GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Insulating miners’ containers – Ulaanbaatar, Mongolia

albertrooks | Posted in Energy Efficiency and Durability on

Andy lives in Ulaanbaatar, Mongolia, the coldest capital in the world. Right now night time temperatures are hitting -40 F and C. He is an ecologist, fly fishing guide, and now learning green building / energy efficient building / sustainable design. He is doing some work with a local construction firm and is trying to figure out better building methods for his climate.

Andy came to us with a real poser of a question that I became uncomfortable with. We were just helping him with ways to attach insulation to a shipping container, but when I looked at the layers, I became nervous, and now, I’ve made him nervous about it too.

I’d like to help him by getting a few well thought out opinions on this since it’s in such unforgiving conditions and the occupants need to rely on it’s safety.

There is a need to insulate a number of shipping containers for mining and construction sites. As you can guess, there is a limited supply of materials. Andy can get access to XPS or Mineral wool. No spray foam or much else.

His initial thought was 4″ of XPS inside the container with an air/vapor barrier at the face. My concern is that any air+vapor that leaks through a broken barrier will become trapped behind the XPS and grown mold during times when it’s not frozen solid to the steel. Remember the irregular shape of container walls. Those voids would be full of air since we don’t have spray foam.

Would it be better to use mineral wool with with no vapor barrier and let it dry to the interior?

Does any one have any experience in making habitable shipping containers in extreme climates?

All comments appreciated.

Albert Rooks
The Small Planet Workshop

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. GBA Editor
    Martin Holladay | | #1

    I lived in a shipping container when I was doing earthquake-relief work in the mountains of Armenia, and I've thought about this question a lot.

    It's possible to insulate a shipping container on on the interior with XPS, but you would need to pay a lot of attention to air sealing. Moreover, steel shipping containers are always expanding and contracting with changes in temperature, making air sealing challenging.

    The best solution, of course, is to insulate on the exterior, but that requires adding roofing and siding.

  2. GBA Editor
    Martin Holladay | | #2

    Further thoughts on insulating shipping containers:
    These homes are probably temporary, so minor flaws (for example, occasional condensation) may not matter if the insulation works "pretty good."
    The shipping containers are being used for housing because they are available, not because they are ideal. Probably the miners and construction workers who live in them will be more forgiving of flaws than American homeowners.
    If the disadvantages of shipping containers start to outweigh the advantages, think about other ways to provide housing.

  3. user-757117 | | #3

    Interesting concept...
    Unlike Martin, I haven't really ever thought about it...
    But wouldn't a steel shipping container already be quite "tight"?

    If so maybe an exterior insulation, vapour open assembly is better?

    A housewrap or felt-like membrane to act as a half-decent "wind breaker" over mineral wool...
    Or maybe just cladding with "pressure moderation" over mineral wool...
    Cladding could be anything...
    Corrugated steel/tin for example...

  4. GBA Editor
    Martin Holladay | | #4

    Yes, shipping containers are tight. When converted to housing, these containers are usually retrofitted with doors and windows that allow ventilation.

    If you try to insulate the interior of a shipping container with fibrous insulation, the warm humid air in the container will migrate to the cold steel, leading to condensation. Eventually the ice will melt and drip to the floor.

    You're right that exterior insulation is best. But by the time you buy the roofing and cladding, you lose some of the savings you hoped to achieve by starting with a shipping container.

  5. user-757117 | | #5

    I wonder what other materials are available "for free"?
    Are there any nearby dumps or salvage yards?
    Or maybe abandoned or disused facilities that could be disassembled for parts?

    Old warehouses sometimes have plenty of old panels (steel, polycarbonate etc.) that might make functional cladding.

  6. GBA Editor
    Martin Holladay | | #6

    At 40 below zero, that drain will freeze. (Actually, the drain will freeze well before the thermometer hits 40 below...)

  7. user941025 | | #7

    If you aren't doing the mineral wool solution, couldn't you run channels with a shallow slope just above the floor, in a small gap between the steel and the XPS, to drain the condensate to the container openings?

