I've been on here for a while looking at different building options/techniques as I, someday, hope to build my own house. I know relatively nothing about building practices other than what I've read in these forums and ASHRAE journals etc. I'm in the Oklahoma city area and I'm not sure what zone that is (quick side note: if someone could please tell me what zone it is and what exactly it means by zone, climate zone maybe?), but I've become particularly interested in ICF construction because of it's overall strength and potential for high thermal performance.

However, through my readings I've seen that average ICF walls (2.5" EPS on either side) will yield approximately R-22. Now let me get to my question. . . some may say that R-22 is sufficient for my region (I believe R-19 is the minimum recommended and yes I could check codes, etc.) but for the sake of discussion let's say I want more. Not a specific number in mind, just ideas for a little extra. A couple of ideas I've been mulling around are:

1: Adding just a simple 2x4 frame on the interior and fill with blown-in celluose or fiberglass batts

2: Using SIPs on the interior (or maybe even just extra XPS or polyiso on the exterior it could be as simple as "More Foam!!")

Now let's assume that the necessary precautions regarding water, air, and vapor barriers are to be taken. I'm simply trying to gather options for enhancing thermal performance while taking into account buildability and durability. If someone has ideas other than my 2 please post them. I love looking at all options. I'm not so much interested in cost either. That's for later as it will be at least 5 years before I attempt to build.

I appreciate the input/opinions and look forward to hearing from you guys/gals.

Many ways to build and have increased insulation.  Some systems will result in more work and materials which will drive up the cost.  When given a choice, I prefer a simple system with high R-values that is fire, storm, water and termite resistant. 

Your two ways would increase the R-value but at what cost and how much used up space.  Using SIPs would have more continuous insulation than adding a simple 2x4 frame filled with cellulose.  (I would not bother building a wall to fill with fiberglass.)

Oklahoma City is US climate zone 3:  http://www.buildingscience.com/docu...ns_web.jpg

Your IRC 2012 code minimums can be found here, in the zone-3 row.  An R22 ICF would run well over code-min for mass-walls, since it has over R10 on the exterior.  For a CMU or a poured concrete mass wall with foam on only the exterior you could get away with 2" of EPS rather than the 5" of an R22 ICF and still meet code. If all of the foam is on the interior of the mass it would take R13, or about 3"of EPS. 

R20 would be the code-min cavity fill for 2x6 stick-built (not R19).  That's more than a subtle distinction- R19 batts only perform at R18 in a 2x6 cavity even during mild weather, but are so low density that they perform much worse than that at the temperature extremes.  High density fiberglass R21s or rock wool R23s don't have that problem and increase slightly in performance during cold weather, decrease sligthly during mid-summer heat wave  temps, but always beat R20 if installed properly (which is another issue you have to deal with when using batts.)

1: An insulated 2x4 studwall gives you another ~R9-R10 after factoring in the thermal bridging of the framing, but putting it on the inside isolates the conditioned space even further from the thermal benefits of the thermal mass of the concrete.

2: A SIP on the interior has a similar issue of isolating the thermal mass.

If you want to go higher performance with ICFs, use asymmetric foam, with the lesser amount on the interior, thicker stuff on the exterior. Many vendors sell asymmetric ICFs. Though most people opt for the 2.5"/2.5", if there were a 2"/3" available, it would outperform the symmetric version.  Some vendors still sell 2"/2", but I've yet to see a 2"/3" being marketed.

From a cost/benefit point of view, once you're north of R22 for whole-wall R the window performance dominates the heat gains/losses through the sides, so you have to upgrade those to something more appropriate, and start looking at your absolute air-tightness, attic-R and sub-slab R, as well as reducing thermal bridging factors where walls/floors/ceiling meet.  Take a peek at table 2, p10 of this document, which is a starting point for where the economics of going higher R may go super-long on financial returns.  For zone 3 they're calling R20 whole-wall about it- going higher than that requires adjusting all of the other R-values commensurately, and adding more to the wall without doing so wouldn't be the best bang/buck. An R22 MASS WALL would slightly outperform a lower mass R20 whole-wall solution.

