Wall Component & Assembly Suggestions for Repurposing Shipping Container in Northern Climate
Hi folks,
The project is envelope design & construction for repurposing a 20′ shipping container in Toronto, Ontario – USDA Zone 5. Primary goal is high energy efficiency – within a medium budget range – while meeting interior conditioning requirements & comfort demands of users for different applications, including but not limited to:
– cricket farm
– vaccination and/or testing site (“clinic”)
– temporary isolation living quarters
– office space
The design will be a generic model as a base case for Ontario’s humid continental climate, modified according to different application scenarios, such as those mentioned above.
In general, I’d love your input on wall, roof, and flooring assemblies (it will be a radiant floor system).
Is spray foam a must? Is it more advantageous to insulate the interior, the exterior, or both? Is there any value in a thermal ceramic barrier coating applied directly to the steel panels (or elsewhere?)
All experiences & suggestions welcome. I’m new here and a student, so please go easy on me.
Lastly, the shipping container is not my choice, but rather what I have been given to work with.
Thanks for your time and wisdom! I enjoy learning from this community.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Erin,
Is your client absolutely wedded to recycling shipping containers? I've read that the cost savings are pretty minimal at best. It's also my understanding that you have to be careful about sourcing containers since they are sometimes used to ship really nasty stuff.
Performance-wise it probably makes more sense to insulate the outside. You could use reclaimed rigid insulation or new rockwool. On HVAC, split mini heat pumps would likely cost less while giving you AC when needed.
Thank you, Steve. I appreciate your response. A shipping container wouldn't be my choice personally, but my final year college project is to make it as effective as possible at heat retention, within a practical price range. It's a cold climate up here. I've been reassured that the inside is not contaminated.
Erin,
I'm not an expert. (Maybe they will chime in.) But the PERSIST method would be appropriate. You could use reclaimed polyiso for cost savings and higher r-values. Martin has a PERSIST article here: https://www.greenbuildingadvisor.com/article/getting-insulation-out-of-your-walls-and-ceilings.
I've read people get in trouble by designing in more openings than absolutely necessary.
Thank you for taking the time to make a few suggestions. Reclaimed polyiso hadn't at all occurred to me. I'll look into resources for it in my area now.
And yes, my plan is to minimize openings. I'm looking at one man door and a single window.
Thanks again for the input, Steve.
Erin, it's clear that the shipping container is not YOUR first choice, but could you just say a little more why it's been chosen as the platform you have to work with?
Just as there are numerous threads on this site exploring the pitfalls of how steel containers just aren't compatible with "green" building, there are many enthusiast resources all over the internet. The only thing I can figure is that these containers are incredibly tough and lend themselves to portability--IF you have the appropriate truck and load system. That it represents a "re-using" of some existing structure just doesn't seem to redeem the effort and expense in working around that fact that it's... well, a solid steel box designed for heavy freight.
But you didn't come here to get interrogated. You came for ideas on how to make it work. Like Steve above, many have given suggestions.
https://www.greenbuildingadvisor.com/question/what-kind-of-insulation-would-you-recommend-for-a-20-shipping-container-being-renovated-into-a-home-studio-space
https://www.greenbuildingadvisor.com/question/how-to-insulate-a-shipping-container-home
My guess is that it is being used as a platform because a) there was one available, b) it fits on campus, and c) the college likely has some additional purposing of it in mind for down the line, on top of its current use for my program's objectives.
I'm in agreement with you, Andy. At one time the idea appealed to a younger version of myself - that interest has had an inversely proportional relationship to my increased awareness of what's entailed in repurposing these annoyingly narrow and restrictive steel boxes.
I'm in my third and final year of a diploma program, as a mature student, coming from a very different career sector. I began researching sustainable building practices for a personal construction project, and came to realize that participating in that industry in some form of professional capacity seemed more interesting & meaningful to me than my employment at the time.
The pandemic's impacted the programming at my school and it would appear that the shipping container is somewhat of a provisional project.
Thank you for directing me to those two articles!
Erin,
Some semi thought out ideas:
The only reason to use containers are a) They give you a ready made exterior shell, and b) They look cool. So exterior insulation or additional cladding are out.
The simplest effective wall assembly I can think of is a 2"x4' stud wall kept 1 1/2" away from the inner profile of the container and the irregular space behind filled with closed-cell spray foam. The cavities are then available for services and can be fill with batt insulation.