    Edit: or alternatively, drain the condensate to be used as drinkable water?

  8. user941025 | | #8

    Give me one uncharacteristically warm winter and I forget the obvious.

    Still: draining condensate to a warm collection point: an openable insulated/gasketed panel from interior or exterior.

    What's wrong with the mineral wool solution?

  9. homedesign | | #9

    I did a Google-Earth Fly-over
    looks Like Yurts are very popular there

  10. user-626934 | | #10

    Albert -

    WWTS? "What Would Thorsten Say?" - put all the insulation to the outside. Also, as I've heard many do-it-yourselfers in the Northeast know - taped XPS can be effective "cladding" for a number of years.

  11. GBA Editor
    Martin Holladay | | #11

    Q. "What's wrong with the mineral wool solution?"

    A. Mineral wool on the inside results in condensation on the cold steel. Mineral wool on the outside needs wind-washing protection, roofing, and siding.

  12. user941025 | | #12

    Understood, Martin, and as I'm referring to the solution proposed in the OP ("Would it be better to use mineral wool with with no vapor barrier and let it dry to the interior?") is drying to the interior unlikely or insufficient?

  13. GBA Editor
    Martin Holladay | | #13

    As long as the interior is warmer and more humid than the cold steel, there will be continuous condensation on the steel all winter long. It won't be drying to the interior; it will be puddling on the floor.

  14. albertrooks | | #14


    Yes, that's what I was worrying about. Even with dehumidification, the humidity could spike to quickly and go right to the steel. It seems that this is just an extreme case of a cold climate wall. Place the air and vapor pressure boundary at the inside of the insulation. Nothing new.

    I think you've made the point that vapor hitting the steel will condense quickly and not have anywhere to go... The only solution is to keep it in the room and away from the steel. The XPS route is sounding like the most reliable method. I can (or Andy can) tape to XPS well to make a good flexible joint.

    The only other thought that I had was to turn the the steel into a ventilated rain screen to give it some venting in each channel. It's probably "too fancy" and would just plug up as you pointed out unless they were pretty big cuts.

    I really appreciate the comments.

  15. albertrooks | | #15

    John Semmelhack:

    WWTS? "What Would Thorsten Say?" - put all the insulation to the outside. Also, as I've heard many do-it-yourselfers in the Northeast know - taped XPS can be effective "cladding" for a number of years.

    Good call. I think that would be his thoughts also.

    The whole working on the inside vs. outside is based on portability and security.

  16. MICHAEL CHANDLER | | #16

    I was talking with a Tyvek representative at IBS last week who said they are constantly wrestling with the thought of putting out a camo or clapboard pattern product as they see homes under construction that have had the house wrap patched numerous times over very extended build times. (they can date Tyvek from the street by the logo updates, she said she had photos of one home with three or four different logos representing over ten years of putting off the siding project)

    Put the foam on the outside, tape it and wrap it with tar paper or roll roofing, skip the windows and put in a wide screen TV with a live feed from a warm climate.

  17. GBA Editor
    Martin Holladay | | #17

    You've got to have a few windows for ventilation. Containers are airtight, and they have a low ceiling.

  18. user-757117 | | #18

    I came across this website but have no idea if it is of any use:

    It may not be easy to properly insulate the exterior of a shipping container with XPS...
    What thickness XPS would you use?
    Maybe the containers in question don't have such an irregular surface?

  19. user-757117 | | #19

    How are these containers supposed to be heated?
    Electric resistance?

    If ventilation is through a cracked window, it's that cold outside, and the surrounding region is dry to begin with, I wonder how humid the interior could possibly get?
    I'm guessing cracked knuckles and creaky furniture are not uncommon this time of year.

    How many people will sleep inside?
    Will people be cooking or showering inside?

  20. user-659915 | | #20

    I think Michael's idea would work in Ulaanbaatar's dry climate. But I rather think the XPS interior lining would be just fine too. There's potential for condensation for sure, but so long as it could drain and/or evaporate at winter's end I don't know what the harm would be. Moisture alone is not sufficient to grow mold, organic material material is also required which neither the steel nor the insulation would provide.