If you took an R22 ICF up to R26-R28 with an extra inch of foam on the exterior, that would be about it, since EPS & iso aren't exactly lowest-cost R, and from an annualized energy use point of view at R26 it would be close to behaving like R30.  If you look at the table in that document, R30 is their estimated economic balance point for zone 5, so if you do that, use the zone 5 numbers as the starting point for R values on the rest of it- putting an inch or two of EPS under the slab (running right up to the inner foam of the ICF, floating the slab with no thermal bridging) and another 8-10" of cellulose above code-min in the attic would have a bigger thermal benefit than putting more foam into the walls when you're in the mid-20s for mass-wall R. Better class U-U0.22-0.25-ish low-E gas-filled double-panes for windows start to look pretty beneficial by comparison to fatter foam on the walls too.

It pays to avoid large areas of glass on the east & west exposures, because low-E or not, mass-wall or not, even with low solar gain window areas on those sides (particularly the west side) will cause cooling loads to soar.  Put in enough for daylighting, but beware the "sunset view" picture windows- they roast you, and add real tonnage to the cooling system requirements. Big windows on the south side can always be shaded by overhangs, but that's impossible to do well on the east & west sides, since the sun is much lower in the sky on those aspects.

Posted By HVAC-Engineer on 18 Jun 2013 03:49 PM
I've been on here for a while looking at different building options/techniques as I, someday, hope to build my own house. I know relatively nothing about building practices other than what I've read in these forums and ASHRAE journals etc. I'm in the Oklahoma city area and I'm not sure what zone that is (quick side note: if someone could please tell me what zone it is and what exactly it means by zone, climate zone maybe?), but I've become particularly interested in ICF construction because of it's overall strength and potential for high thermal performance.

However, through my readings I've seen that average ICF walls (2.5" EPS on either side) will yield approximately R-22. Now let me get to my question. . . some may say that R-22 is sufficient for my region (I believe R-19 is the minimum recommended and yes I could check codes, etc.) but for the sake of discussion let's say I want more. Not a specific number in mind, just ideas for a little extra. A couple of ideas I've been mulling around are:

Oklahoma City is shown to be a "humid subtropical climate" and you are Zone 3.

In the end when it comes to insulation, it all comes down to cost vs. ROI. You can build a R-50 wall in a Zone 3 climate but why? You won't see a ROI for 150 years or so. With ICF you will get an air tight home (which is very important) and a structure that is more resistant to severe weather.

For a climate like yours the ICF R-Value of R22 is more than sufficient. I am in Zone 4 and R-22 passes for the 2012 IRC and I would not spend more money trying to bump the R22 walls. I would focus on the windows, roof and doors. If I were you I would try and get a R-50 roof and get some great quality triple pane windows with an R-5 or higher. I am oversimplifying it but that is what I would focus on.

I would agree with Lbear and Dana1, Ive just spent my first year in a new ICF home and after a standard 2 x 4 older house, then a 2 x 6 tract home new in 1983, then a custom passive solar double wall homebuilt in 1987, which I took great pains in sealing, the ICF home wins hands down. Pay attention to all that has been mentioned, window placement, thermal specifications and size, insulate under the slab and bring it to the foam of the wall, 2 inches of blown foam on the attic floor and top up with blown in insulation. You will be more then happy with the results and qulaity of living in your new home! John

With a good design for low thermal bridging, better (and well placed) windows, it's possible to hit Net Zero Energy (with a photovoltaic array that actually fits on the roof) and a few mini-splits with R22 mass-walls in Oklahoma City, but it would be a bit easier at R30.

High performance houses are more than simple ROI measure in IRR or NPV terms on the energy use savings- the comfort factor at the temperature extremes is real. With an R22 mass wall the interior surface temp of the finished wall will never be even 2F away from the average air temp in the room. An R60 attic / roof and better-class windows would have similar comfort benefit.

