Tiny house envelope questions
Hello GBA community,
I am starting the design process for a tiny house which will be mobile. The current idea is to build a very light and very small (72-100 sqft) off grid home for myself. I live in Minnesota (Climate Zone 6) but I would also like to take this home to other climates potentially for extended periods.
My current question is a broad one, I’m looking for resources, ideas, input on the structure and building envelope. My initial thought is steel stud construction along with steel panel siding and roof, exterior rigid mineral wool insulation. Strong, light weight, simple construction.
I build for a living, but not with steel so I am very interested in info on designing structures with steel studs. I am also wondering about what amount of insulation would be optimal, this is a tiny space, but I do plan to use solar for electric heat if possible along with passive heat from windows. Another big question for me is how to achieve continuous control layers on a trailer? I’m very keen on implementing perfect wall principles.
Please suggest any previous threads on GBA or elsewhere, and any articles, resources, etc… Thank you!
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Derek,
An off-grid home can't use electricity as a space heating fuel. For more information, see "How to Design an Off-Grid House."
Thanks Martin, just read the article and looking at Backwoods Solar now. Congratulations on 40+ years off-grid!
Steel readily allows heat conduction, R=0.0031. Bigger R-values are less conductive. Wood studs are R-1.25 per inch, so 2x4 studs are R-4+, far less heat loss than R-0.003. Fiberglass insulation batts between the studs are R-11, or higher for higher density fiberglass batts. Roxul ComfortBatt mineral wool batts are R-15. So using steel studs would pretty much defeat using insulation between those studs. Too much heat would escape via steel studs. In a portable tiny house, you likely want walls to be thin (take less space) yet well-insulated for those very cold Minnesota winters. Wood studs with ComfortBatts or high density fiberglass batts would be a good choice unless you have other criteria important to you.
If you want higher insulation levels, 2x6 wood studs and R-23 ComfortBatts would work better but take up more space (thicker walls).
Personally I'd be concerned about wind effects, assuming you are creating a trailer that would be subject to travel speeds of 65+ mph plus the winds present during travel. 1/2" exterior plywood sheathing or equivalent would be more structurally sound for wind's high racking forces. You could consider interior continuous insulation from rigid mineral wool, but if used on the exterior, would complicate the exterior structure, perhaps making it more susceptible to wind shear.
Overall, you need to compute the total heat loss from your home. The better insulated and airtight, the less heat needed in winter. That includes quality windows and doors. Two-thirds of the time during winter, its dark outside, and you likely using window coverings (curtains, shades, or better-insulating window quilts or rigid insulated panels). Upgrade to higher R-value window coverings. Your window area likely needs to be higher on one longer side of the home facing south for wintertime solar gain, and the trailer parked with that side facing north during summers to avoid solar gain when its hot outside. You probably need more window glazing area overall to make you feel less cooped up in a tiny home.
Calculating total heat loss for average and "design" (coldest) temps is not difficult, especially for a small, simpler structure like you are designing. You can do this yourself with a spreadsheet (like I do) or use various web-based sources. Others can suggest good web-based sources. The results would indicate how big a heating/cooling system needs to provide. If you plan on staying somewhere where AC power is available to you, electric heating is an option. As Martin noted, to be off-grid you will never get enough electricity from solar panels to heat your home, but direct solar heat gain from windows can be substantial. Unfortunately you would need propane heating in the winter, unless you are very willing to accept low interior temps or move south every winter.
Steel exterior cladding would be fine, as it doesn't transmit heat into the interior of your home.
Insulating the floor of a trailer gets complicated, assuming you have a steel trailer. Insulation between the steel "joists" structure in the floor is pretty much a waste, as the steel transmits too much heat past the insulation.
You might install batts between wood floor and ceiling joists, maybe with additional continuous insulation if you prefer and have sufficient space. Keeping the structure roadworthy, avoiding getting too wide (>8') or too tall can be a challenge.
Hi Robert,
Thanks so much for your detailed reply!
I am aware of the thermal issues around steel, and have been thinking and reading about the options. Thanks to Building Science Corp I am aware that basically there is little to no benefit of insulating between steel studs. What I had in mind was a frame built of steel, with sheathing (plywood or equivalent), water & vapor barrier, wrapped in rigid mineral wool, furring strips attached to studs and steel siding exterior.