The floor and ceiling are probably best dealt with using a layer of EPS covered by plywood. use G1S Birch on the ceiling, and a t&g subfloor as a base for finished flooring.
Thanks, Malcolm. This is close to what I had originally pictured, but feel like I'm getting a bit of pushback from my instructors, about the foam - I haven't deciphered what's behind their ambivalence as yet. In Ontario, spray foam needs to be installed by a professional so perhaps it's pandemic-related or that there is some apprehension about it being cost-prohibitive in some way that has not been made transparent to me.
Do you have any experience with sheep's wool?
Thank you for the floor and ceiling recos.
My suggestion for how to use the shipping container is to use it to store the tools and materials that you use to build a wood-framed building. Then you can sell the shipping container for re-use for its intended purpose.
Ha! I'd likely follow that suggestion in other circumstances, but the container is what I've been assigned as a final year project in my college program related to energy-efficiency. The pandemic has effected most of the promised programming that lured me back to school as an adult in the first place - namely hands-on internships and in-person integration into the industry... and so here I am, working on a shipping container.
I'd be remiss to not acknowledge that this challenge has gotten me thinking quite a bit.
Thanks for taking my provocative suggestion in the right spirit, and for clarifying the scenario you are working with. I'll read through some of the other comments and try to be more helpful if I have suggestions that haven't been covered.
Sure. I got a good laugh - it's one of those funny-because-it's-true situations.
Steel is an excellent vapour barrier. Since the container is meant to keep out the elements, it is pretty well sealed, so you are starting with a very solid air/vapor barrier. Normally this would be great, but it is on the wrong side.
If you want to keep the container aesthetic, than you need to insulate on the inside. Since the skin is a vapour barrier, you now have a cold side vapour barrier. About the only option at that point is lots of spray foam. You can do other materials but you'll get condensation in the walls and mold down the road.
If you don't mind changing the outside, you can insulate it on the outside with pretty much any material. Make sure the insulation is continuous, you don't want any part of the container structure poking through your insulation as steel is an excellent conductor of heat and it is very easy to create large thermal bridges. This is especially true of the of the floor which is hard to insulate without a way to lift it up.
P.S. The strength of the container comes form the continuous outer metal skin, as soon as you cut into it for doors and windows, you loose that unless you reinforce the opening. Not something that can be ignored as you might get a surprise if you try to move or lift it.
Hi Akos. Thanks for chiming in. 2" foam was my first thought - I presumed it was assumed, given our climate and the parameters of the assignment (energy efficiency, w. reduced/eliminated need for fossil fuel input, more so than an eliminated or minimal embodied carbon footprint). I was startled yesterday by some feedback from an advisor, that continues to seem counter-intuitive and puzzling. I'll share it here: "[Spray foam is] pretty bad moisture management as outside is metal (read: no smart vapor barrier). Recommendation would be combination of sheep's wool, and water tanks for thermal floor mass. Wool is much better than foam for this application and much easier to install than foam."
The person who relayed this to me mentioned issues of water being trapped between the steel and foam, causing rust/ degradation issues.
Any comments/ thoughts?
PS I had planned on raising the container with levelling supports. Something like in the attached a pic. Also, I was thinking the man door would be triple-glazed and installed in one of the 2 swinging barn doors at the end of the container.
Closed cell SPF of excellent for managing moisture. Anything over 2" thick is considered to be a vapor barrier plus it is also an air barrier. This means there is no chance for any interior air or moisture to make it to the steel and no chance of condensation. If the moisture can't get in, there is no way for it to be "trapped".
Fluffy is fluffy. There is no magical fluffy insulation (wool included) that will keep air that gets into your walls from condensing on the steel surface. In Toronto the outside of that container will be cold enough that any air from a typical house interior will condense a large part of the year. You can bin the seasonal temperatures to calculate this, a lot of time is spent bellow the 3degC dewpoint (equivalent to 70F room with 30%rh).
Since the steel has no way to dry towards the exterior, your only drying can happen in warmer weather to the interior, but in the mean time you'll have a lot of moisture sitting on the steel, perfect conditions for mold.
You can run some of your wall assembly ideas through this calculator:
https://www.ubakus.com/en/r-value-calculator/index.php?
It can show if you have condensation risk anywhere in your assembly.
2" of rigid is not a lot, but because it is a small space it won't matter all that much. It is still well bellow OBC code min (~4" of polyiso). Rigid foam is not the same as spray foam though as it is can't be made air tight. If you do go with rigid, it needs to go on the outside of the container.
