CMU vs ICF R-Values, cost and efficiency

[Dean M Rantala]

Initial disclaimer: I am not a contractor or any type of mechanical/civil engineer!!

To begin, I have been considering new home construction here in northern Colorado. Given the ridiculous prices of lumber right now, it got me thinking about alternative building methods - specifically ICF. Yup - the ones with styrofoam on each side.

Now, a little background about myself before I continue... I grew up around construction until I was nearly 21. By 2003, I had decided construction was not what I wanted to do for a living and went down the path of software development and systems engineering. That said, I am not a complete idiot in regards to common building practices and methods.

But my interest in the title subject of this post is deepened by my experience as a part-time expat (Hidalgo, Mexico) and the construction methods used there.

Most homes in Mexico are built with virtually no insulation. Instead, we rely on very dense (clay-based) red brick to help form a thermal mass. Now, granted - this is certainly not the best in the hotter or colder regions, it got me thinking about the ICF vs CMU debate.

The ultimate question boils down to this: would it not be more efficient to have a simple CMU wall with a single 4-inch layer of rigid foam insulation on the outside -vs- a ICF wall with 2 inches on each side?

My reasoning is as follows: with an ICF wall, the concrete itself is essentially serving as a thermal-break (WITH thermal mass) between the two layers of insulation. However - if a simple CMU wall had a single 4 inch layer of insulation on the outside, then you would have a much better advantage: You still have the same effective total R-Value worth rigid foam, but you are now utilizing the thermal mass of the concrete to directly help regulate the interior temp of the home.

This question further ties in with the fact that I would consider concrete and CMU for the entire construction (interior walls as well as roof - as is typical in Mexico). This additional concrete not only adds even more thermal mass to the interior, but further eliminates wood needed for construction.

Homes in Mexico do not usually use drywall - the raw concrete walls are finished smooth with a stucco-like blend. Very durable as everything in the interior is basically solid concrete. Very common to see a home in near-pefect condition that is over 100 years old. Mice can't eat the walls, a kitchen fire cannot burn your home down, a busted water pipe will cause virtually no interior damage (just mop up the water)...

Finally, it seems to me (after having watched a few dozen YouTube videos at least) that the idea behind ICF is one of them "death by a thousand cuts" scenarios. Sure, it is "quicker" to stack the ICF forms - vs - laying traditional block. But that seems to be where the savings end. Plenty of wood must be purchased to properly brace the ICF walls while they are poured (one-time use!) and the additional zip-ties, inter-locking rebar and extra adapters to attach the bracing (rental in many cases) adds to the cost and time. By comparison, a CMU wall is pretty damn straight-forward. Cheap mixer from harbor freight, some straight line, a plumbob, only a few straight 2x4s, cheap concrete saw (found all over craigslist) and the materials.

Watching the instructional videos from some of the ICF vendors starts to make my head hurt. All the extra steps, all the additional "little" things that need to be tended-to. The additional cost of the forms themselves (after which, you are STILL paying for the concrete). It just does not add up - when a simple CMU wall with 4" of rigid foam seems to effectively accomplishes the same end-goal.

Now, part of the cost-aspect of this is assuming a DIY job. A contractor who does this every day will likely already have all the bracing, special attachment hardware, will be familiar with the assembly, caveats, etc. But that same contractor will likely negate any possible cost savings that could be realized.

 

[ChorasDen]

Hi Dean,

- Colorado homes often have basements below the first floor, is that your intent for this project? I could see condensation potentially being an issue if you have bare CMU at the interior.
- CMU have different seismic response coefficients compared to wood framed structures. Not sure if you are located in a significant seismic design zone, but this will increase the design loads that will need to be resisted by your structure.
- Others may have to respond to how CMU fits into IRC, if there is not detailing for CMU construction found within the IRC, then you'll need engineered plans, unless your building department waves this requirement. What is the design snow load in this area? You may note meet IRC provisions if your design snow load is high.
- Special inspections and lateral reinforcement are different for CMU compared to wood framed. Can the plans examiner and inspector provide this service?