  21. user-757117 | | #21

    In Ulaanbaatar tonight,
    If I were to take outside air and warm it to just 0C the RH drops to 10%.
    The RH would drop even more if I kept warming the air.

  22. user-942951 | | #22

    Yes its pretty dry here in winter, so managing interior moisture via ventilation should help. For heating electric resistance baseboard will probably be the first option. Radiant floor has been suggested, but I think better to spend that on good doors, windows, air sealing etc.

  23. albertrooks | | #23


    The side of a shipping container looks much like the BSC section: Alternating cavities like a sine wave. If you had a guy with a cutting torch who cut a quick slot across the top and bottom of every outward wave then they could function as a vent. Much like a ventilated rain screen. If/when they froze closed it probably wouldn't matter. All of the water would be frozen too.

    When things thaw, the cavities can vent and dry.

    That brings it back to XPS or Mineral Wool with a VB at the inside face. The VB can be polly, or really a good quality membrane might bet better if possible since polly doesn't last too long.

    You could screw through the slot from the outside to hang the framing lumber that carries the wall.

    Here is Lucas's picture again. The red line is the "venting slots" on the outward waves

    The lawyers could be pretty simple: From out to in: Framing lumber + Tyvek or? + Insulation + VB + Batten + finish layer. If the framing lumber was just 1/4" proud of the inner steel wave, then you could vent the whole wall and perhaps bring the vent around to the roof. Lay tyvek across the farming lumber faces and then screw the battens back through the insulation into the farming lumber. Then VB + Finish layer.

    Non moldy, fire safe and durable.

    That and a decent HRV will take you a long way.

  24. albertrooks | | #24


    "I was talking with a Tyvek representative at IBS last week who said they are constantly wrestling with the thought of putting out a camo or clapboard pattern product..."

    The sad thing is it would probably sell well... Especially the clapboard print. Of course if they don't hang it level, it's going to be really annoying... I'll feel like visiting a vortex.

  25. user-659915 | | #25

    Simple lawyers and farming lumber. The new face of spelling thanks to auto-correct?

  26. homedesign | | #26

    This Tyvek Exterior "Wallpaper" concept could catch on....
    not only clapboard and brick ...but modern fashion too!
    Frank Gehry Fish Scales
    Mondrian Grids ...
    rainscreen siding with fake shadows

  27. user-659915 | | #27

    Out to in:

    Steel container structure. No vent slots - more likely to let moisture in than out.

    2" XPS temporarily secured with adhesive dabs tight to the inner wave of the steel section (no furring, no mechanical fixings), joints taped on the interior.

    2 x 3 metal stud interior box frame: wall, floor, ceiling, built tight to XPS with no through fixings, secured only to itself so no thermal bridge, air or vapor gap through the XPS.

    Mineral wool framing fill. Interior finish with vapor retarder, no poly. No concealed wood framing or any other material that will support mold growth.

    Shower units obviously need special attention to control both water vapor and liquid water at the shower trays. They will still smell moldy after a couple of years in a mining camp but at least you should be able to make sure the mold is on a visible surface where you can scrub it with the local equivalent of Clorox rather than buried in the structure.

  28. user-659915 | | #28

    @ John Brooks:
    You mean like this?

  29. user-757117 | | #29

    UB is now officially the worst polluted capital in the world, and in the top ten worse polluted cities in the world.

    Not sure if you're still checking in...

    Just curious...
    Are they mining for rare earth elements?
    Are the mines near UB?
    If so, how much of the pollution would you say is related to the activities of the mining industry?

  30. user-942951 | | #30

    Lucas, just checked in.

    The pollution is coal smoke smog (plus some traffic fumes). Biggest source is the informal settlements surrounding the city where everyone burns dirty coal (and even trash), to cook and stay warm.

    Also heat only boilers for schools and public buildings, and coal fired power stations add to the mix.

    Mining is distant from the city, but is driving the boom and unsustainable urban growth. There's coal gold, copper, oil, uranium, rare earths... everything.


