For our new house (not yet under construction), I hired someone to do heat/cooling load calculations, select HVAC equipment and do design for duct work. This person is doing the calculations and spec only, and is not invoved in the sale or installation of the equipment. I provided a lot of the info you would expect, such as all the details on foundation, wall, roof construction, window specs, house orientation, our plan for air sealing, etc.

A question came from the person I hired asking about the level of air tightness that I expected to achieve. We are acting as GC and doing a lot of the grunt sealing work ourselves. For the sake of the calculations, I was asked to consider the following as a general framework:
 - typical code house, 7 ach50
- upgraded construction with good attention to air sealing, 4 ach50
- passivehaus, 1 or less ach50

After a brief discussion where I explained I was reluctant to pick a number because of my lack of experience in building/measuring-my-building, it was recommended that the calculation use 5 or 5.5 ach50. Honestly, I feel like I'm doing a lot on the air sealing front; Zip/taped walls, caulking ALL stud bays, caulking in the attic and any chases, caulking and rigid insulation between floor joists... It seems like a lot of effort and I'm happy to do it for energy efficieny and comfort, but I would think it should yield better than 5 in the albeit informal 'range' listed above ... but jeepers I really do NOT know. Granted, there is no attempt (or claim) that we're going to passivehaus standards on insulation (not sure what all their air sealing calls for).

Is the process I describe above typical for how these calcs and specs are done? Seems rather chicken or the egg to me. You need to know on how much infiltration will happen to accurately predict heating/cooling loads, but you can not know that number (via blower door measurement) until it is too late to put in duct work and very late to be ordering hvac equipment.

Are there guidelines for estimating air infiltration based on construction techniques, sealing options selected, etc?

Any advice or education is appreciated.
Thanks

BabyBldr-

Yes, it is a bit a guess work.  However, let's assume that 2/3's of the heat losses will be due to conduction, and 1/3 due to infiltration.  Then a 50% uncertainty in the infiltration will only amount to a 17% uncertainty in the heat loss calculations.
 
There is a report by Max H. Sherman and Nance E. Matson "Air Tightness in New Houses in the U.S., A Preliminary Report," LBNL 58671 that provides data from the LBNL database for "new homes" as shown below for "natural ventilation," although this report was dated 2002.  If we multiply those natural ventilation results by 20 to convert (approximately) to ACH at 50 Pa, then the average for new homes at that time was a mean of 6 ACH50.  I would suggest that a careful job of sealing should reduce this to about 3 ACH50.  I have a very simple house design, the guys did a careful job of sealing every stud box and every electrical feed-through, I used infiltration reduction boxes around every electrical outlet, and the measured ACH50 was 2.4.  Dana1 has pointed out that it is possible to do much better than this (and Canadian standards would require better), but if I were pulling a number out of the air, I would suggest 3 ACH50.  Windows that are casement or fixed along with doors that include storm doors might reduce the value, while sliding doors or windows might increase it.  Lots of "features" in the outside framework might also increase it.   



Lee Dodge
http://www.residentialenergylaboratory.com
in a net-zero source energy modified production home

Infiltration is a bit of a guessing game and heat load calcs are often not accurate as well. To be concerned about what the actual amount will turn out to be is noble. In my opinion trying to achieve a perfect tight home has more negative health impacts than the energy saved. Indoor air quality should be higher on the list than some random number. Depending on where you live, climate conditions,etc. should be what determines the degree of infiltration.

I think the only way to really quantify it is to measure it with a blower door test. There are just too many variables.
I think it is time we stop this BS about not building houses too tight.

“In my opinion trying to achieve a perfect tight home has more negative health impacts than the energy saved. Indoor air quality should be higher on the list than some random number. Depending on where you live, climate conditions, etc. should be what determines the degree of infiltration.”

I don’t think anyone here is advocating building very tight buildings without adding controlled ventilation. If you are relying on random infiltration to supply fresh air, you indeed have a problem. Build tight, ventilate right.

Depending on the heating source, you may be able to use something that has a significant range of heat outputs, ie a modulating gas boiler. If you size it for 4 ach50 you should be OK.

cheers,
Eric

I just took my 1958 house from a 4.07 down to a 2.43 ACH 50, only focusing on the rim joist and exterior sheathing. Still have a foolishly leak open wood burning fireplace, and a couple of HRV vents that the blower door guy wouldn't let me tape shut.

If you're going with taped ZIP and paying attention to your wood-to-wood joints, there's no reason why you couldn't get a 1.5 easily.

Had my energy evaluation for energy star today. Got a 1.43 ACH50. I used lots of caulking on the inside of the framing on the ground floor, foamed the band joist with 2#cc, caulked the outside seams of the OSB, caulked all the plates. Foamed where I couldn't caulk, including every penetration. What the modular builder did with the main floor is a mystery, but probably not much if anything, but I went through and foamed all the main floor electric boxes from the inside, and caulked where the walls met the floor. The front half of the roof (Cathedral) has 2#cc, and I foamed the marriage wall and blew in R70 cellulose in the back half. All the walls have R21 fiberglass batts. I taped all the seams on the ground floor but the main floor modular installation was not as meticulous.

If you plan on being thorough, and it sounds as if you are, I would say a 1.5ACH50 is not unrealistic.

-Rosalinda

Thank you all for your input and personal experiences. I shared this thread with the person doing our heat/cooling load calcs and designing the HVAC system.
Rosalinda and Boardom congrats on doing the work and actually measuring it ... and on getting such good numbers, that must be very satisfying!

Hi, original poster here. I've got the results of the analysis and a recommendation for our hvac system (mini-split, which is what I had asked for). I'm back with a request for wisdom from the forum.