As you said wrapping this around the underside of the trailer and dealing with the frame gets tricky and perhaps your suggestion of continuous interior insulation is the way to go. I went to a trailer dealer yesterday and learned that an aluminum trailer may be viable, additionally I looked at aluminum enclosed cargo trailers. These are very interesting because the entire frame is basically and aluminum cage, with exterior aluminum panel. The more I think about it what I am talking about building is more like an ultra-light RV trailer than a typical tiny house.
I would be interested in any resources you can point me to on creating a spreadsheet to calculate heat loss. The way I am thinking about this is to create a building envelope that gives me the best balance between insulation, size, weight, structure, etc... Then working out the basic shape size of the structure and the openings.
Using one side as the primary window side like you suggested is a good idea, like you mentioned facing south in the winter, north in the summer. I'm not super concerned with having tons of windows and feeling cooped up. I work on houses for a living and spend a good amount of time working outside. In the summer I plan to use the tiny house as shelter, but also hope to find parking places that allow me to use outside space for dining, relaxing, cooking at least part of the time. In the winter, I work most of the daylight hours, but that said a nice large window for lazy Sunday reading will definitely make the space more enjoyable.
Derek,
The downside of exterior insulation and empty stud bays is the thickness of the resulting walls. In most projects it doesn't matter that much, but in a tiny house where floor space is at a premium, it doesn't make much sense.
Exterior insulation also makes the connection between the framing and cladding less robust. Not something you want for a structure subject to high winds.
Tiny houses have been around for quite a while now. I'd look at existing ones and see what works, rather than inventing a new method of construction.
Thanks Malcolm,
I completely agree about the empty stud bays. Doing some research on trailer and camper manufacturing I have found they are using 1” square tubing aluminum framing with aluminum exterior sheathing/siding. For wood frames 2x2 walls and simple arched trusses are used. The trusses are basically a 2x2 joist with a 2x2 arched over a few blocks, about 4-5” at the peak. The Wood framing is simply connected by nailing plates. This is current industry standard and holding up decently in the real world.
What’s interesting to me about this is that the wall framing remains decently thin, not 3.5” thick empty framing.
I am potentially able to build 1’ total wall assemblies. 8’ exterior width trailer, 6’ interior. I’m not saying I am definitely going that route but I am exploring the option.
There is all sorts of styles of construction for tiny homes, but a lot of what I have seen does not really address building science principles, especially thermal bridging. I have seen people spraying or stuffing steel stud bays thinking they are doing a great job insulating. Just like traditional home building there is a lot of potential for building science to improve these structures.
A tiny house is inefficient on the volume/surface area front, but it is TINY.
So that means, even an inefficient structure, will use little energy.
There is little to no energy saved in my area (zone 5) above a R20 whole assembly wall. With a small footprint, giving up 2' of space just for insulation is silly.
Take a 200sqft 9'tall structure (~600sqft of outside surface). In my area, at design temp, with R20 walls/floor/roof that would only loose 2000BTU. Bumping that up to R35 would drops it to 1200BTU but you also loose 17sqft of space. So you are sacrificing almost 10% of your floor space to reduce the heat load by the amount of heat from a fridge. It just doesn't make sense.
Maybe insulating more the floor for a bit of extra comfort, but outside of that the math on thick insulation just doesn't work.
Working with steel is not too hard. The challenge is that it needs some special fasteners (plywood to metal and larger self tappers) which you won't find at the box stores but are readily available at commercial suppliers.
The good part is that if you design the structure up front, you can get all your studs pre-made exactly to length. This means there is no additional cutting on site, very quick build.
The studs are always straight, very nice for taller walls. You can also do moment connections, which means you can do a gabled roof without collar ties.
They are also a bit lighter than equivalent dimensional lumber, definitely noticeable for larger sizes.
As others have said, thermal bridging of the steel reduces wall R values.
I've have used these in the past:
https://www.steelform.us/documents/Delta%20Stud%20Table%20Commentary.pdf
The extra punch outs reduce some of the thermal bridging, not as good as wood but much better than standard studs.
Attaching cabinetry to steel studs is not a problem provided you have the right fasteners (as I mentioned above). Sometimes it is easier to add in some wood blocking if you know exactly where things will go.
Hi Akos,
Thanks for your reply here and your other reply above. I'm interested in learning and experimenting with steel more and glad to here you have had success with it. I work with wood every day but I have had to construct details in houses that required steel studs around gas fire places. It was quite interesting to be able to build a 12' tall wall and pick up and set it in place with little effort.
I also appreciate your point above in your second comment about the wall thickness and practical insulation level. I don't think I will end up doing a full 1' thick wall but like you suggested there is a sweet spot where the balance of BTUs, cost and practicality align. I'm excited to use Roberts spreadsheet idea to work out that balance.