At those kinds of gross R-values, the key to actually reaping that performance lies in limiting thermal bridging and paying close attention to air sealing, eg: A 2x12 (R13) top plate to an R22 ICF with R50+ cellulose above looks like a red-hot stripe in an infra-red image, and moves far more heat per square of area than any other part of the assembly. If it's leaking air due to lack of attention to air sealing it looks like a red hot stripe with flames.

Posted By Dana1 on 19 Jun 2013 03:08 PM

At those kinds of gross R-values, the key to actually reaping that performance lies in limiting thermal bridging and paying close attention to air sealing, eg: A 2x12 (R13) top plate to an R22 ICF with R50+ cellulose above looks like a red-hot stripe in an infra-red image, and moves far more heat per square of area than any other part of the assembly. If it's leaking air due to lack of attention to air sealing it looks like a red hot stripe with flames.

My architect drew up a detail that put the top plate in the center of the ICF wall and 2.5" of foam on the interior portion and exterior portion. He also used a ripped 4x6 top plate. This provides a thermal break and bumps the top plate area to R-23 ' ish. The 12" steel SIP lays down over the top plate at a 5:12 pitch with a 24" overhang. The steel SIP gets screwed down into the wood top plate.

The windows are also "innies" and recessed into the ICF wall and this has been proven to show higher R-Values on the frames and protects the windows from wind washing (I think that is the term).

A lot of small "details" but it adds up in the end for a better energy efficient home.

That's definitely where the focus should be for optimizing the design. The cost of adding that very substantial thermal break at the top of the ICF is huge compared to cruder methods at that critical transition point, far less than adding an inch of foam to the whole thing, with a much larger performance improvement.

I wonder if HVAC-Engineer is even reading this though.

Yes Dana1 I've been reading and I must say wow. This information is incredibly helpful. I knew things like better performing windows, roofs, and slabs would definitely play into whole house performance, but I didnt know that extra insulation on the interior would actually hinder hinder the ICF performance. Especially when you finally do factor in the cost/performance ratio it certainly is not worth it. Thank you so much for the information and if you have any other tips please keep them coming. I never even thought of the asymmetric ICF, or the SIPs at the top plate. It does sound like thermal bridging is definitely the hurdle to overcome here and it will take much more research on my part. However these tips are excellent and will definitely make sure they are implemented.

As of now, my rough idea for the house is to have 2 floors with basement. I'm guessing it will end up being 2500-3000 Sq. Ft. I'd like to do a complete concrete enclosure simliar to nd96's thread in the ICF forum. I like his design for the most part. I'd just like to add a basement as well. I don't think I would do a concrete roof but i am certainly open to suggestions. Thanks for tips too. I agree it's the little things that add up to huge performance and savings. A net zero house would be amazing for me.

Posted By HVAC-Engineer on 20 Jun 2013 12:44 PM
....I'd like to do a complete concrete enclosure simliar to nd96's thread in the ICF forum. I like his design for the most part. I'd just like to add a basement as well...

He has a "basement" under that new ICF home.  Read all about it here:  http://www.greenbuildingtalk.com/Fo...fault.aspx

If you have occasion to head south to Austin TX, the building codes there are ramping up to "Net Zero Ready" by 2015, meaning the designs will be required to be able to hit annualized net-zero-energy with PV that fits on the house. I'm sure there are several examples of existence-proofs using a number of different methodologies. If you google on the terms:

austin texas net zero homes

You'll find an number of them. While the annual average heating loads in Oklahoma City (zone 3) are higher than Austin (cool edge of zone 2), the overall methods of getting there won't differ by much.