 

[phamENG]

CMU has prescriptive requirements in the IRC, and I believe many jurisdictions accept ECR reports for ICF and their load tables.

An important thing to consider: vernacular architecture evolves regionally because it best expresses the wisdom of years, decades, or even centuries of building and living in those regions and their climates. Much of what you're suggesting is taking the vernacular architecture of Hidalgo and attempting to transplant it to Colorado. I'm not overly familiar with Colorado, but I do understand that the local climate can vary significantly from depending on what part of the state you're in, particularly with regard to your proximity to the mountains.

Most of your question actually falls outside of traditional structural engineering and more in the realm of architects, architectural engineers, and 'building scientists' (whatever that really means). Placement of vapor barriers, insulation, and structure varies from place to place with temperature and humidity. A sequence that works great in Florida could mean mold growth in 6 months in Canada, and vice versa. A local architect can guide you on local design and construction practices to make sure your building envelope is properly designed.

CMU does have a number of advantages. Chief among them is the ability to inspect it as it goes up. There are several methods that can be employed, and all of them are fairly reliable. ICF, on the other hand, can be prone to honeycombing - large groups of air pockets in the form - but you never know because the forms never come off. A good contractor can prevent this, but you're suggesting a DIY and you wouldn't know how or be able to recognize it if it happened.

ICF has an advantage in seismic design. A good engineer can throw out the prescriptive ICF details and design it as a real engineered concrete structure, which tend to perform better than masonry in seismic events. Masonry can still be designed to do the job, but cast in place concrete is generally better if high seismicity is a design consideration in your area.

You mention using concrete for the roof. That'll be tough if you get snow. You'll either have a cast in place roof that is flat and collects lots and lots of snow, or you'll need to do precast slabs that are sloped. Either way will probably be more expensive than just having wood trusses delivered and installed.

Wood prices have dropped pretty sharply. Steel prices have not. Steel is still used in ICF and masonry construction, so you probably won't recognize as much of a savings as you're thinking. DIY projects are also a lot more expensive than they look or than what they add up to on paper. If you have nothing else to do and have the skills, go for it. But it sounds like you've been away from construction for the better part of 20 years and you have a day job. So you probably won't be as efficient as you were (you're also 20 years older), your skills aren't practiced, and you won't be able to spend as much time doing the work if you have to keep working your day job. Neither of these construction techniques lend themselves well to the amateur building.

 

[Brad805]

ICF is not complicated to install. My supplier showed me in the am and I built a 2,000sqft home using it. I rented the braces needed from the supplier. There is not a lot of wood needed. That said the forms are not cheap given the volume of EPS foam. Some of the different block styles are also very expensive to ship due to their volume.

Comparing how one builds in Mexico vs. somewhere in North America is not exceedingly useful given the cost differential of trades. Material costs will be far more important in Mexico than for many of us. In my locale, we cannot find masonry trades so the ones we have price that in. It is hard work and we have lots of jobs where you can sit in an air conditioned cab and make a great living. Here we put up an ICF wall and finish the two sides. That is typically the same trade. The masonry solution involves more trades of differing costs.

Thermally one can build any of the solutions to achieve the same thermal value. The cost and how one might achieve that will vary. Our trade costs are high, so we aim to minimize the number of steps. ICF fits the bill well here, but I favor concrete over masonry. If it is a DIY project, then material costs will be very important. As a DIY project your skill set will be important as well. The masons I have worked with are very fussy and highly skilled. I do not think laying block is a DIY project.

If you want some insulating data on different wall assemblies here is a report by a very reputable company: RDH Report

The Europeans are figuring out how to build more thermally efficient blocks, but I do not believe that has found its way here: EU Blocks

 

[Dean M Rantala]

So, my comparison to construction in Hidalgo may be un-fair, I will admit.

But at the same time, I think it is healthy for discussion - and will also be relevant for myself on my next build DOWN SOUTH.

On this side of the border - it is common practice for everyone to take things at face value and "go with the flow of the masses". Well.. that is actually just human nature everywhere in the world, I suppose.