A more thorough description of this can be found in Martin Holladay's articles:
https://www.greenbuildingadvisor.com/article/how-to-perform-a-heat-loss-calculation-part-1
Heat loss is a function of resistance to heat flow (R-value), area (A), temp difference between the indoor temp (Ti) and outdoor temp (To), and time (T). This is probably obvious, but...the higher the R-value, the less heat gets through. THe bigger the area, the more total heat is lost through the bigger area. The larger the temp difference, the more heat is lost. The longer the time, the more heat is lost. The formula is:
BTUs = A x (Ti - To) x T / R
Area is what you measure (or are planning) in SQFT.
Temp difference is degrees F outdoors (e.g., 30F in winter) and desired temp indoors (e.g., 68F in winter). You should look up the average January temperature for your location, and the design temperature. The design temp is the coldest temperature expected for your location. For example, you can look these up here:
https://up.codes/viewer/connecticut/ipc-2015/chapter/D/degree-day-and-design-temperatures#D
You can use R-values that you find on reliable sources the Internet, such as:
http://www.coloradoenergy.org/procorner/stuff/r-values.htm
or
https://www.archtoolbox.com/materials-systems/thermal-moisture-protection/rvalues.html
or
https://en.wikipedia.org/wiki/R-value_(insulation)
Time typically would be per hour (1) or per day (24), depending upon how long you want to calculate the total number of BTUs of heat loss. You can look up the amount of solar heat gain per hour or per day, or the BTU capacity of whatever propane heater or electrical heater you may decide to use. The usual cheap portable plug-in 120V electrical heaters, for consumers to use in homes, throw off about 5,000 BTUs/hr.
Unfortunately it gets tedious computing the total heat losses for a room or home, because you must analyze the heat loss through different components that face the outdoors...walls, ceilings, floors, windows, doors, everything that is different. That's one reason spreadsheets are nice to do all these separate calculations, then sum up to get the total.
You could use metric units instead for all these variables, if you live in a more logical country vs. using the odd traditional measurements the Brits gave us. :-)
Normally you use the average winter temp in January (coldest month) to determine things like your heating costs and insulation levels desired, but you use the design temp for your heating system capacity, so that you can handle the coldest expected winter weather. Similar for cooling but you indoor temp likely around 76F and outdoor design temp might be 90F.
For a spreadsheet, use Microsoft Office Excel if you have it, or Numbers spreadsheet free on Apple Mac computers, or Google Sheets for PCs:
https://www.google.com/sheets/about/
In case you haven't used them, spreadsheets are just a 2-dimensional table of blank boxes. You type in numbers or words in some of the boxes, and can use arithmetic operations to create sub-totals, totals, or labels. There are built in formulas to compute averages etc. You can use spreadsheets to solve equations, or create nice looking charts. Not difficult but takes a little time to learn how to create nice looking tables with the data calculated the way you want.
Once you start using spreadsheets, you'll likely use them for other computations, or to create nice looking tables and charts. For example, see:
https://www.greenbuildingadvisor.com/article/a-quantitative-look-at-solar-heat-gain
Robert thanks so much for the info. As soon as I read this I started imagining the sheets I could build to compare different assemblies side by side. Needless to say I love a good spreadsheet and I think this will be the perfect way to balance out the R-value, cost and construction process structure. I'll definitely post some comparisons here once I get that up and running.
Hello Derek. Small/tiny homes have a high surface area to volume ratio. Thus thermal bridging becomes a significant source of heat transfer quickly. Heat loss, simply from the high surface area to volume ratio of the tiny house form factor will be a big issue in your area.
I am also building a tiny house on a trailer, 187 sq ft. exterior, ~153 sq interior, in Portland, OR which is not nearly as cold, climate zone 4c. To achieve nearly zero thermal bridges, I am using structural insulated panels (SIPs) that are ~3.5" thick with 1" of exterior continuous insulation and 1" ThermalBuck around the window rough openings. I am also importing European triple glass windows with insulated and thermally broken UPVC frames. It is pretty overkill for my climate zone 4c but likely not overkill for your climate zone. Reducing heat loss by building a very airtight envelope will also be a challenge. Airtightness is even more important than insulation for a number of reasons.