Managing solar gains is essential for any Net Zero design, balancing the trade offs between desired passive solar gains in winter and the extra cooling loads too much glass would add in winter. There are some reasonable freebie tools out there for making adjustments out there, but they're not all super-easy to learn. BeOpt is an interface to an energy use simulation tool (DOE2) for the purposes of zeroing in on the relative cost effectiveness of different aspects. It's not a tutorial on green building techniques, but can be fairly useful for adjusting window sizing/types and site-shading and other design factors when net-zero is an underlying goal, since you get to see the net effects on energy use, and compare the relative costs. Designing a high-performance house is an iterative process, so it's good to have a good handle on the general principles before entering the whole design, but it's a very reasonable tool for getting a handle on things. See:

http://beopt.nrel.gov/downloadBEopt2

It's likely that a 2500-3000' house designed to Net Zero would have peak cooling loads well under 3 tons, and peak heating loads under 2 tons, which would put the house within range of very high-efficiency ductless heat pumps, which would be cheaper and about as efficient as geothermal heat pumps in your neighborhood. The 99% outside design temperature in O.C. is about +15F, which is well within the operating range of most of the better ductless heat pumps these days (some are good to -10F or colder), and has become a common way of heating & cooling Net Zero homes even in much colder climates. But of course, being point-source heating/cooling, it works much better with open floor plans than plans with high loads in doored-off areas, since the cost-per interior head or cassette is significant. (I probably don't have to go into too much detail for someone with the handle "HVAC-Engineer", eh? ;-) ) I'm not sure how many people are heating with ductless in your area, but designing the floor plan with ductless in the back of your mind could put you into HSPF 10+ & SEER 20+ territory, blowing away most ducted systems on real-world efficiency, reducing the size of the PV array needed to hit Net Zero.

If you are going with a truss roof, one of the simplest ways of increasing the r value of the roof to wall junction is to raise the wall an addition six or so inches and having the truss designed to drop down inside the wall. That way the ceiling will be 6" lower than the outside wall similair but slightly different than a high heel truss.
I would like to see the top of the concrete core covered with foam as well such that the concrete is totally encapsulated in foam.

Excellent information everyone. Gimme all you got!! I love all the information

arkie6-thanks for the link, I was wondering about his basement/storm shelter/nuclear fallout safe as I was reading about his house. I'd love to do something very similar as his but I'm afraid to even imagine what he's spending on a beast of a project like that.

Dana1-thanks for your link as well. It will come in real handy for sure, especially when I'm ready to consider utilities and HVAC design. (I can't wait for that. :-D) I will certainly have to plan a recon mission down to TX and see what all they have to offer. It will be amazing to see those types of houses up close and in the flesh (or concrete).

FBBP-since you touched a little bit on roof construction, please give me your full opinion/preference (others welcome to comment). Would you go with concrete or truss or something else? Why? And how would you layer it? I've read ASHRAE Journal's May '07 Article "The Perfect Wall" as I'm sure many of you have, and it also touches on "The Perfect Roof" and "The Perfect Slab". If you were to build a truss roof, would you layer it in this manner? Please enlighten me on the specific layering as I know squat about building roofs. Also if the preference is a concrete roof, would you layer similarly to the ICF walls (with proper adjustments to acheive the appropriate R-value of course)?

Posted By Dana1 on 20 Jun 2013 11:29 AM
That's definitely where the focus should be for optimizing the design. The cost of adding that very substantial thermal break at the top of the ICF is huge compared to cruder methods at that critical transition point, far less than adding an inch of foam to the whole thing, with a much larger performance improvement.

Speaking of wood top plates on an ICF wall. Can one apply a peel & stick membrane on the wood top plates and then transition that membrane onto the ICF/EPS exterior wall? Or is P&S not recommended to touch EPS?

when detailing my ICF house, I had a hole pile of details to create thermal breaks, air sealing, etc. Come time to do them in the field, I realized how challenging they would be so I revised some of the details. Some may result in a small energy ding, but I think it made for smoother construction. I was going to recess the top plate into the forms to give me an insulated top plate as well. However connections were more challenging for the trusses, not to mention the concrete pour being a smooth 1.5" down. I decided I valued strength in connection more than the small thermal bridge ding in energy. My detail for the top plate was this: 2x10 plate set 1/2" from the face of the exterior foam. I installed sill seal between the top plate and concrete, and laid a bead of expanding foam onto the outer foam prior to placing the top plate. This should create a semi decent seal between the plate and the concrete, but the foam bead is on the outside for the first protection against air coming into the house. The trusses have a 14" energy heal, allowing nearly the full r60 to be above the top plate. I then ripped sheathing and nailed it to the trusses and then down to the top plate. This created a much stronger connection of the roof diaphragm to the ICF. I still installed truss straps on the inside to the top plate as well. I then had the attic baffles spray foamed in place prior to the insulation blown in. This seals the truss heal and forces the venting up the baffle and reduces any wind washing of the attic blown insulation. I also had the attic foam sealed prior to blowing insulation.