"All my friends and neighbors are doing it like this, can't see why I should do any differently.". Codes aside for just a moment - that basic ethos is essentially a large part of what drives the overall decision in most cases.

Yes, we pour the entire floors AND roof (hand-pour, actually) as a single steel-reinforced slab. When done correctly, they will NOT leak and last decades before any slight possibility of water ingress becomes an issue. The slabs are usually flat and often have a sealant applied if water drainage IS an issue. Snow load is not an issue down south, but we do place multiple 2500-gallon water tanks on those same roofs for water storage and it never poses an issue. These homes generally have steel-reinforces vertical columns every 9-10 meters, with the red brick filling the void's in-between. This is due to the rather seismic region in which I live down there.

In many areas, the house sewer also runs directly out into a local stream or river. So don't think I am trying to defend one cultural or regional building method over another here.

The flip side of this is: my next home I build down in Hidalgo, I would like to apply the inverse: most homes in my region have no climate control (no HVAC at all). And I would like to have my new home about to be built down there to feature some level of insulation. So the idea of rigid foam on the exterior and covered w/ stucco would be at the top of my list.

Regarding CMU in residential.. not sure these days, but back in the 90's I lived in the Tampa bay area and there were MANY residential homes built back in the 50's and 60's using CMU's. No insulation, of coarse - and not very energy efficient.

Then again, I have become rather accustomed to the very solid build quality of my current home down south. After years of living down there, you come back into a stick-built home and it just feels... cheap and fragile.

I am going to continue my research into ICF as well - not exactly ruling it out here. But the question of ICF -vs- CMU in my application (especially from an energy efficiency standpoint) still stands to some degree.

I can attest to thermal mass in construction from my lifestyle down south: entering a buildings down there in 100 degree heat or dead of winter and feeling the difference when you walk into these buildings is amazing when you consider - most these places have absolutely no form of heating or cooling. Personally, I have become a firm believer in having high-thermal-mass on the interior of a home.

Some of the really old homes down there are built using very large (2-foot diameter) rocks and were set into walls that average around 2 feet thick. My uncle-in-law lives in one of those types and I tell you: it can be 20 degrees outside or 100 and yet the thermal mass keeps the interior at a near constant 65-70 degrees year round. Not that building a 2-foot thick wall is very practical these days.

Finally, the climate in which I live in Hidalgo is surprisingly the same: 16-20 inches yearly rainfall, similar arid climates, lots of patches of land covered in sagebrush, joshua trees everywhere, 7,000 foot elevation, we just don't get the snow Colorado gets. So labor and material cost aside, applicable building methods are not that far apart.

And I just took a look... yeah - lumber prices have dropped off a cliff from where it was this summer.

 

[Brad805]

"The ultimate question boils down to this: would it not be more efficient to have a simple CMU wall with a single 4-inch layer of rigid foam insulation on the outside -vs- a ICF wall with 2 inches on each side?"
Part of that question will involve what finish being applied to the exterior of the 4" layer. If something lightweight (metal,board/batten..) where you need no other framing, then thermally that could be better. If you need a lot of secondary framing to support your finishes, then it is less likely. The more thermal breaks you have the poorer the wall will perform. ICF only has plastic webs connecting the two sides, so they claim they provide an R-22 wall for the common thickness (I included a study below). If you want to delve into the effect of thermal breaks buy yourself a Flir Camera and go test it out. At the right time of the day I can find every grouted core in a masonry wall with a FLIR camera. The same applies to wood framed walls.

ICF Study: ICF Study

I applaud the effort, and this thought process will become more and more important as our energy costs increase. The problem you have here is to answer these questions scientifically you usually need a mechanical engineer that performs an energy model of the overall assembly. We have worked on net zero projects and LEED projects where these people have been involved, but I have never encountered a residential client willing to pay for that type of study. I am sure many architects may suggest they can do this with Revit, but I would be very suspicious that would be a case of GIGO.

Another thing you need to consider is the percentage of windows. You can drill down on all the walls, but if you have a very large percentage of windows, the thought exercise can lose its value. We have worked with many architects that get totally stressed out about the wall insulation, but then use crazy amounts of glazing. One LEED Gold building in our area has one complete wall elevation that is a curtain wall. The curtain wall is a high quality, but the space can be virtually unusable at times.

