What's new
What's new

Advice on building an exterior door for house in extreme cold environment

crzypete

Hot Rolled
Joined
Jun 8, 2003
Location
New York State
I have a potential job building a pair or doors that will live in interior Alaska. Regular temps of up to -40ºF on one side of the door and standard house temps on the other. The Client has already specced triple pane glass. I was curious if anyone else had experience in building doors for such an extreme environment? Anyone ever sneak insulation in in any capacity?

I'm planning on stave core construction. Wood TBD.

Thanks

Peter
 
I have a potential job building a pair or doors that will live in interior Alaska. Regular temps of up to -40ºF on one side of the door and standard house temps on the other. The Client has already specced triple pane glass. I was curious if anyone else had experience in building doors for such an extreme environment? Anyone ever sneak insulation in in any capacity?

I'm planning on stave core construction. Wood TBD.

Thanks

Peter

No need to go as far as an Alaskan winter to become aware that LOTS of exterior doors have insulating cores.

Wood? "Only"?

Welll... earliest residenters hung animal skins. Fur-on. More than one layer. And over the tiniest opening they could get in and out through.
 
Solid wood is a terrible insulator. You need lots of insulation and good weatherstripping. The really high R-value doors don't have a single closing plane. They have a rabbet around the outside so you get good sealing. They also tend to have multi-point latches to seal up well. You'll also have to consider thermal bridging.

Houses are getting better up there, but really high performance doors are still rare. Last time I was out in the Bush, everything was your bog standard fiberglass or steel door. The easiest way to improve the door is to add another one (arctic entry).
 
...The easiest way to improve the door is to add another one (arctic entry)...

This......or...

I've often wondered why, with all the effort being put into improving the insulation in houses, that standard doors are still normally
no more than two inches thick. No reason that I can see that a door couldn't be 3-1/2"-4" thick. This would allow for a lot more
insulation. Perhaps build it like an Arctic wall where the studs are staggered from inside to outside so there can be no direct
conductive heat loss through the wood...
 
No reason that I can see that a door couldn't be 3-1/2"-4" thick.
You start running into issues where the width of the door across the (plan, not elevation) diagonal with hinges at one end is significantly more than the width of the door flat on. The vertical corner diagonally across from the hinges starts running into the jamb. So you need more clearance in the jamb (bad for sealing) or hinges somewhat fancier than a simple pivot (expensive, nonstandard) or a door/jamb with a taper on the edge opposite the hinge side (expensive, nonstandard).

Look at a thick bank vault or safe door. They are almost always tapered on the edge opposite the hinges.
 
Hey Pete, will you get to go there and install them? Hopefully in the summer....

Anyway, I'd think balsa would be the best insulating wood, lots of air in there. Or maybe just a sandwich of wood, insulating foam, wood. Will the door see direct sunlight in summer?

Gotta believe that the Alaskans (and Siberians, Tibetans, Tiera del Feugans, etc) have this figured out by now. Field trip!!!
 
You start running into issues where the width of the door across the (plan, not elevation) diagonal with hinges at one end is significantly more than the width of the door flat on. The vertical corner diagonally across from the hinges starts running into the jamb. So you need more clearance in the jamb (bad for sealing) or hinges somewhat fancier than a simple pivot (expensive, nonstandard) or a door/jamb with a taper on the edge opposite the hinge side (expensive, nonstandard).

Look at a thick bank vault or safe door. They are almost always tapered on the edge opposite the hinges.

That's another reason the thick, high R-value doors have rabetted edges.
 
DSCN0466.jpgDSCN0481.jpgDSCN0473.jpgThe standard overhead crappy door will always be the weak link in a shops insulation.
So when a friend and I built an addition onto my house last summer we insulated the swinging carriage doors with 3" thick R30 foil backed rigid foam insulation.
 
View attachment 217363View attachment 217364View attachment 217365The standard overhead crappy door will always be the weak link in a shops insulation.
So when a friend and I built an addition onto my house last summer we insulated the swinging carriage doors with 3" thick R30 foil backed rigid foam insulation.

The MAIN thing you gained was better ability to control the EDGE sealing than paneled roll-ups allow and can still move.

I just replaced the seals on my 18-foot wide one last Sunday. Major improvement, but still..

Next-up is to frame the rebated opening outside of the door for removable stack-in insulated panels to turn it into an all-round edge-sealed air-gap. Too much damned rails, rollers, and moving "hardware" to seal those doors from the inside.

What with other doors to the area, and not a lot of large stuff in or out, I have the advantage of being able to go weeks at a time without need of opening the roll-up at all, so.. not a universal solution, but...

"Carriage" doors, preferably bi-fold, with removable vertical posts, could be added to the outside to supplement the roll-up instead if I needed that.

What with BOTH door sets and the air-gap, they could be much lighter than your ones, and also allow for half - or less- being opened at a go.

Other than wintertime, the pushbutton-remote powered roll-up is more convenient "naked".

2CW ... well... probably closer to $200 - $500, but still...
 
Thermal breaks will help a lot. If they're large doors, I'd be tempted to start with a Structural Insulated Panel, and let in the necessary framework around the edges, creating a box-beam like structure. Tapered or stepped jambs. Magnetic weatherstripping, like your fridge. (Used to be available via Brookstone, of all places.) Drop-down or air-inflated threshold seal. Except for the SIP, I've used these techniques in recording studio door construction, where elimination of air gaps is key to acoustic isolation. Many parallels to thermal isolation.

If it's not just a hinged personnel door, then use semi-truck rear-door hardware to create a compressing seal when closed. Should work for sliding barn doors too if you mount the hardware horizontally.

For the finish surface, laminate on whatever the decor calls for. If the window size is not huge, no sense not doubling up on the triple-pane glazing. Make the inside one removable for the inevitable cleaning.

