Attic truss vs. shed roof: Cost, complexity, energy and finished living space
![](https://secure.gravatar.com/avatar/4083c71ce872750fd23b3d094f5f89f5?s=52&d=https%3A%2F%2Fimages.greenbuildingadvisor.com%2Fapp%2Favatars%2F200x200%2Fc.png&r=g)
We bought a lot in the Twin Cities, MN and have an architect builder friend helping us with designs. Much of his work is on projects that are paid for by the city and also remodels, so cost estimates are hard to determine.
We are trying to build a simple home (800 sf on first floor) and later convert the second floor to a living space as well. My wife and I both like the coziness of an attic space and do not require a full second floor with 8′ or 9′ walls.
The shed roof would have two sheds. One would sit higher (say sloping west) and be the roof over the living space, allowing fixed lights on the vertical bearing wall above the lower shed roof (sloping east) that would be above the garage. We want the option to have living space above the garage as well some day.
The attic truss option would extend over the garage and living space together until the garage ends (28 feet deep) and then the house extends another 22 feet with another truss design.
Which style would cost less?
Which style would be easier to build?
Which style would be simper to make energy efficient?
Which style would create a better living space?
Thanks
First post
Christian
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb1.png)
![](https://www.greenbuildingadvisor.com/app/themes/greenbuildingadvisor/dist/img/modules/detail-library/thumb2.png)
Replies
Christian, it's hard to visualize the two options you're proposing. Could you upload a drawing or sketch?
It will still be hard to give accurate answers to your questions, as different builders will have different preferences and approaches, and it also depends on your intended performance level.
As Michael Maines pointed out, your question is hard to answer without some drawings, so here's a non-answer ;)
While it will add some additional costs to the design phase, if you are trying to make really cost-optimized design decisions, and make plans for ease of future remodeling, it may be worth considering getting a contractor on board earlier in the design process, paying them for some pre-construction services, and working collaboratively with the contractor and the architect to develop a design. For this approach to be effective, you need a trusted contractor, and your architect and contractor need to be able to work well together. I would consider asking your architect what they think about bringing on a contractor to collaborate with and seeing if they have any recommendations of folks they've worked with before.
The term you're looking for, with a pair of offset shed roofs with a window wall at the ridge, is a 'Clerestory Roof'. They're beautiful, but difficult to do efficiently for conditioned spaces.
Unfortunately it sounds like you want to do architecturally the opposite of what is commonly advised for climates that suffer extreme cold, like MN.
Per WP:
> Owing to its northerly latitude and inland location, the Twin Cities experience the coldest climate of any major metropolitan area in the United States. However, due to its southern location in the state and aided further by the urban heat island, the Twin Cities is one of the warmest locations in Minnesota. The average annual temperature at the Minneapolis–Saint Paul International Airport is 45.4 °F (7.4 °C); 3.5 °F (1.9 °C) colder than Winona, Minnesota, and 8.8 °F (4.9 °C) warmer than Roseau, Minnesota.[41] Monthly average daily high temperatures range from 21.9 °F (−5.6 °C) in January to 83.3 °F (28.5 °C) in July; the average daily minimum temperatures for the two months are 4.3 °F (−15.4 °C) and 63.0 °F (17.2 °C) respectively.
> Minimum temperatures of 0 °F (−18 °C) or lower are seen on an average of 29.7 days per year, and 76.2 days do not have a maximum temperature exceeding the freezing point. Temperatures above 90 °F (32 °C) occur an average of 15 times per year. High temperatures above 100 °F (38 °C) have been common in recent years; the last occurrence was on July 6, 2012. The lowest temperature ever reported at the Minneapolis–Saint Paul International Airport was −34 °F (−37 °C) on January 22, 1936; the highest, 108 °F (42 °C), was reported on July 14 of the same year. Early settlement records at Fort Snelling show temperatures as low as −42 °F (−41 °C). Recent records include −40 °F (−40 °C) at Vadnais Lake on February 2, 1996 (National Climatic Data Center)
You purchased a lot in a bizarrely extreme continental climate that humans have no place living unaided by technology. As such, if you want a durable, efficient house, ideally we would employ all the tricks in the trade.