What is the typical roof insulation in Mexico?

 

[phamENG]

I grew up in one of those 1950s CMU bungalows in St. Pete. They still do it that way down there now often than not, but it's reinforced masonry now. Most two story houses seem to switch to lumber for the second floor, but that's just from what I see when I'm down visiting.

I agree that both options will give you a generally stronger structure. But as far a the energy part goes, I agree with Brad - you'll need an energy study with a mech and a building envelope design from an architect.

Good luck with the process.

 

[Dean M Rantala]

Roof insulation? Not a drop of it.

Overall, modern home construction (in MX) is not very energy efficient. Fortunately, the government has a simple solution for this.

Homes in the north and along the beaches (where the climate is rather hot and air conditioning is virtually a requirement), the electric rates are VERY cheap via government subsidies (think less than a few cents per kwh).

Closer inland in in the more moderate climates (where my home is), the rates are downright oppressive - as in nearly a dollar per kwh. The idea is: the people who do not "need" electricity end up subsidizing energy costs for those who do.

Any sort of HVAC is very rare in most homes - even higher end luxury homes down there. My solution at the home I have down there now: paid a $3000 bribe to a CFE official to hot-wire a 100-amp drop that by-passes the meters.

For this reason, most homes wired so that the **entire** home is fed w/ a single 30 amp breaker. I am not even kidding.

Back to the topic at hand.. I am ultimately on a quest to merge the best of both worlds. And I would love to apply the "best of both worlds" to both my new home I am going to build in MX as well as a home here back in the US.

On a side note... I write software for a Supercomputing company. I bet one of the production guys may just be able to model this to some degree for me. I personally do not do any scientific modeling (strictly end-user interfaces, reporting and workflow handling) - but we DO have PhDs working at the company with access to serious computing power.

This has me thinking now.... time to bring a buddy into this discussion...

 

[Dean M Rantala]

phamENG - I lived in both St Pete and Seffner. Last CMU house I liven in was just behind the old Lazy Dayz.

Man.. nothing beat the party scene in Ybor city

 

[RabitPete]

I've seen several DIY ICF projects, CMU on the other hand takes some skill to do right. Renting wall forms like those used for basements and pouring RCC walls might be faster and easier than CMU. Hiring a professional masonry contractor will cost you arm and leg, especially if you want it properly reinforced and grouted, think $5 per block just in labor. And you still need to baby sit them as contractors tend to cut corners all the time - reduce rebar splicing length out of convenience, forget to put horizontal reinforcement, use "shovel" method to measure amount of sand in the mortar etc.
I've also seen DIY jobs done much more carefully and thoroughly than most professional ones, it all depends on the abilities and dedication of a particular individual.

I would consider concrete and CMU for the entire construction (interior walls as well as roof - as is typical in Mexico)

You mention using concrete for the roof. That'll be tough if you get snow. You'll either have a cast in place roof that is flat and collects lots and lots of snow, or you'll need to do precast slabs that are sloped. Either way will probably be more expensive than just having wood trusses delivered and installed.

It might depend on the floor plan. RCC roof will most likely require interior load bearing walls, further increasing the cost. Truss construction is dominant in the US as it allows for large clear spans at very low price point. RCC is very common in many other countries, but their houses are typically much smaller. RCC is by far superior in terms of longevity, impact resistance, fire resistance and thermal mass. Also, in most cases small RCC structures have so much dead weight, that properly designed and built structure will naturally resist tornados and hurricanes at no extra cost.

 

[Dean M Rantala]

The roofs down south typically have a built-in rebar-reinforced support beam every 2-3 meters. The integrated support beam usually measures anywhere from 8x14 to 12x20 (depending on the span). When the forms are set up, this is part of the form. The rebar is positioned into the channels (special pre-built latices are sold super cheap just for this application) and either wired together, or welded together. When the floor (or roof) is poured, the concrete is not only being filled for the floor/roof, but also fills into the support truss network as one integral piece when cured.