If you want to reduce the amount of foam you use, one option is to go with a hybrid insulation option. This means 2" of cc SPF on the inside of the steel followed by 3 1/2 of batt insulation (this bat insulation can certainly be wool to make your advisor happy). You can have a similar assembly with rigid insulation (~2") on the outside with fluffy in the inside. In either case the rigid/SPF will raise the temperature enough to avoid condensation inside the walls.
Raising the container up will work, keep in mid that the OBC requires a lot of insulation for exposed floors (R31) so make sure to plan on how to attach that. Probably best on the outside. This can be fluffy insulation.
Akos, thank you!!! This is exactly what I want to demonstrate to my group members and advisor, but wasn't sure how to go about it - I have tables and charts and mins & maxes of weather data but don't know how to present it so that it clearly illustrates the condensation issues especially particular to this climate.
I've been aggravated for weeks because I think I'm being directed towards bad ideas and wrong decisions. I'm in a team for the project, but interacting with GBA has convinced me to go rogue - even though it will be a lot of extra work independently, I will save time arguing with the others and be able to follow my own visions and learn from the experience instead of being stonewalled. This thread of responses from everyone here has been validating and I'm so glad to have reached out.
I don't know how to bin seasonal temperatures. I'll see if I can figure out how to do this now.
Edited to Add: I now have histograms to clearly demonstrate critical factors of consideration and support my choices for the design. That is immensely helpful. Thank you for breaking it down for me in that manner, Akos.
I'll share with you a chart that I've done that I'm rather proud of comparing various insulation materials:
Cost/board foot R-Value/inch Cost/Rvalue
Wool batt $0.41 3.6 $0.11
Polyiso $0.58 6.5 $0.09
Rockwool batt $0.23 4.1 $0.06
Fiberglass batt $0.11 3.5 $0.03
Rockwool board $0.81 4 $0.20
Cork $2.10 4 $0.53
XPS $0.47 5 $0.09
EPS $0.39 3.9 $0.10
HFC Closed cell Spray Foam $1.00 6 $0.17
HFO Closed Cell Spray Foam $1.50 6.3 $0.24
Open cell spray foam $0.50 3.6 $0.14
Dense pack cellulose $0.10 3.8 $0.03
Blown wool $0.50 4.3 $0.12
These are for where I am (Washington, DC) and you'll have to research your own area because prices vary a lot by geography.
Given the parameters you've stated, that final column is important, price per unit of r-value. But rather than us doing your homework for you, I'd rather guide you on the path. if I were grading your project, here are the things I would want to see discussed:
1. What are the insulation requirements for a habitable structure in your climate zone?
2. Each of the options listed above requires some sort of supporting structure, what is the cost of that structure?
3. There's more to a wall than insulation. Joe Lstiburek, PhD, (also Canadian) writes about the "perfect wall" having four control layers. How are those four layers manifested in your construction?
4. Inhabited structures require ventilation. How is that provided? What steps are taken to manage humidity and maximize efficiency?
5. In your climate an inhabited structure requires heating in the winter and cooling in the summer. How is that provided?
If you make a good faith effort to research each of those questions I will help you arrive at the answers.
DC,
Good advice!
Thank you DCContrarian,
I'm pleased to report that I've been asking myself similar questions along the way, and so have a bit of a head start on the 5 you've proposed. Your response is both helpful and reassuring, in that perhaps I have been meandering along the right track. I'll piece my notes together in a more organized answer format, and look forward to coming back here with it shortly.
I'm familiar with the Perfect Wall, yay! I've referred to that Lstiburek article several times over the past week. :)
If you have any comments about the details in my #17 Reply to Akos, above, I'd be happy to know of them too (re: about the sheep's wool & water tank thermal mass floor direction I received from one of the project's partners).
Honestly -- and I'm a bit of a Contrarian around here on this -- when I hear someone say "thermal mass" I run for the door screaming.
If you have an active heating and cooling system -- ie, a thermostat controls the temperature -- the less heat capacity your structure has, the more comfortable it will be. It will achieve the thermostat set point more quickly and be less prone to overshoot.
Unless it's a course requirement to have radiant heat, I would recommend a mini-split HVAC system. This is actually a pretty good application for it. The only issue I would see is finding one small enough if you go with thick insulation. This box is going to require both heating and cooling and radiant is going to have a lot of issues with cooling.