Design is important as mentioned above. I spent a lot of time on window sizing and solar studies besides the focus on r value and tightness. Now that I have gone through a winter and am now entering summer, the angles and sun projections are scary accurate compared to the modeling, which I am honest amazed....The ICF forms I am using are averaging about r28 on my build. My original goal was closer to 40 with double stud framing (zone 6) however there are other values to ICF I ranked higher than the energy savings differences my reports were spitting out going from 28 to 40, which was a lot smaller than one would think. I elected to go with tuned triple pane glazing instead of increasing r value past 28.

Posted By Lbear on 20 Jun 2013 10:00 PM

Posted By Dana1 on 20 Jun 2013 11:29 AM
That's definitely where the focus should be for optimizing the design. The cost of adding that very substantial thermal break at the top of the ICF is huge compared to cruder methods at that critical transition point, far less than adding an inch of foam to the whole thing, with a much larger performance improvement.

Speaking of wood top plates on an ICF wall. Can one apply a peel & stick membrane on the wood top plates and then transition that membrane onto the ICF/EPS exterior wall? Or is P&S not recommended to touch EPS?

 I don't know of any restrictions for using those types of membranes in contact with EPS.  Some solvents like those used in standard construction adhesives are a definite problem with EPS, but I'd be surprised if they used anything that aggressive & volatile for the adhesives in the membrane products.  You could ask the manufacturers, I s'pose.

TAMCO 60 mil peel-and-stick waterproofing membrane (TW-60) is what I used to waterproof my ICF basement walls. No harm to the ICF foam at all. The TAMCO mastic that you use with the peel-and-stick to seal the edges for waterproofing does melt the foam a little bit as it is a viscous ruberized asphalt mixture. But once it has dried, it is thoroughly fused to the ICF foam.

I will say this also. We put some of the TW-60 up without primer and some with primer. It definitely stuck better to the ICF with primer applied first. This is after thoroughly cleaning the ICF beforehand in both cases. I couldn't locate any of the TAMCO recommended primer, so I used 10 year rated elastomeric roof coating as the primer. It worked great. It is a thick white silconized latex paint. It is rated for application on EPS foam.

Posted By Dana1 on 20 Jun 2013 11:29 AM
That's definitely where the focus should be for optimizing the design. The cost of adding that very substantial thermal break at the top of the ICF is huge compared to cruder methods at that critical transition point, far less than adding an inch of foam to the whole thing, with a much larger performance improvement.

Here is a similar detail (wood SIP instead of steel) but it has the foam on both sides of the ripped top plate. I would also probably run the stick & peel membrane between the top plate and SIP:

I like the idea of peel-and-stick membrane. I assume it would cut some construction times.

Lbear - On this detail, would the membrane go over the top plate to the interior side of the inner EPS and then down the side of the exterior? With The SIP, can you just put roofing directly on top (with proper moisture protection) or would you need an air space for venting? If you're not going to do an attic space and it is conditioned I mean to say. The R-value should be sufficient enough with 12" (48-60) right? I'm just not too familiar with how roof construction changes when you are doing a conditioned space directly underneath (no attic).

Futhermore on roof construction, what do you build to fasten the SIPs to? I've seen a few threads stating you don't necessarily need a truss system but I can't wrap my head around it. I know houses have cathedral ceilings and say I want to have a finished, conditioned space directly under the roof (like a finished attic) how would you build the frame to support the SIPs? *whew* so much to consider when building a house