Trick 1 is a fully vented attic with good air-sealing. That means sacrificing the area between the eaves of the roof and the ridge of the roof, and dedicating it to a single purpose: holding a big bathtub of cheap cellulose insulation, two feet thick. It means not even using this space for storage - because sealing access hatches is a nightmare. It means if you want can lights or fans below that level, you would ideally frame in a whole floor platform underneath the insulation, so that you can get that pristine seal and not leak much air up through the ceiling.
Clerestory roofs are a variant on cathedral roofs with some complications; Reverse-lapping just below the clerestory wall is a common issue (so you can't make the slope too shallow), and structurally you're competing between showing structure (which blocks some light and can look ugly) and supporting the roof planes adequately for large snow loads.
Cathedral roofs in general have a lot of issues. They're expensive and finicky to get properly detailed, and leaks, condensation, and air sealing issues are not rare. All roofs benefit from being steeper in terms of shedding water, so the optimal cathedral roof is fairly steep, but this gets expensive and inconvenient fast - the steeper it is the more material you add and the more surface area you have to protect.
Cathedral roofs in cold climates have a serious additional issue: Ice damming. Ice damming is when the underside of the snow pack on a roof gets warm enough to melt, which trickles down the roof until it finds a crack in your waterproofing or it reaches the eaves. The eaves aren't insulated, so they're cold; Since the water is just north of freezing, when it reaches this level it freezes, creating a dam. Ice dams can get huge and heavy and dangerous if they fall; They create spectacular icicles. Behind the ice dam is a reservoir of water soaked into the snow. If it gets far back enough, it may find someplace it can get into the attic, and thus into the walls and ceiling of the house. This may be immediately visible to you, or it may find a place to drain and merely saturate your sheathing & studs, setting you up for rot in the springtime.
How do you deal with this? You want a way to keep the roof cladding cold. You could just add tons of insulation, but this is expensive and it has its limits: Since snow itself is a good thermal insulator, and we're heating it some infinitesimal amount from the bottom, there will be a gradient from 70F inside temperatures to 0F outside temperatures, and if there's enough snow-insulation competing with your roof-insulation, the ratio can easily push the bottom of the snow pack above freezing.
What we prefer to do instead is insulate the hell out of an intermediate surface, then have an airspace above that surface where outdoor air is free to flow from eave to ridge, and then have the roof cladding above that airspace. In the vented attic model, the attic serves as this small wind tunnel, with air continuously flowing from eave soffits to the ridge vent. In a traditional cathedral roof, you can't do this; You've pushed the insulation right up against the underside of the roof sheathing. So what's recommended instead with a cathedral roof is to install wooden spacers ("battens") between the roof sheathing and the roof cladding (shingles/etc), or possibly install a second layer of roof sheathing above those battens, and then vent the airspace it from eave to ridge. This can be tough to do, because every nail you lay down is going to interfere at least a little bit with your weather-sealing, and because of the possibility of "reverse lapping" situations wherein you make a ramp for the rain to go into your home directly. One of the big downsides of such a "Vented over-roof" is what happens when the roof leaks. Unlike in a vented attic, there's no place to stand where you can actually find out where that leak originates, your only option is to start tearing off drywall and insulation from below.
Clerestory roofs are even harder over conditioned spaces in cold climates, because you're opening up very-poor-thermal-insulators ("windows") in your roof, and hoping they don't melt any snow, while also hoping they don't leak or condense any water and drip into the middle of the room below.
https://www.youtube.com/watch?v=Ld8pzIu45F8
https://www.greenbuildingadvisor.com/question/ice-dams-on-almost-complete-new-construction-house
https://www.greenbuildingadvisor.com/article/martins-ten-rules-of-roof-design & critique http://thehtrc.com/2011/ten-rules-of-roof-design-review
One of my new favorites on basic good ideas for new construction is Martin's new talk:
https://www.youtube.com/watch?v=OZ3uPiGwHhA
Additionally: One of the things you start to see is that retrofit construction is a nightmare. Energy retrofits don't have anywhere near the same criteria for success as a new-construction home, because there are so many things that get broken or elided past the transition which are harder to fix than knocking the house down and rebuilding. House additions are a key place where good building practices get bypassed by people separated by a decade or two temporally who just have different ideas and levels of knowledge about how things work. A house will happily leak ten times as much heat from the room you added as from the rest of the house combined, if you forget to insulate one wall, or screw up the air-sealing strategy, or leave a flue space next to a duct open to an unconditioned attic.