As others have noted, you're going to have significant issues with moisture and humidity. In winter it will be cold enough out that the humidity that occupants produce will create a substantial risk of condensation on cool surfaces. In summer, the outside air will be hot and humid enough that you will want to cool the interior, and where outside air leaks in there will be a risk of condensation.
The standard advice is to put a vapor barrier on the side that is warmer most of the year. In heating-dominated climates that is the interior, and in cooling-dominated climates that is the exterior. If the other side of the wall can dry the heat will drive moisture through the wall and out. Toronto is very much a heating-dominated climate, and my understanding is that the norm in Ontario is to put plastic sheeting directly under the drywall. If you don't allow the wall to dry to the cooler side you'll get moisture problems. I think it's in the perfect wall article that Lstiburek talks about this, and gives the example of houses in Florida where people put vinyl wallpaper on the inside, or even a large mirror, and then get a mold problem.
The metal skin of the shipping container is going to be a vapor barrier. You could have it be the inside and insulate on the outside. This would give you vapor drive in the proper direction. This is also complicated and expensive, because you'd have to build an exterior cladding, which takes away part of the appeal of the shipping container.
If you put the insulation on the inside, the vapor barrier is on the outside, and your wall won't dry*. In that case, you have to build with the assumption that it can't dry and won't dry. This is the same situation as below-ground basement walls. The solution is to do everything possible to keep moisture out of the wall, which means air-sealing it relentlessly, and building it out of materials that are impervious to moisture. Usually this means closed cell foam, in either spray or board form. You might be able to use cork, but I'm not sure how impermeable it is.
I know that sheep's wool is advertised as being moisture-resistance but I don't think it would be appropriate for this environment.
*(This is a bit of an exaggeration. It will dry during the summer, when heat flow from outside to inside will drive moisture into the interior, where it can be removed by the air conditioning. But the air conditioning season is short.)
I'm not married to the curbside appeal of showing off the identifiable steel paneling, particularly at the expense of basic practicality. I think that I'll design and energy model 3 versions of the envelope - an exterior, an interior and a hybrid version of insulation systems, along with their various price points.
I had been eyeballing this little HRV:
https://lunoscanada.com/products/copy-of-lunos-e2-hrv-system-9
I'll have to do more research and side-by-side comparisons with a split mini heat pump, because I'm not informed or experienced enough to immediately recognize which is better suited for this project. I'm glad that both you and Steve Knapp CZ 3A Georgia, above, thought to highlight this for me.
And I'll certainly investigate alternatives to the radiant and thermal mass flooring. Now that you've raised the point, I realize that I don't know if it's an actual stipulation, if it was imposed, or if it was a group determination. This may save any screaming and running on your part, while adding another learning dimension to mine.
I'm starting to lean towards removing the 2 swinging doors at the one end entirely and building a wall that will house a door and a window - instead of cutting, structurally reinforcing and inserting a man door into a steel door or panel. I also like the idea of a simple and low pitched roof with an overhang, for shading, precipitation, and possibly to house a PV array.
>I'm starting to lean towards removing the 2 swinging doors at the one end entirely and building a wall that will house a door and a window - instead of cutting, structurally reinforcing and inserting a man door into a steel door or panel.
This sounds extremely sensible. Saves a lot of trouble.
Check local codes but I think you'll need 30 CFM of ventilation, which is about 50 M3/hr in your Canadian units. It will be an interesting question whether a heat recovery ventilator will be justified.
To do an A+ job, I would include a back-of-the-envelope Manual J, because that allows you to weigh the impact of different insulation schemes. I get around 5000 BTU/hr with R20 throughout (assuming a 40x8x8 container). I'm not aware of a minisplit smaller than 6K BTU but that would be a pretty good fit. That's over 15 BTU per square foot of floor area, which is starting to get into the upper limit of practicality for radiant floor heat.
>This may save any screaming and running on your part.
I appreciate that.
At 30 CFM, interior temperature of 72F, exterior 5F (I used the design temperature for the closest US city, Buffalo) I get over 2000 BTU/hr heat loss for ventilation. So up my estimate from 5K to 7K unless you use a HRV. That's 40% going to ventilation, so I would say it's significant.
Best for thermal performance would be exterior insulation. Rigid foam is easiest, but you end up with a lot of "wasted" space due to the corrugated nature of the container. Spray foam gets into those gaps, but you're never going to get a smooth surface with spray foam. Exterior insulation is best for performance, but as others have mentioned, it will change the exterior aesthetic from a shipping container to a sort of "streamlined" shipping container since you'll end up with a smooth exterior box shape after insulating.
