarette smoke migration, because you almost certainly won't succeed. And, if you want to save energy in existing buildings, what are you doing mucking around with surfaces that have no delta-t?
Damn, we still have a lot to learn....
You bring up and interesting discussion Don, and I suspect you're right, there could certainly be some communication across corners ect...
However, I would suggest in your example that the number of fans can still be reduced to a manageable number, say 3. On both the first and third floor, install a fan to de-pressurize the hall, open the doors to each unit, and you've now neutralized the leakage between the center unit and the floor above and below. Obviously the possible test scenarios and test setups is nearly endless depending on the exact building configuration. And yes, you ask a great question... and yes, we still have much to learn.
I've been doing some multifamily testing during the last few months and been looking at it from all these different angles. In one building we did some testing of a middle unit (double loaded corridor) on the second floor. We first did at total leakage test - 1 blower door, and then did a guarded test depressurizing all adjacent units and the hallway. The results were very similar - less than 100 CFM difference. My best guess is that the interstitial spaces are pretty well connected throughout the building. Of course this is just anecdotal, but led me to test total leakage for all the units.
I agree completely with Don that you don't have to worry about where the air comes from, just stop it. As far as I'm concerned it's too complicated, too uncertain, and too expensive to test anything other than total leakage in this type of situation - especially new construction. I just don't see the down side to doing the total leakage test and keeping my life simple. Does anyone else?
For existing buildings, you have to consider a whole building blower door to get at the leakage to common areas and non-living spaces which can have a large impact on energy usage. Again, I agree with Don - no need to muck around with surfaces that have no delta-t. I have much less experience with existing construction however.
This is a great discussion and I look forward to hear and learning more from you more experienced folks out there.
Tom Holmes (AEA New York) appears not to be a member of this group, so he isn't here to brag about some VERY illustrative work he's done in New York state. He reported on it at ACI in San Francisco.
He actually did all the work needed to go through an existing eight-unit "garden" style building both ways; testing all eight units individually, then doing guarded tests on all units. His numbers showed that testing units one at a time gave CFM50 reading that were all too high. The error ranged from 15 to 30 percent, in units with "pre" infiltration values from about 1,500 to 2.300 CFM50.
So that indicates (the same as the little bit of experimentation I've been able to do) that you probably do over-state the leakiness of units, by a significant amount, when testing them solo. Test like that would not be all that useful -- It wouldn't be hard to fool yourself into thinking there was a lot more room for improvement than the building actually offers.
The one thing I don't know -- he reported also "post" blower door numbers, but only those coming from guarded blower door tests. I should ask him whether he repeated his individual unit tests after they did their air sealing. If the error stays about the same, then it might not matter so much. I doubt that it would though--If all the air sealing were focused on the thermal shell (not affecting the defective air barrier between units) then then the unit-to-unit leakage would be a much larger proportion of the total leakage, when testing just one unit at a time. That would fool you into believing you hadn't changed the building much -- even though the thermal shell could be MUCH tighter, and the post-sealing energy use could then drop significantly.
You might also fool yourself into thinking that the unit didn't need supplemental ventilation, when it might in fact be pretty darn tight. Tom got a one-bedroom unit down close to 600 CFM50 to outside -- pretty darn tight to depend solely on infiltration for a ventilation/moisture strategy!
Thanks for passing this information on Don. Definitely different results than I got. I was testing a 4 story, 66 unit double-loaded corridor building, so construction type may be the difference or just construction technique. I guess my conclusion is that we can't say for sure whether testing 1 unit at a time will always significantly overstate outdoor air infiltration.
For new construction, we only see buildings with mechanical ventilation, but I will definitely heed the warning that one could be fooled into not needing a ventilation strategy in an existing building.
The second phase of the building I tested is being built right now, and I will have access to test as much as I want before it is occupied. Unfortunately, I'm limited by having to do any extra testing on a volunteer basis. I wonder if there might be a source of funding for this testing out there somewhere.
It looks like Wisconsin is going to wade into doing blower door testing on low-rise MF buildings. We did a training last week on auditing small MFBs and ran a two-fan test on a 12 unit building. That demonstrated that the building was about as leaky (per ft2 of shell) as a single-family home where air sealing would be a good strategy.
The state office is going to "encourage" weatherization agencies to test 5-24 unit buildings "whenever possible." This could get interesting...