For heating and cooling, I have specified a very efficient 42 SEER/15 HSPF ductless mini-split. Carrier and Bryant both sell this mini-split under the names INFINITY and Evolution respectively, although I am certain both of these manufacturers are simply rebadging a heat pump made somewhere in Asia. Modern heat pumps can do 100% Heating Capacity at 0° F (-17° C) and a heating operating range down to -22° F (-25° C) but they are hardly efficient at those temperatures. In Portland, OR it never comes close to getting that cold, but in your area, Minnesota, it does dip down to those temperatures. How often does it dip down to 22° F (-25° C) in your part of Minnesota? And for how long? If it is just a few dozen nights a year, you could likely still do a ductless heat pump and sleep under an electric blanket/heating pad that uses only 100-200 watts.
Solar and batteries have come a long way. I will be going off-grid using eight 300-watt panels each running on its own Enphase microinverters. Those microinverters will sync with the sine wave generated by a battery inverter. I will be using 2.4 kWh of LiFePO4 batteries. I might need more batteries than that but I'll start with that much.
Indoor air quality is a concern in such a small space. I'm opting for zero combustion products. Induction cooktop, heat pump water heater, ventless heat pump washer/dryer combo, and ductless heat pump mini-split. I will also be running a 12v fridge and 12v LED lights so I can schedule shutdown the inverter at night. Will my 2.4 kW of solar and 2.4 kWh of LiFePO4 batteries be enough to run an all-electric tiny house? My energy modeling says yes, but I don't necessarily trust the software to be that accurate for tiny homes that have very high surface area to volume ratios.
Hi Edward, Thanks for your input! I'm glad to hear from a fellow tiny builder out there also pushing the "envelope" of green building techniques.
Your project with SIP panels seems like a great way to build a high R value wall that is strong, simple and light weight. I had considered sips myself and I may look into them more. Would you mind sharing where your sourcing those? I'd also be interested to learn more about the windows your using. If I keep my window package small I may be able to afford something similar and if I'm going to build a high R-value wall I should definitely not negate that with leaky windows.
Great idea on the heat pump, I will have to look into the number of days below 22° F (-25° C) in my area to see if that is an option. Perhaps I could use that primarily and invest in simple back up solution like you suggested. In my case I would likely use a combustible back up as I do plan to spend some time up in the farther North during the winter. There is something magical about being out in the middle of nowhere when its -20° F and everything is covered in white crystals of snow, but it is also very dangerous to be without heat.
I'm going to do more research on solar once I get my general structure worked out. Do you have any suggestions of sources to find more info?
You need strength and low weight. I suggest using 2x4s at 16 or 24 on center. Then use mineral fiber in the stud bays, and continuous foam insulation under the floor and a rigid foam insulated cathedral roof. Vinyl flooring inside (or laminate) is likely the lightest choices.
Alan,
I agree.
Hi Alan, Thanks for the input. I have been doing "research", watching factory tours of camper trailer manufactures. It is amazing to me, but they are building very thin (2x2") walls and these are holding up, mostly because of the continuous metal siding which also acts as the sheathing. I think there is definitely something to this that I could incorporate. Its interesting to have the challenge to build a structure that can withstand hurricane force winds and earthquake like shock and also keep it ultra light weight, yet this is what every RV trailer manufacturer does every day.
I definitely have been considering the cathedral roof to give some more head space. I also think Vinyl may be a good idea too. I just finished a basement with vinyl and was really happy with how it turned out. The simplicity of install and durability is a great advantage.
Derek,
But isn't the big knock on RVs that they are lipstick on a pig? They are designed and built for appearance, but deteriorate at an alarming rate. I thought the idea behind building tiny houses was to get something more approximating the quality of a conventionally built house. Otherwise, why not just buy an RV?
An engineer can calculate loads and building for natural disasters such as hurricanes is a thing (i think GBA even had an article or few on this). Racking resistance does come from the sheathing or even metal, but without the ability to do the math your not going to get very far. I am not qualified to make such calculations but maybe you could find an engineer who is?
That said tipover is a risk, i cannot calculate its likelihood either.
Metal is heavy, metal strong enough to make a difference is heavy. I don't know of engineers who specialize in tiny houses, if any exist, but there should be those whose skills are relatively transferable.
Also bear in mind your probably not going to be able to get house insurance so make sure your not spending any money you care about losing, and that also includes content insurance, your not likely to find any. A fire will undo any strength you add, even if its metal sheathing.
I had attended a conference on tiny house building many years ago, the presenter used 2x4s and fiberglass, and you can build full sized houses with 2x4s, so for a tiny house i would use them with plywood and mineral fiber. And bear in mind thermal bridging.
I would be interested in seeing heat/cooling loads for a tiny house, there must be someone who could run them for you.