If you insulate on the interior, you have the same two options. The problem is that you don't have much space to work with, especially vertically -- if you add insulation and a subfloor/ceiling, you've probably lost at least 5-6 inches of vertical space. You'll lose that on the walls too, and you'll only be getting about R13 if you go with polyiso here. I'd absolutely go with high R per inch materials here since you have so little space to work with to begin with.
I'll suggest a potentially interesting way to go if you can figure out a good way to do it. Support the container on wood timbers. You typically only need to support the corners, so you can use what would basically be wood "piers" for this purpose. That helps with thermal bridging. If you use treated lumber for this purpose, and you probably should, remember that you need to protect the steel from corrosion (this is why you have to use multicoated or stainless steel fasteners in treated lumber). I would use a thin sheet of stainless steel as an interface material hear, formed to make a cap for the wood. I'd probably use 0.050" stainless sheet here, in 304 alloy, which is thin enough to not be overboard but still thick enough to not get damaged.
Spray foam the exterior, with closed cell spray foam, but instead of trying for a smooth surface SCULPT the exterior. Make it look like igloo blocks, stone, whatever you want. Make the "impossible to make smooth" spray foam surface into an architectural feature. Make it art! When you're done, find an encapsulate coating that can be sprayed on. You want something that will dry an form a hard protective coating for the spray foam. It needs to be UV resistant, and it needs to shed water. I know there are coatings like this out there, but I don't work with them so I can't recommend a specific one. I'd try contacting someone at PPG and asking them. If you're tied in with a university, or some type of charitable project, you might even be able to get the material for free.
Bill
Around here spray foam on interior walls needs to be painted with fire retardant ("intumescent") paint or covered with drywall or 3/4" plywood.
It's an interesting engineering question whether it's more energy-efficient to have the containers on piers where you can put a thick layer of foam on the bottom, or on the ground where it's harder to insulate the floor but you're dealing with ground temperature instead of air temperature.
It's basically the question we talk about here a lot, whether a conditioned crawlspace is better than a vented crawlspace.
Hi Zephyr7/ Bill - thanks for the input. This is fun to think about. I came across a photo of one with a spray-foamed exterior showing run-of-the-mill lumps and lines and it occurred to me that the imperfections could have been embraced, embellished, and painted into three-dimensional representations of ivy or something.
Hi again, Erin. Not sure how this link has been overlooked, but here's the definitve treatment on the topic: https://www.buildingscience.com/documents/insights/bsi078-ship-shape
It begins with the fascinating history of shipping containers, but quickly turns to detailed drawings and ideas that should be considered authoritative for your project.
Author Joe Lstiburek, foremost expert on building enclosures and fellow Ontarian (I'm a UW grad), somehow manages to work in several references to the village of Tuktoyaktuk, so you know it's legit advice for cold climates.
Andy - thank you for this link, very interesting! I'm familiar with Lstiburek - how odd to have missed this particularly relevant article, when I've been researching the very topic for quite some time.
Back in the day, I attended UofGuelph and am also a Concordia alum. These days I'm the oldest kid in the class (!) at Humber College in Toronto, hoping to transition into a very different field than my last. While studying here @ Humber I have access to Revit and RETScreen, so it will be fun to model and play around with some of the suggestions from this GBA post and in the linked articles. I really love learning about all of this and feel lucky to be doing so at this stage in the game.
Thank you again for your support in my current quest. :D
Looking at the assembly drawings in this article - although the steel wall and roof panels are labelled as corrugated, they are drawn flat/flush to the rigid insulation (wall) and OSB plywood (roof). There's no mention of the alternating 25mm deep concave/convex property of the steel - would those air pockets not have any implications or are they irrelevant once sealed with 2+ layers of rigid ins?
There's a company called InSoFast that sells all sorts of EPS insulating systems particular to shipping containers. They have "inserts" that can be applied either or both internally & externally to side, end, and roof panels. I haven't sourced or priced them in Canada.
I've used InsoFast*. It's a good product that does what it says it does, but it's crazy expensive -- about $3 a square foot for the regular panels and $4 for the ones cut to shipping containers. It also doesn't meet code in most places in most situations.