Similarly: Changing an attic to add living space down the line is hugely sub-optimal. Too many opportunities to screw up. Adding things like dormers to get that natural light and egress window for an attic bedroom interferes with roof venting and sets you up for a number of potential problems.
Best practice:
Design for the largest exterior envelope you could see yourself possibly using. If you need to for budget purposes, leave an upper floor unfinished inside (skimp on trim, leave it to subfloor, don't paint, maybe even skip the drywall if you have secondary airsealing), but get the framing, HVAC, and electrical put in, and get the exterior insulation in its final form. Those second-floor walls are cheap *right now*, but they can be much more expensive to add in later.
Combining garage and house is its own series of posts. Code wants you to keep them as separate as possible HVAC-wise because of the risk of carbon monoxide and paint thinner leaking from that area into your kid's bedroom through ducting. It often prescribes a fire-wall, a complete ventilation separation, and even a concrete-step-up (to stop flammable fluids) between them.
Thermally, if we're conditioning the garage, we would want to keep conditioned spaces as closely connected as possible, leaving less surface area to leak heat, and use a separate mini-split or something. On the other hand... it can be tough to condition a garage at all. The big problem is the garage door. Normal segmented sliding-down garage doors leak tons of air and are uninsulated; Upgrades are better-insulated, but the air leakage renders this superfluous, and is fundamental to that type of door. People say carriage doors ("A massively oversized version of your front door") can work better air-sealing-wise, but they come with a large price premium and are often done custom.
If you're going to do an *unconditioned* garage, the last thing you'd want to do either thermally or from an air quality perspective is to put living space directly above it, and try to seal five of the six sides of this room. This is nevertheless common in more forgiving climates, for lower-performance homes.
The worse your insulation & more air your house leaks, the more fuel you're going to burn. But at the same time, the better your interior air quality and the more moisture issues solve themselves through evaporation. A house from 150 years ago might be uninsulated or negligibly insulated, but the framing lumber is fine because we've been using the house as a dehydrator by burning huge amounts of fuel in it; A century and a half of nosebleeds and zero condensation issues. When you make a house efficient by adding insulation, you have to deal with the prospect that even minor moisture sources could get trapped in those walls. Air-sealing is partly an attempt to reduce thermal losses, but also partly an attempt to address those moisture sources and prevent them from damaging your house.
Thanks for the help. We met with our architect and are considering a "flat" roof to simplify the build, optimize interior space and allow for simpler insulation.
When I look out my window, 90% of the houses around me have had the attic converted into living space over the years. To me, an unusable attic is a huge waste. Even if you don't use it for living space, it makes for much better storage than a basement. If I was building new and didn't need all the space, I would definable build the attic to be usable down the road.
Overall, to keep costs down, you want to stick to simple standard construction methods. Keeping your house envelope simple will go a long way in making it cheap to build and energy efficient.
This does generally mean a simple box shape with a gabled or shed roof. For structure, you want the roof to be either trusses or TJIs. If you want it be be cheap and work well, than you are looking at a vented roof with blown in insulation.
If the inside is cathedraled, you want the slope to be low enough for blown in insulation (I think it is around 4 in 12), this avoids any of the expensive insulation (SPF, rigid or dense pack).
Adding in a clerestory is definable a nice thing, but it will be more expensive to build. My current home has it and I consider it one of the central features of the place. If you are careful with your slopes and clearances, it can be doable as a vented roof, but the lower sloped section is usually easier to do as an exterior rigid+interior batts un-vented roof.
One thing I did is to move the structural supports for the clerestory inside the house behind the windows and used a wider top and bottom plate. This allowed the windows could be installed as one big array with the sections clipped together. This saved a lot of trim/detail work and gave more glass area.