Chip
 
Thanks for the replies. These are two standard house entry doors. They will have some glass, possibly quite a bit of glass on one of them.

Unfortunately, the ship has sailed on an air lock, this is the main line of defense against the weather.

And Richard, Of course I need to go on site to do the install!!!

Anyone have any core material suggestions? The last door I did, I used engineered lumber, which was amazingly straight and didn't require any prep work, but I feel like moisture may be getting in and swelling it a bit, will probably only use those for interior doors in the future. Thinking solid wood for these.

Peter
 
Thanks for the replies. These are two standard house entry doors. They will have some glass, possibly quite a bit of glass on one of them.

Unfortunately, the ship has sailed on an air lock, this is the main line of defense against the weather.

And Richard, Of course I need to go on site to do the install!!!

Anyone have any core material suggestions? The last door I did, I used engineered lumber, which was amazingly straight and didn't require any prep work, but I feel like moisture may be getting in and swelling it a bit, will probably only use those for interior doors in the future. Thinking solid wood for these.

Peter

The three main insulation options are EPS, XPS, and polyiso.

EPS (expanded polystyrene) is the lowest R-value of the three. The only thing it has going for it is it's cheap and it has high vapor permeability.

XPS (extruded polystyrent) is the standard pink/blue foamboard insulation. It can form a good air barrier and is vapor semi-permeable (depending on thickness). R5 per inch.

Polyiso (polyisocyanurate) is whitish and usually comes with foil facing on both sides. It has a nominal R value of 6.5 per inch, but its real world R-value goes down a lot in cold weather. It's not a great choice for the outermost insulation layer in Fairbanks.

Insulation is also available as spray foam, which is easy, but more expensive than sheet foam.

I would probably use XPS as the core insulation. A solid wood door has a terrible R-value. Alternatively, build a case and fill it with spray foam.
 
Last edited:
This......or...

I've often wondered why, with all the effort being put into improving the insulation in houses, that standard doors are still normally
no more than two inches thick. No reason that I can see that a door couldn't be 3-1/2"-4" thick. This would allow for a lot more
insulation. Perhaps build it like an Arctic wall where the studs are staggered from inside to outside so there can be no direct
conductive heat loss through the wood...

No point of putting excessive efforts to 2 doors when you still have 20 windows that are about as bad or even worse.

Having said that: 76mm or 3" thick door is the standard here nowadays. Double sealing/weatherstripping , EPS insulation and if there is window its a 3-panel low-e/selective glass.

Ulko-ovi Pihla Termo 198 mittatilausovi - Ulko-ovet - Taloon.com

Metric U value of 0.8 to 1.0 soooo... that is R6 to R5 american "R"
 
I see the ship has sailed but thought I would add something for anyone in the future who reads this thread.

I have worked in Germany and was amazed the first time how everything was built so tight. The doors in the hotels were built like someone else said like a bank vault door. A stepped door and frame and they had rubber insulation on the ledges. I was so impressed I took some photo's..lol Also the windows are made so they open 2 ways. The lock handle in the middle inside of the window is you turn it 90 degrees and the window swings open from the side. Then if you turn the handle 180 deg's. the top tilts back and it hinges on the bottom. Then when you close it pulls itself in against a seal. I see they are made in USA too:
European Tilt and Turn Windows

DSC00039.jpgDSC00037.jpg
 
I did what Rusty Tool did, but my shop doors are 6" thick with insulation inside. Not as fancy as his - was in a hurry 25 years ago to get the old barn weathered in, and just screwed on skins of 1/2" flakeboard. Hmmm...25 years and still strong....

On house doors, i agree, it would not be that difficult to make, say 3" thick doors. there are lots of weather seal options. Most people still don't get that a very substantial jamb system & an excess of high quality, oversize, ball bearing hinges is the more critical part of the package if you want door operation and weatherseals to remain "faultless" for the life of the product (other than needing to replace the weather seal and grease the hinges from time to time). The jamb system including hardware, thresh hold, and weather seals usually costs the same amount as a "nice" but relatively simple door. Sometimes more. I still dovetail the heads of doorjambs for "significant" installations. I have not had to make "nice" doors that were insulated beyond being 2-1/4" thick multi-layer construction.

smt
 
I built my shop doors the same way as well, they have been holding up well now ten years in.

But these doors are entry doors to a nice house. Stephen, I remember reading your dovetail casing detail and used that on the front door I built for our house last winter. That door was the inspiration for this door in a way in that my client saw it and wanted me to make similar ones for their home.

Now unfortunately the Rough opening is already existing and the existing doors are oddly 34" so making an extra heavy jamb is not an option. Making a jamb out of 5/4 hardwood made hanging the door here so easy. I rabbeted that jamb which makes for a cleaner look.

I should dig around for some pictures.

Quick internet research says triple pane glass is R5, hardwood is R .71 and softwood is R 1.41.

Peter
 
First greenbuildingadvisor.com

Second heat loss is delta t Times square footage over r value

21 square foot door at that temp has a max heat loss of about 2500 btu
With r5 glass and a 3 inch thick door you are down below 600

Doubling the r value will net you another 300 btu at the cost of?

Air sealing is always most important
 
First greenbuildingadvisor.com

Second heat loss is delta t Times square footage over r value

21 square foot door at that temp has a max heat loss of about 2500 btu
With r5 glass and a 3 inch thick door you are down below 600

Doubling the r value will net you another 300 btu at the cost of?

Air sealing is always most important

Btu's for how long?
Bill D
 
The client will eventually reconsider not having an airlock/mud room.

As to wood for the thermal break, dimensional stability is my concern.

I'd investigate the availability of interlocking extrusions having foam between them.
 








 
Back
Top