When is a wall too thick?
I’m in the planning stages to build a 2-3 bedroom 800-1000 sq. ft. super-insulated house in Western Massachusetts. I recently took ‘Super Insulation for Zero Energy Building’ at Yestermorrow in Vermont. From what I learned in the class I have decided to go with a Larsen Truss wall system, dense pack cellulose (such an amazing insulation), and the airtight drywall approach as my air barrier.
In the class they suggested R20 under the slab, R40 in the walls, and R80 in the roof. To get R40 with dense pack cellulose you need about 12 inches of insulation. Since the main difference in cost between a 12” Larsen Truss wall and a wider one is the additional cellulose (materials and labor to build the trusses should be negligible) when does it make sense to stop? 18”, 24”, when is the wall too thick? Are there structural issues once you make a Larsen Truss wider than 12”? At what point does the payback period for saving on energy usage vs. the cost of the extra insulation become outrageously long?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Before you can do a return on investment analysis you would need to be able to determine the efficiency difference between different R-value assemblies. You can do this with energy modeling.
Some of the most challenging issues with thick walls aren't structural, they are detailing correct water management and fastening the cladding to the structure. I have not designed a building with a Larsen Truss but the advantage I see with this system is that your truss extends out to serve as a nailing base for your cladding.
If you are trying to take advantage of passive solar gains thicker wall assemblies can shade you windows significantly and reduce the amount of solar gain. You should question where you can get the most bang for your buck. Is it increasing the wall thickness are investing in better windows or assuring air tight construction.
Jamie,
Q. "When does it make sense to stop?"
A. In New England, about where you stopped -- R-40 and 12 inches, for a variety of reasons:
- Thick walls limit the amount of light entering the building;
- Very wide window sills seem cave-like are foreboding;
- The extra cost to go to 18 inches of cellulose would be better spent elsewhere.
Jamie,
Your instructors, Bill Hulstrunk and John Unger-Murphy, probably mentioned my wall system in their class, since I've been pioneering the modified Larsen Truss (Riversong Truss) wall for 20 years.
I agree that 12" is a good place to say good enough, and I put only about R-65 or so (18"-20") in the ceiling.
I disagree with the other two, however, that deep window boxes are intimidating (home owners absolutely love them, as do potted plants and cats) or that they reduce either the amount of daylight or passive solar gain.
They may reduce the view, but if a reflective or white surface is used for the deep boxes (I use drywall returns with wood or tiled sills), then they actually increase the amount of diffuse daylight (which is much better than direct glare) and also reflect the solar gain to more surfaces.
The primary limitation for wall thickness is adequate structural and visual support. I always use a foundation as thick as the trussed wall, either a shallow frost-protected grade beam with exterior foam or a 12" Thermomass wall.
More at http://www.builditsolar.com/Projects/SolarHomes/LarsenTruss/LarsenTruss.htm
Jamie,
It sounds like your project is a perfect candidate for PHPP modeling. There are many Passive House Consultants in your area that can help you get your project entered into PHPP http://web.me.com/polytekton/CPHCmassachusetts/massachusetts/massachusetts.html
Even if you decide not to reach for PH Certification, the PHPP software will give you the information you are looking for to make informed decisions regarding your building investment.
You can download a trial version of the software here http://www.passivehouse.us/passiveHouse/DesignTools.html Naturally the tool is at its best when a trained PH consultant is at the controls. Good luck with your project! Will you have a blog to track the progress of your project?
I agree with Robert - thicker walls with outie windows and deep reveals can give a wonderful quality of light. Even better if you splay the reveals as is common in traditional stone and cob buildings, whose walls can be as much as 2' thick. To do that with a wooden truss wall I guess you'd have to oversize the truss opening then furr in for the window R.O. - losing some wall R in the process but you'd probably realize a net gain overall as you could use a smaller window to achieve the same degree of daylighting.
JC: Cost analysis is tricky; what are you going to heat w/ in 20 yrs, and what will it cost? Will wood be available? Oil? Tough one. Go as thick as you can afford, then push it 2". ("afford" meaning money and lack of interior space if the outside dimensions are set in stone).Good luck. j
Robert, Bill and John did mention your system but the link you provided gives me much more detail to go by. Thanks.
Skylar, Thanks for the link to the trial version of the PHPP software. Since I’m not trying to achieve PH Certification I hadn’t considered purchasing the software but would like to see what the trial software says about my design.
John, the house will have a southern orientation with a solar slab. My main heat will be a wood stove (hopefully with all the wood sustainably harvested from my own property) and a ductless mini split as the backup. I know the COP drops in the winter but this site mentions a Mitsubishi model that still gets a COP of 1.8 in 0°F (https://www.greenbuildingadvisor.com/blogs/dept/musings/heating-tight-well-insulated-house).
Robert is correct that with 'outie' windows there would not be a reduction in solar heat gain. That was an oversight on my part as I have dealt with designs where the windows are set midway into the assembly.
Another apparent advantage of the Larsen Truss and 'outie' windows is the flashing details are much easier to integrate with the WRB.
I've worked on a couple of PassivHaus designs. I think trying to model your design in the PHPP will be an informative process however unless you are actually attempting to meet the standards you may not get accurate outputs from modeling your design. The input for the software is extensive although some of the more detailed inputs have less impact on the overall results as the overall square footages R-values etc. If you are not using Passivhaus certified windows or ventilation equipment then it will be more challenging to get meaningful output.
The strength of the PHPP is it accounts for solar heat gain. But you can do a 'heat loss calc' for the entire assembly simply by multiplying U-value, square footage of thermal envelope, and the difference between the design outdoor temperature and the interior temperature that will be maintained in the winter. There is also a factor for outdoor air infiltration which is significant. ASHRAE outlines this calculation and I assume this is the standard formulation behind most domestic energy modeling programs. In this scenario again passive solar heat gains aren't taken into account.
I think the point that the windows can feel 'cavernous' with extra deep assemblies is worth considering. Whether they end up looking like awkward portals or attractive deep set window wells is a matter of design. Larger windows will be fine. The smaller the window the more thought is needed about how it will end up looking, admitting light, and allow views to the outside.
from a design standpoint, i'm generally not a big fan over walls greater than 10" for a sub-1000sf house. they start to feel really clunky beyond that point. the exception would be rammed earth/heavy masonry walls.
a 1000sf house (gross) w/ 10" walls loses about 11-12% of net floor area. a 12" walls loses about 12-14% of net floor area.
this has been my greatest issue with passivhaus, as echoed by the folks at transsolar, etc.
Mike,
As long as your lot is large enough, just build a bigger foundation. (Of course, that increases the foundation cost).
Or, switch from double 2x4 walls to PERSIST with exterior foam. The foam can overhang the foundation.
martin,
most of the projects i work on are urban or semi-urban, so small lots. after setbacks, thicker walls only work to make these lots smaller (not always a bad thing).
i've been an advocate for REMOTE/PERSIST-type systems rather than a larsen truss or double stud wall to achieve similar R-values for a while. as a homeowner, i wouldn't have issues with upcharge for increased net floor area and a better looking/functioning building.
This is one way to understand the ramifications of thick walls, but I believe it's counterproductive.
Is the glass half empty or half full?
With thicker walls (higher R-values with more benign insulation), we gain so many wonderful things, including energy efficiency, ecological responsibility, acoustic isolation, moisture buffering capacity, and the feeling of living in a protective "castle" that the mathematical decrease in interior living space (which is almost invariably far more than is necessary for shelter) is the smallest possible price to pay.