When discussing with architect be careful with what you ask for. Their job is to provide design you, the client, asks for but they are generally not fully aware of the costs/energy implications. A simple looking detail could easily be a $10k line item on your build. Doesn't take a lot of these small "details" to completely blow your budget. Getting your builder involved early in the design phase is important.
> We met with our architect and are considering a "flat" roof to simplify the build, optimize interior space and allow for simpler insulation.
Did they provide a reason why? A flat roof is not optimal in a cold climate, and requires skilled installers and a knowledgeable designer working together.
Are you trying to squeeze under a code-restricted building height?
If not:
A simple gable roof of moderate pitch over a vented attic made of trusses is the bargain in this design space. Extra walls (a second floor beneath that attic) are cheap.
Akos:
I have met the enemy, and he is us.
> When I look out my window, 90% of the houses around me have had the attic converted into living space over the years. To me, an unusable attic is a huge waste. Even if you don't use it for living space, it makes for much better storage than a basement. If I was building new and didn't need all the space, I would definable build the attic to be usable down the road.
You are entirely correct. This is how it looks to a homeowner well after build time. When you're mapping out a green building strategy, though, you have to ask: How many problems is a door into the attic going to cause? How many problems is walking on the bottom chord of a truss going to cause? How do I even make a frame that's high enough off of the bottom member so that I'm not trudging through two feet of blown-in cellulose insulation? Is this framing going to dramatically increase heat transfer? What would it cost me to mitigate these issues? How hard is it going to be to explain the plan to the builders and ensure they follow through? How fragile is all this going to be for the homeowner? Is the homeowner and the airsealing that I am making for him going to be *safe* (not cause structural failure) cramming hundreds of pounds per square foot of storage stuff up there? Attic trusses cost around three times as much as standard trusses for equivalent loading. Any kind of storage space up there invites the homeowner to pretend it's a liveable room (just like our non-waterproofed midcentury basements!). Then they want to add expensive dormers and screw up the roofline with shallow valleys.
Moving all of our crap (storage, mechanicals, ducting, lighting) inside of the carefully designed conditioned envelope, giving the homeowner a whole extra floor by building them proper walls (or constructing a wider building footprint), and strongly discouraging the homeowner from using the attic (eg by not giving them an entryway except through roof access) seems like it makes more sense. If you're going to build storage space, build it as an actual floor. See also https://www.greenbuildingadvisor.com/article/all-about-attics https://www.greenbuildingadvisor.com/question/gable-access-to-attic
If you absolutely must store things up there and you're not going to be tempted to live there, change the whole strategy. Forget trusses. Build the whole thing from huge I-joist rafters with a proper I-joist floor, a glulam ridge beam, and some structural reinforcement to the gable endwall to hold it. A bit more expensive and requires a skilled carpenter and additional time with the structure exposed to weather to frame, but you could plausibly construct a sizable room up there and not have to worry so much about bringing the house down around you, or landing a foot wrong and falling into a hallway (a family member had this privilege).
I find it hard to understand what you are planning without a drawing. But if you are talking about a story and a half they are energy nightmare and not the greenest design.
I say borrow the money at today’s very low rates and build a two story now. I say cut you budget and go with low end finishes that you can upgrade later when the budget allows.
If not build with the plan that you will remove the roof add the full second story and put the old trusses back on top.
Walta
I have a two shed, one facing south (2:12 pitch, second story, 7.5' to 11' south to north) and one facing north (2:12 pitch, first story, 9' to 13' north to south). The second story is living space above the garage, everything is spray foamed, some foam and batts, and standing seam metal roofing. There's about a 4' vertical overlap and some conductive heat transfer, which heats the second story. As you can tell, no attic and the 6x12 beams (running north-south) are without thermal breaks, suboptimal, but aesthetic. You may wish to design with solar in mind, which means snow load + solar cell load + direction + pitch. I've lived in the Twin Cities, not sure I would recommend a flat roof for residential due to snow and/or rain. We had one rain storm that flooded the streets in less than 30 minutes (cars inundated). And, I remember January rain followed by -20 F in less than 12 hours, some streets were 1' thick in ice.