Pretty common to see a 20 or even 30 foot span depending on the integral truss size and count.

I am waiting to talk with a co-worker to see what it would take to compare a model comparison of a single 4 or 6 inch poured concrete wall with 2 inches of foam on each size vs 4 inches of foam on a single side. The guy plays around with models of the thermal expansion of the universe - modeling 4 walls and a roof should be a walk in the park for him.

Thank you all so far for all the great info.

Brad805: I read through that study, very interesting. Even though some of the terminology is certainly over my head - the point is pretty clear: ICF is still pretty kick-butt stuff it seems.

 

[masonrygeek]

I echo many of the comments related to energy efficiency, in that you need to run some energy calculations for each climate zone to get a real picture of what is going on. Thermal mass works best when it is connected to the interior environment. CMU does exactly that when you insulate it on the outside. An ICF wall has 2" of insulation separating the mass from the interior. With energy efficiency, try to stay away from the prescriptive insulation tables found in the codes and try to use COMCheck or other energy simulation software. The prescriptive tables don't address thermal mass very well since it addresses steady state conditions, which the outdoors doesn't seem to follow.

In regard to construction, there is nothing faster than CMU construction. All you need is a footing and masonry construction can start right away. No forms to build, no waiting for concrete to cure, etc. For experienced mason contractors they are often waiting for sparky (electrical work) or floor joists to continue the construction. The Florida market has obviously seen the benefits and that is why you see so much CMU construction there. And it could be the resistance to mold, floods, fire, wind, etc. that make CMU more appealing than wood. As pham mentioned you can get honeycombing of ICFs, although all products have their own set of concerns that need to be addressed to get quality construction.

Finally, the cost of CMU can be less than ICF construction and even wood (when the prices climb). Check out the cost studies from different parts of the US of CMU vs other materials:

https://www.buildingstudies.org/initial_cost_construction_multi-residential_structures_reports.html.

When the cost of wood is high, masonry is competitive and masonry is often 5-10% cheaper than ICF construction. This is for multi-family construction and may not be comparable when discussing single family homes.

 

[Brad805]

Unless you add some insulation to the roof I am not sure there is much value in fretting about the wall insulation. There are lots of ways to insulate the roof and add a trafficable surface.

 

[RabitPete]

@masonrygeek: You probably mean a plain unreinforced CMU, right? I am not familiar with FL practices and mostly deal with commercial projects. After adding rebar, grout and union mason labor, CMU is typically comes up more expensive than RCC. On larger jobs tilt up RCC, PEMB and traditional steel frame seem to prevail but it might be area dependent. For CMU we have to have someone on site 24/7 to baby seat masons and verify reinforcement as it goes up, while RCC can be inspected once, after all rebar is in place just before the pour.

I've only seen one guy so far using RCC to build the whole house, he did it all DIY with his son and had a degree in civil engineering, though never practiced long enough to get his PE. A typical residential basement takes experienced team a few days to setup forms, pour in 1 day and strip after 1 week. You only need 1 inspection, and 1 nice day to pour. Masonry work has to stop if weather is not favorable or require extra measures at extra costs. I just cant see how a quality reinforced CMU can go up faster than poured in place on a single story structure. Could be a region dependent though, our area is heavily unionized and non member masons have little chance of getting any project larger than a typical 2 courses foundation around a residential slab on grade.

I personally prefer monolithic walls, as they are stronger than fully grouted CMU and literally bullet proof. ICF is a cool idea but as others stated, inspection is not possible and they are prone to consolidation issues and thermal mass ends up sandwiched between the insulation which is not ideal. CMU is great for interior partitions and remodels, especially if fire rating or better durability than drywall is required.

@OP: Another aspect to consider is financing. If you are paying all cash for the land and materials it does not matter, but if you are planning on taking a construction loan/mortgage, most banks don't favor DIY, they typically want you to hire an experienced general contractor to manage the project who will then hire licensed contractors for all the trades. If construction type is not common in the area, appraisers might have difficulty estimating the house value.