I’ve read that orienting the house to have south side high SHGC glazing with sufficient roof/porch overhangs and night time thermal covers (for winter) can make a big difference in the amount of energy you need to condition the house. I set up our house to do just that. I also planned for some easily added thermal mass in a tile floor and soap stone counters in the south side dining room and kitchen. It is not a full-out passive solar design, but makes some use of those principals.

In a Manual J calculation, are these types of passive solar design elements accounted for? I had assumed they were, since the Manual J is used to size the equipment – and I don’t want to get it oversized. But through a series of interactions with the company I hired to do the analysis and system design, I’ve come to find out – if I understand what they’re telling me – that Man J is peak load. Since peak load occurs for the coldest night time temp, there is NO accounting for solar gain in that number. I asked about the possible benefit (reduced load) of solar heat radiating back out from the thermal mass - but they said that was not part of Manual J.

They also treated all windows as ‘naked’ even though I spec’d insulated window coverings. And they set the desired indoor winter temp to 72, even though I typically set ours (in our leaky 2x4 FG bat house) to 65 and never above 68. Also they didn't want to figure in any contribution from a small wood stove as backup option in the coldest parts of the year, if the heat pump couldn't keep up for some reason.

Ok - so I’m wondering if the system they designed might be oversized to the extent that I should attempt to hire another professional to redo the calculations. Or is it something along the lines that a mini-split heat pump, for example, will work for such a huge range of BTU needs, that plus/minus some large percentage on the heat load calcuation isn't going to change the equipment design. I know the readers of this forum can’t answer the question (oversizing) specifically - without all the data and specs. But I’m wondering if those familiar with these types of calculations and how they’re used to spec equipment could help me understand the real world issues here.

Thanks for your time and help.

Babybldr-

I have not used Manual J, but it makes sense that you do not want to depend on solar heating to size you HVAC for the coldest night.  In my case, I used the computer model HEED and it computed something slightly less than 30,000 Btu/hr for a 1600 sq. ft. well-insulated house, but the smallest size furnace of the type my builder was installing was 60,000 Btu/hr.  This natural-gas-fired, hot-air furnace has no trouble keeping the house warm on the coldest nights, but is very slow to heat the house up if I leave for a few days and set the thermostat down to 50 F or so while gone.  You might consider that limitation for small heating systems if it applies to the way you use your thermostat. 

For my 7000+ heating degrees days (degrees F), modeling predicts a 21% reduction in seasonal energy use with a 5 degree F reduction in house temperature.  However, since the design temperature used in the Manual J calculations is much lower than the average wintertime temperature, the change is heat loss at the design temperature is reduced by only 7% for a 5 degree F reduction in house temperature.  If you are in a colder climate than 7000 HDD(F), then the sensitivity to house temperature would be even less, or visa versa for a warmer climate.  Therefore, you could estimate a slight reduction in the Manual J results to account for your lower house temperature, just computed as the ratio (65F - T_design) / (72F - T_design). 

Lee Dodge
http://ResidentialEnergyLaboratory.com
in a net-zero source energy modified production home

I have studied Manual J (borrowed from the library) and there are lots of things they don't take into account. Whoever did the Manual J for you should use whatever design temp is called for by code as long as your comfort level is equal to or falls below code. For example NY State code in my area calls for 68 degrees when it is 1 degree outside, so a delta T of 67 degrees. I have to be able to show my code guy that the heating system can achieve this level of heat, despite the fact that if the house is above 60 at any time, I am way overheated.

If you size your heating system to meet code requirements, you should be ok. In an energy efficient house, ALL the heat input and heat savings should be taken into account when designing your heating system, since the house will not lose heat the way a leaky poorly insulated house will. You need to make sure the full insulation value of your structure was used, and they should include the R value of your night time window coverings in Manual J, since that will give your windows a much higher R value at night. You should be able to figure out a BTU/hr input from your solar tempering and thermal mass design, and though it wont be included in the Manual J, you should use it in your heating system design, and for your code official, in addition to adding in the design output of the wood stove.

For the things that Are included in manual J, such as the window R value, you should probably sit down with them and tell them what figures to input - they ARE working for you.

-Rosalinda

I wouldn't think they would take into account the Window insulated shutters, unless they were automated. When you move/sell, and the next owners say screw these annoying shutters, the house itself should still meet the basic heat requirements.

You'd probably be better off using something like the PHPP (passive house planning package) or something along those lines if you want a more accurate load calc.

J

Posted By boardom on 08 Apr 2011 12:00 PM
I wouldn't think they would take into account the Window insulated shutters, unless they were automated. When you move/sell, and the next owners say screw these annoying shutters, the house itself should still meet the basic heat requirements.

You'd probably be better off using something like the PHPP (passive house planning package) or something along those lines if you want a more accurate load calc.

J

I agree boardom, right on point!

Taking the shutters into account would depend on how long BabyBldr plans on living in the house. If it is for the long term, then the shutters should be used in the calculation, along with every other conservation practice BabyBldr plans on using. The specific Manual J should be for this specific house and its users, not some generic house owner that might purchase the house in some unknown future. On the other hand, if the plan is to build, sell and move in a few years, that is a different story.


-Rosalinda

e-house has a 30,000 btu/hr. peak heating load, but almost 50 percent of that is through high-performance windows comprising less than 13 percent of the building envelope surface. In a worst-case scenario, less than 5 percent of the heat loss is from infiltration, 20 percent from ventilation and 25 percent wall/roof heat loss.

This is an energy efficient example, but it demonstrates that after a point, one needs to look at windows and ventilation as the places to save energy. In general, I don't see much emphasis on shutters and methods of reducing ventilation requirements (eg, air purification, moisture reduction, dehumidification, etc).