I believe exterior walls in Toronto require R5 of continuous insulation plus R13 of framed wall with cavity insulation. That's 1-1.25" of foam insulation and a 2x4 wall filled with fluffy stuff. Given that, I don't see any advantage to the Insofast inserts. The Insofast installation method is to glue it to the wall with construction adhesive. You could just do that with regular foam board, the recesses behind the foam wouldn't affect anything. I would take great pain to make the foam airtight, taping the seams where pieces meet and using spray foam or sealant at the top, bottom and outside edges. Then I would build a 2x4 wood wall. The top and bottom would be just glued in place, if it's cut so it's a tight fit it's not going anywhere. Then fill the cavities with fluffy insulation -- you can use that sheep's wool and make your instructor happy -- and drywall over that. There's a lot of benefit to a 2x4 wall when it's time for things like wiring.
(*The spot where I used InsoFast was a basement bathroom, where I needed to minimize wall thickness while creating an insulated, finished wall. It worked well for that, and since it was only about 50 square feet the cost wasn't too bad. But I wouldn't recommend it for a large area.)
Ah, ok - that's good to know. There's a company in Brantford Ontario that recycles styrofoam by chopping it up into molecules and I had some hair-brained visionary scheme about jerry-rigging something very similar to InSoFast's inserts - and so was both amazed & slightly disappointed to discover InSoFast shortly afterwards.
I thought it would be clever to make use of those little pockets by stuffing them with something, but wasn't sure how to determine the added R-value to effort ratio. If negligible, just one more idea to scrap.
Still air has an R-value of about 4. However, when an airspace gets above a certain size the air in it starts to circulate because of the difference in temperature between the warm side and the cool side driving convection, and the R-value drops. How much is the kind of things nerds argue about.
If you want to show the other kids in the class how smart you are, here's how to calculate the R-value of an assembly of different materials. You use the U-value, which is the inverse of the R-value (ie, U=1/R). You multiply the area times the U-value and add them all up, then convert back to R-value.
So let's say the wall is 8x40, 30% of the area is indents and they are 1" deep. You're using EPS which is R3.9 per inch and 1-1/2" thick, or R5.85. How much insulation do you get by filling those indents?
70% of the wall = 320*.70=224 sf, R5.85= U0.17, 224*0.17= 38
30% of the wall = 320*.3= 96 sf R9.75= U 0.10, 96*0.10=9.6
Total U value of the wall =38+9.6= 47.6, or 0..149/sf. R-value is inverse of u-value, which is R6.7. If you just did the whole wall in inch and a half EPS it would be R5.85, assuming those air pockets give nothing. So filling those pockets only contributes R0.85 to the assembly, which isn't much. It might pay off if it would allow you to go down a size on the insulation and still meet the R5 that you need, but it doesn't.
I'm not quick at it yet, but they covered this in my program! :D
I was stymied by this particular layer though, wondering if the steel zigzagging through the EPS panels need to be accounted for in some manner. I see from what you illustrated above that I was over-thinking it - not unusual.
Thank you.
In my calculations I assumed that the outside indentations wouldn't be filled, only the inside.
For the zig-zag steel you'd have to calculate its length (diagonally through the wall) and cross-sectional area, and use the r-value of steel to figure out what it contributes to the U-value of the assembly. I'd predict that steel is so conductive that it negates any contribution of the insulation on the outside.
Because of the steel zig-zag with pretty thick steel, the R value of that assembly is essentially zero. Essentially what you have a giant thermal bridge through the foam.
When it comes to steel structures, the only R value that matters is continuous insulation. You have to take great care to make the insulation continuous, any exposed metal surfaces through the insulation can kill the energy efficiency of the whole assembly.
You have equal face areas of steel on either side of that insulation, and fairly thick steel connecting them over a relatively short path, maybe 35-40mm or so. The result is that the two exposed faces act as radiant surfaces, connected by a metallic thermal link. You've just short circuited all your insulation as a result. You'll need to either put all your insulation on one side, ideally following the contour of the steel in this case, or put equal amounts on BOTH sides so that the steel is completely encapsulated within the insulation where it can't "bridge" to anything.
Bill
Gotcha. Thanks!
Filling the ones one one side, in addition to more foam over them, still has some benefit, but filling them on both sides does not.
Erin,
Your project might be over by now, but I thought you would be interested in this announcement for a shipping container neighborhood that is being planned for a city near Atlanta: https://urbanize.city/atlanta/post/shipping-container-homes-for-sale-south-fulton-evo-haven.