  31. user-757117 | | #31


    I was curious because there has been a lot of exploration for, and discoveries of rare earths and other minerals in northern Ontario.
    Getting them out of the ground will take billions (probably public money) in infrastructure spending - they're the not-so-low hanging fruit of mineral deposits.
    I'm not excited about the prospect of more coal-fired power generation...

    But hey, the world needs it's ever increasing annual alottment of new ipads and other electronic do-dads I guess.

  32. user-942951 | | #32

    Albert, thanks for posting my question here, I've learned a great deal from GBA.

    Exterior insulation would of course be best, but we rapidly start to get to the point where we might as well build prefab another way (next project), and whatever way we do it they are going to be less portable. For this project we want to keep the steel exterior for ruggedness (durability) and security: the idea is "plug and play, lock and go".

    Containers are already quite popular for a variety of uses, and those insulated on the interior are normally done with mineral wool or fiberglass, I've just seen one done with open cell spray foam (i.e. a sponge lining). The one container shower block I've seen did smell very moldy. So I'm trying to figure out if there's a better way to do it. Not only is the challenge moisture, but also how to insulate effectively without loosing too much space (i.e. 4 to 5 inches).

    Potential uses are site office (mine or construction), temporary worker accommodation, ablutions (toilet and shower), weekend cabin (vandal proof, like Lucas' image posted above). One company I know just ordered 3 x 40ft containers for temporary offices while they build a new training center; six desks to each can; the Canadian expat specified a poly vapor barrier on the interior of the mineral wool. We've just been asked to quote on a 20ft container for use as a shower/toilet.

    Martin's comment about finding a "pretty good" solution is spot on; I'm wondering if we've been 'over-thinking' this. Maybe mineral wool with air barrier/vapor retarder might be good enough for a site office, especially if we've installed a small HRV and therefore manage indoor humidity.

    Closed-cell spray foam would be a good solution, especially high humidity uses like shower block, or bunk room and I am sure it won't be too long before someone offers closed cell.

    Should the focus be on vapor diffusion or air transported moisture?

    A polyethylene class I vapor retarder (vapor barrier) on the inside of mineral wool will need to be perfectly installed. But if any moisture gets in there through holes/tears/abuse etc its stuck in a vapor impermeable sandwich and not going to dry. But whats it going to do in there? It might rot the studs and corrode some fixings. It might cause mold but will the spores get back into the occupied space? But its not going to be a catastrophic failure of the structure. So is a poly vapor barrier interior to the the mineral wool 'good enough'?

    Alternatively is an air barrier "good enough" for the mineral wool option? Most interior lining materials will function as a vapor retarder to some extent, and if we have good air sealing this might effectively manage the moisture problem. Again this would need to maintain integrity (expansion/contraction as Martin noted, and also punishing 'roads' to get the can to site). I'm leaning to fiber cement board sheets for the internal lining for toughness, MgO board or plywood are alternatives, we could install these ADA style or tape?

    As Albert mentioned I have also thought about XPS. My concerns are fire safety in addition to blowing agent considerations. I was thinking two layers of 2 inch (5cm) XPS, sheets offset, glued together, and maybe taped for good measure, getting us to about R20 for the walls. That would be air tight and a 'pretty good' vapor retarder. This presents some fixing challenges, we can fix wood studs in the corrugations to use as a base for screwing to through the foam. BSC's swimming pool roof might provide some inspiration for this, see attached. Maybe this is a solution for a shower block? But for accommodation fire safety takes precedence and we might be best to avoid interior foam?

    Lots of container house how-to websites out there, but mostly lacking in construction details, and most seem to think ceramic paint is all you need ;-)

    The google earth image that John Brooks posted shows the yurt or 'ger' districts that surround Ulaanbaatar (UB as we call it). UB is now officially the worst polluted capital in the world, and in the top ten worse polluted cities in the world. The ger districts where everyone burns coal in a yurt or a rough and ready house are a major source of the problem. Yurts are not that energy efficient, they work out on the steppe, but best not to build a city out of them. I lived in one for a winter when the temperature hit -50C.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |