Like Archinect on Facebook.
Sign up to our mailing list.
I have a new one-story office building, slab on grade with footings at about 42".
Am looking to run continuous 2" rigid insulation on exterior of the sheathing and then essentially continue it right on down to the footing. This of course leave 8" or so exposed above grade, which then needs a cover board, or applied protection coating.
I like the simplicity of the idea of getting continuous insulation with no thermal bridging on the exterior, but the cover board thing sounds a little cheap-ish.
Any experience with the cover boards or applied protection coating in terms of durability and appearance?
I've never done it but have considered a masonry board product for a similar application. It can be parged to finish. Probably need a water barrier over the insulation, too.
I have also seen 22 ga. prefin. alu. sheet metal.
ejm beat me to it!
The problem with that detail is metal flashing is subject to damage from weedwackers and lawn maintenance as well as weathering.
On my last project I isolated the wall and footings by insulating under the slab with a thermal break at the exterior.
Just found this.
Building science is a great resource. This is a really tricky location. Try to get insulation between two layers of foundation wall. Or interior insulation that comes right up to the slab edge.
Brick veneer can also cover the insulation at this location.
But there is no good product to cover exposed insulation. Just various stucco type coatings. All leave it exposed to abuse.
Great info and links -- thank you.
The reason I'm poking around on this method of insulation -- there is quite a bit of R-value testing showing that the continuous exterior foundation insulation outperforms the "interior' methods -- by quite a bit. Mainly, I think, because there is no thermal bridging, as there is with the other methods. Plus, except for the dilemma of the 8" exposure, it's pretty simple to build -- essentially one plane of insulation.
The flip side is having to figuring out how to finish the 8" exposed part. -- the links pretty much lay out what's been used typically, which helps a lot. I'm just not digging the look of the stucco coatings. Plus it looks like even the fiber-cement backer board type ideas end up needing a stucco coating.
Yeah, you hit the nail on the head. Post back your proposed details, I geek out on this stuff.
I will. But right now, I got nuthin'... What did you have in mind on the brick veneer? Like run it -- (and maybe some 4" CMU where it goes subgrade) -- all the way down to the footing? I'd guess you could through-wall flash above grade at about fin.flr., then instead of solid grouting down a foot or so below grade as is done quite a bit, you'd solid grout between the brick and insulation all the way down to the footing...?
I don't understand the reason for insulating the perimeter of a buried foundation wall. Trying to keep the ground warm?
Make a thermal break and insulate the slab, which I assume you were going to do anyway. Expose the concrete outside and deal with the plate detail. Or make it really complicated and expensive.
^expanding on this...
Miles -- the reason for discussing exterior-side foundation wall insulation is that is outperforms the interior methods such as the one you are suggesting -- not that what you're describing is bad. But a continuous insulation plane, essentially from the top of wall to the footing is a simple build with no thermal bridging -- your method still bridges. With continuous exterior insulation, a phased temperature gradient is created under the building so that the temperature of the earth directly below the slab nearly matches the interior temp.
(Also, if you have a monolithic slab / grade beam or whatever, exterior is your only performing choice -- insulating the slab does nothing in this case.)
So to solve the protection / cover board problem would be great since it's the simple, performing scheme. But faux stucco just ain't cool.
saint, I've worked with Jim D’Aloisio who has a lot of insight on these issues. Here is a brief article where he discusses this detail:
He also recommends using sill sealer (the thin layer of foam) in commercial construction, under the bottom track of a steel stud wall. It's a normal detail in residential, but not commercial.
He is a great resource.
I personally think that the main issues with exterior insulation are:
1) it cannot be exposed to UV radiation
2) the affordable ways of protecting it don't necessarily look good
I don't think that damage is really an issue, unless in a high traffic area.
In my last building where this was an issue we tried to do this:
1) run insulation from the footing up to just below grade
2) cavity wall insulation (with brick) down to just below grade.
at this point, the two layers of insulation did not align. So we detailed 3" of high compressive strength rigid insulation UNDER the bottom course of brick. I calculated that it would compress ~ 1/16" which I thought was acceptable. this layer of insulation tied together the other two layers.
Of course, in the field, the contractor did not build it this way, so I have no idea if it'll work or not. but I thought it was a cheap solution.
Two pours seperate the slab frm the wall. Now it's down to thermal breaks that are detailed subject to bearing issues. The slab can float inside the foundation or bear on a ledge on 700 kPa insulation. As gruen mentioned, the plate connection needs attention too.
I'd like to see a cost benefit analysis of a 15-20 degree difference in subsoil temperature below an insulated slab. The vast majority of heat loss is infiltration, glazing, etc. and more money spent on these items would logically seem to have a vastly greater impact on efficiency than what you propose (subject to climate - ground temp here is about 55).
If it's a monolithic pour you might consider reversing the whole arrangement and insulating the top of the slab. A different flooring system would be required but there are potential benefits to that for mech & elec.
Jim D’Aloisio has done some interesting calculations on heat loss through steel brick ledgers/headers, etc. It's an amazingly high number compared to the relatively small area of the steel angle.
It's because of the massively high conductivity through the steel as compared to other materials (wood, etc)
I think the slab edge detail is critical, because even though it's small, it's exposed to the air (not to the earth). And, the concrete does go inside of the building, so while the exposed edge is small, it acts as a huge heat sink to pull heat out of the building.
You have to insulate basements per energy code anyway.
I've been trying to come up with energy efficient details for attaching metal panels (or any type of rainscreen wall) through the exterior rigid insulation without using a continuous metal Z-strip. The Z- strip pretty much kills the efficiency of the insulation.
And, top of wall (parapet) details. Same thing. It's a big heat sink on top of the building, but difficult to wrap with insulation.
Thanks, Gruen -- good link. Funny -- the Figure 2 is pretty much what was attempting to describe in a previous post.
Are you familiar with Thermomass? Essentially non-conducting specialty ties to connect two wythes of concrete with insulation between -- can be precast, tilt-up, and poured in place. Very interesting to me because the insulation could then be continuous from footing to top of wall, with concrete on both sides -- so concrete can be your interior and exterior finish material, which I'd be diggin'. Not for everyone's tastes, granted, but the sheer simplicity and performance is very cool.
You are correct, too, in that the way you deal with the slab edge makes a very big difference.
I'm still poking around on all this. I'll post up a sketch for your input later on.
You're reading my mind on the metal panels attachment problem -- that is also on my agenda for this project.
Miles -- you're trying to sell me on interior methods since you're apparently not buying the performance argument / data for the exterior. Your persistence is admirable.
The IECC has also ignited some of these issues for me, and sounds like for Gruen as well. Prescriptively, the IECC calls for 7.5 c.i. over metal studs in the zone I'm in. So, by the time a guy goes through putting 1.5 or 2 inches of c.i. in place over the entire wall, you tend to notice that the continuous concept falls apart near the grade and slab edge conditions. Yes, the problem of heat loss via conductivity and bridging using interior methods can be lessened through the use of several techniques, but it can't be solved using interior methods.
No sales pitch, just tossing out ideas. Brainstorming is fun.
In general I try to live by Keep It Simple Stupid. That often means refining standard details - based on construction experience and performance - rather than trying to reinvent.
This is probably not a solution for you, but I did just discover this product:
I don't think it's particularly attractive. But what it is: you place the special insulation board on your substrate (wall) and then fit "stones" or "bricks" into the insulation. It's then mortared in place. I got confirmation from them that if it's flashed above grade, it can be run below grade. It would protect & insulate the foundation. I would also guess it could be used from below grade to the bottom side of a different siding material. You could use it to do a "brick" foundation.
^ Interesting. I mean you're right -- not that attractive -- but it's a feasible concept.
Points out a continuing problem -- sometimes aesthetics decrease as energy efficiency increases: A few inches of EIFS over CMU could create quite the effective Thermos -- the CMU mass would store energy, there's be no bridging, the "below flashing" material could seemlessly match what's above, and it's economical to build / operate -- but likely pretty unattractive.
Seems strange that this would be a problem -- it's a foundation wall and slab. We're not building a spaceship.
It's only a problem because you've made it one. Can't help it, I guess.
Exactly. I invented the problem. Just for fun.
Well, it's actually a code requirement, although difficult to comply with (at least, economically). And saves tons of energy..
I invent problems all the time. Some of them even have to do with architecture.
Miles, this issue is totally BS to you -- fine by me, I've been polite. What's with the all the unrelated fun facts?
No BS. Design is a process of problem solving. Often the problems must be created in order to define a system of constraints in which to work. I frequently agonize over the difference between design intent and practical execution. The two biggest issues are cost and performance. Aesthetics - which are of course completely subjective - tend to drive the other two.
Once again I'd like to see the lifetime energy cost quantified between insulated slab with thermal break and the same plus insulated foundation wall. Practical experience is often at odds with theoretical calculations. Assuming a thermal break, a small increase in sub-slab R value would certainly make up any difference, leaving you the desired? concrete exterior finish.
Gruen -- how are they enforcing the IECC where you're at -- is it strictly prescriptive or do they allow performance compliance proof reports like Comcheck or whatever? I've found some highly uninformed code officials when it comes to IECC issues -- often, they aren't sure what they want to see.
Com & rescheck are OK, and can allow for lower amounts of insulation. And hey...if you put BS numbers into comcheck, you can get away w anything. But I'm trying to what the code asks for and maybe more. Miles raises an interesting question. I will see if I can find some data.
OK, here's a good article, with visuals, about how an exposed concrete edge can reduce the thermal efficiency of a wall.
the part III of the article shows a spreadsheet.
If I run similar numbers. Let's assume a 8' high x 10' long CMU wall, with 6" of exposed 8" thick concrete at the bottom. Total height 8'-6", total area of 85 SF. The wall is insulated with exterior R-13 insulation per code (climate zone 5, my area), for a U-value of 0.077 plus R-2.56 for the conc block (U-0.39). The concrete is U-.67. The wall is 94.12% of the total, concrete is 5.88%. Running the numbers gives us an effective U-value of 0.1, which translates to an effective R-value of R-10. So, there was a loss of R-5 across the entire wall, just because of 6" of exposed concrete wall.
I'm not sure this is enough data to be convincing, but I find that the problem area is usually not just 6" high. Many buildings, set into slopes, can have exposed foundation walls 1 story high at the high point down to several feet at the low point.
Good stuff, Gruen. By the way, Comcheck seems popular for all involved since mediocre designs with problem areas pass anyway. Voila, your building is "energy efficient".
Miles, "Assuming a thermal break..." is my favorite phrase so far.
So if you abandon the exterior insulation and focus on the slab edge you restrict the problem to a small area. Either float the slab inside the wall or let it bear on a shelf. Either way you can provide a good thermal break - R12 in 2" (which is probably total overkill).
Now you've got thermal performance plus the desired concrete finish on the exterior foundation wall and are free go do some real design work.
Yeah, com & res check can be tweaked to get the results you want. Or, you add insulation where it's easy, and delete it where it's difficult.
Thermal breaks help, but if you've provided a path where the heat can short circuit the insulation, it will go through anyway, it does not care about straight lines.
To get back to the original topic - doing a well insulated base of wall detail is very difficult, especially in an affordable and attractive way.
Looking at large buildings around town here, I see lots of finished stone and brick foundations. I'm going to guess that many of these still don't have thermally improved details at the base of wall. If I come across any great details, I'll post them. It seems to me that really, one wants to build a cavity wall, with some sort of masonry as the exposed face, a layer of insulation sandwiched between it and the foundation. That extra layer of masonry, and a structure to support it, is the expensive part.
Not to mention, it kinda kills the idea of a shallow frost protected foundation, which is another great idea that saves energy and money (less concrete produced and used).
- protect the insulation with something relatively lightweight and attractive. Perhaps exposed fastener metal panels attached to furring strips, attached through the insulation into the foundation wall.
Another point. Say you have brick on the wall, and want to end it before grade. What do you support the brick on? If you choose a metal brick ledge (angle) you will also kill your thermal envelope - because the conductivity of steel is EXTREMELY high.
Look into using stainless steel (still high conductivity but only 1/3 that of steel) or FRP angles (really expensive, but the best thing...)
The point is: you need to not only provide a thermal break between steel and the building wall, you need to insulate behind it. Free floating brick lintels are a good idea. FRP blocks to space it off the wall same thickness of the insulation is also a good idea. Will PT wood blocks work? Probably better than nothing.
Unfortunately, Miles -- it's a not... read my posts, gruen's posts, his links, plus the many links above. You're proposing a slight modification to what's typically done. So, technically, what you're saying is an improvement.
But the problem at hand isn't one that I, gruen, the boys at Buidling Science, or the myriad other websites are simply imagining. It's an actual concern. If you post a sketch of your solution, we can show you the problem.
I'd put a masonry or concrete curb around the base of the building on the outside of the insulation if I was trying to use this detail. that would give it finished transition to grade and protect the base of the building from gardening equipment and the sort of thing. It would just be one course, so it wouldn't need a separate foundation. Just set it in gravel. Make it 12" to 18" high above finish grade and it should do the job. It wouldn't even need to be tied to the wall behind it, which would make it easier to take pieces out and get access to flashings etc. if something started to leak.
More importantly, you need to make sure there's a substantial drainage plane behind the insulation, particularly if it's going all the way to the foundation. Otherwise, it will wick water out of the earth up into your above-grade wall assemblies. By "substantial" I mean 1/2" or thicker. None of this 1/8" drainage mat stuff. At the footing, that drainage mat needs to be tied into through-wall drainage lines connected to the foundation drainage system.I'd also run the foundation waterproofing up the wall higher than normal if I was doing it this way.
Not exactly rocket science. Interior wall insulation dependent on climate zone.
To quote myself: "...a slight modification to what's typically done. So, technically, what you're saying is an improvement."
Compared against the OP, you still have a substantial thermal bridge. Some of gruen's links illustrate the loss when detailing as such.
gwharton -- interesting -- and you're adding to gruen's list of possibilities. I guess I'd wonder what keeps the "loose" masonry units in place and looking good over time. Also, would you run flashing out over the masonry units or use sloped tops? However, I started wondering about a bit heavier gauge flashing / drip edge that would run down over the masonry units 3" or so and could actually adequately hold them in place. Sort of a loose clip. Not sure about the bottom.
You could run a flashing over the top of them, but you don't really need to.. Just think about the curb as a mowing strip rather than a part of the building envelope. It's technically more part of the landscaping than the building. Make the units/blocks big enough, and their own weight will hold them in place. You could mortar sandset them at the base if you were really worried about them moving around. Or cast in place a concrete curb. Or permanently clip them in if you want. I was only pointing out that you don't actually have to attach them to the building.
But Miles' point about not wanting something that can get beat up by weed whackers and lawn mowers is a good one. You need something tough at that level. Most building cladding materials won't survive a run-in with power landscaping equipment without getting pretty scarred up at the very least, even if they didn't have rigid insulation behind them. Metal flashings and siding boards will get trashed in short order.
i assume the sill gasket in the easy to read picture above is really what makes that go from 'slight modification' to 'useful improvement' right? also, it would make the substantial thermal bridge less substantial due to the sealing characteristics of the goo and foam placed beneath the sill plate.
Like I said before:
Expose the concrete outside and deal with the plate detail.
Now it's down to thermal breaks that are detailed subject to bearing issues.
FOAMULAR rigid insulation (R5 per inch) is available form 40 to 100 psi. Use it under the plate to completely eliminate thermal bridging. Or use one of the products specifically engineered for this application such as Isokorb, Armatherm, etc.
It's not exactly rocket science.
That's the conventional way of doing it (in the USA...not elsewhere...continuous exterior insulation is the norm in Europe). But I can certainly see the advantages of doing it the way the OP wants to it. It's much simpler for one thing. And the thermal envelope will perform better with fewer places where the installation can get f***ed up by a bunch of different trades working around it at different stages. And the insulation can go on all at once as one material instead of in parts with different types (as you have to do when you use the conventional method). The main downside of doing it all exterior down to the footing is that you have to be more careful with exterior finishes and the drainage plane behind the insulation. It may also mean you have to change timing for when the foundation gets backfilled during construction.
He asked how we would go about doing it the way he wants to do it. And it's a reasonable thing to ask. Harping at him about using a conventional (in the USA) detail isn't constructive. And when you repeat yourself, it goes from off-topic to annoying.
I tried to use the 100psi insulation under brick, sort of what miles is suggesting. The masons built it the traditional way anyways. Miles detail is a heck of a lot better than most, but fails when the stem wall is a lot higher or lower than grade. A significant loss as shown too-through the plate. Most of my buildings don't do much better. The concrete guys also hate stepping the stem wall like that, and you'll find that the contractor ignores your detail (facepalm).
I guess I'd like details that are build able, cheap and energy efficient.
In the evolution of this thread we've gone through a variety of details. Last post explianed how to eliminate bridging at the plate with an isolated slab. Sorry if you find that annoying, it was pretty clear that Saint didn't get it.
Looking at the exterior insulation detail it seems that the wall will be whatever temperature the footing is and will conduct that into the building (ejm159's link). Unless of course you eliminate bridging with thermal breaks ...
As you mentioned, construction technique is critically important.
^ "...didn't get it" ? Uh, ya sure?
The premise of this thread originated in my 'getting it' -- in my already understanding the nearly standard detail / approach you're touting -- and looking to outperform its limitations. The thread is about managing the challenges associated with simpler and higher-performing exterior insulation approach.
You're convinced of the superiority of your approach -- that's no skin off my nose. Build away.
You set the parameters (no thermal bridging, no cheap cover board), you don't like the simple solution, and you don't have a more effective detail. Talk about persistence ... good luck.
Miles -- it's fine. Really. I take no offense that you don't agree with me on the exterior approach..
so the thermal break is different than insulation. a couple inches of foam will probably provide more insulation in that 3.5" space than the thermal break alone. however, we can all agree that the thermal break is, in fact, a thermal break. i'm pretty sure that means heat leaving or entering the conditioned space is doing so through radiation rather than conduction, which is a big improvement over not having a thermal break.
saint seems to think the thermal break is inadequate, and 2" of insulation needs to be continuously provided everywhere. if that's the case, i don't think you can have any windows. i work with commercial more than residential, so i'm not that familiar with the popular vinyl options, but an aluminum storefront will often have a thermal break in the frame. it's not insulated though. an inch air space in the glass is also probably going to perform less efficiently than 2" of foam, right? also, the gasket and sweep around doors is a thermal break that is not equivalent to 2" of foam insulation.
i'm pretty sure you also have heat loss where you have vent stacks and the like sticking out the roof. if you covered those with 2" of insulation, i don't think they would work properly.
tl;dr i don't think that 3.5" is that big of a deal (3.5" being a thermal break under a wood stud sill-details to other specific conditions will vary).
^ ...uh, wha?
we're talking about insulation. you started the thread asking about continuous insulation on the outside of the footing. you have shown in this thread that you don't like insulation on the interior of the footing. the detail miles posted with a pretty picture shows how one could insulate the inside of the footing while maintaining a continuous thermal break, which you don't seem to like. you referred to that detail as "typical" or "standard" and you claim exterior insulation is simpler and can outperform that detail.
the primary difference between your simple insulation detail and mile's standard insulation detail is that you want to provide continuous foam insulation, whereas mile's detail has a thermal break instead of insulation for 3.5" under the sill plate.
i was also comparing this detail to other openings in the building where thermal breaks, similar to miles' detail, are provided instead of insulation. it seemed to me, and this is just speculation, but you seem hung up on the idea that this 3.5" uninsulated gap under the sill plate is a very big problem for a building. or at least it's something that needs to be addressed to improve the building's performance. i was comparing other places in the building that would have similar uninsulated gaps to illustrate that such gaps are common and not really that big of a problem.
if there is something in this you're still not understanding i would be happy to try to explain.
All -- I definitely appreciate the input on this thread -- Gruen's and gwharton's posts, and Miles first couple posts, all brought up some potential good ideas and things to consider regarding detail management of exterior-side insulation.
As far as the "pro-interior side" argument that developed -- yes, it's done that way in some form on most buildings -- it's a decent detail, it meets code and it is familiar to most builders. I'm not calling for your arrest if you choose to use or improve on that method.