Compressed earth blocks
Hi all – What do you know about compressed earth blocks? I’ve been bumping into them a lot lately, and I wonder about them as an alternative to the steel bolted moment frame design I’ve been focused on. The COVID delay has given me more time to explore alternatives.
They seem a lot tougher than rammed earth, and they look good. R-value is terrible, though, at about 0.25 per inch. I guess I fundamentally don’t understand the physics of insulation, since it makes no sense to me that heavy masonry is so terrible at blocking heat transfer. I don’t understand how stuff like fiberglass fluff or mineral wool outperforms thick, heavy masonry.
In any case, thoughts on compressed earth blocks? Would you expect this sort of earthy masonry to last at least 100 years?
Since it’s going to need added insulation, what’s the best practice for placement around thick masonry – should the insulation go outside or inside? (In hot desert like Las Vegas – I forget the zone number.) I’d probably want to use mineral wool. I’m not clear on what happens re: finishing the inside of the walls as far as drywall or studs. I wonder if a service cavity wall would make sense.
Thanks.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
Having also read your thread about steel framing, I have to ask: What problem are you trying to solve?
Wood framing is cost-effective, sustainable, and well-understood. The framing of a house is only 5 to 10 percent of the cost of the finished structure, but trying to do something non-standard will drive up the cost of every other part of the project. The roofer, drywall guys, plumber, electrician and HVAC installer will thank you if you just stick with wood framing.
I just want an excellent house. I also tend to like innovation, I guess. I like to see new ideas and progress. Wood frame seems lame, maybe because I associate it with decades of cheaply built tract homes in Tucson, Arizona.
Steel is much better than wood with respect to rot, warping, pests, and fire. Though its fire resistance is somewhat oversimplified given that steel rapidly loses its strength in fire conditions. And with the ThermaSteel approach, where you turn the studs sideways, you get incredible R-values – you've basically got solid walls of pure insulation, like SIPs. I'm thinking of designing walls with stiff mineral wool instead of rigid foam though.
Masonry always has a place in my heart. Much more solid than wood frame, of course. There's a primal appeal in having solid earth / rock as shelter. I'm just bummed that masonry has such strangely low R-values.
Here in the Southwest, while wood frame is common, it's not as obvious a default as it is in the East Coast. Out here, you see all sorts of masonry and earthen wall systems, from adobe, slump block, rammed earth, full brick, and even simple concrete block. And straw bale. I grew up reading Buckminster Fuller, and lived a few miles from the Biosphere 2, and I've always been excited by interesting and innovative building approaches.
Is high quality lumber still available these days? I thought it was all cheap now or something, maybe due to the trees being very young and small, the ones they harvest in the privately managed sustainable forests where they're constantly replanting forest and harvesting it on a staggered, long-term schedule. I thought all the good stuff was gone, or maybe very expensive, the big, older trees and so forth, with the strong wood. Am I mistaken? I haven't even bothered to look at or price lumber. I assumed it was all cheap these days. In the Southwest high quality homes tend to be made from masonry.
I'm not sure why you'd expect masonry to have good insulation properties. Haven't you even put your hand on a stone wall or walked on a stone floor at room temperature? It feels colder than most other materials (e.g. wood, plastic). That's due to the high thermal conductivity of the material, essentially pulling heat from you faster. Same thing when it's hot. A hot sidewalk can almost burn your feet, whereas wood at the same temperature just feels warm. I would have thought this to be fairly intuitive, even if you don't know the physics behind it.
I don't remember much of my thermodynamics training, but it wouldn't be intuitive for me that a hot sidewalk was an indicator that concrete was a poor insulator. That something is hot from being in the sun might power the opposite intuition – that it's a *good* insulator. It would seem like it was retaining heat instead of conducting it.
And masonry walls in Arizona release heat at night, so that would make me think they were good at holding heat, not conducting it quickly. In general, I think the lay intuition is that heavy, solid materials should be good at blocking things, including heat. The actual physics seems unintuitive to me, especially for earthen materials, and the way metallic foil works as a radiant barrier is extremely surprising, even though I realize that's only one type of thermal transfer.
Steel is also much harder to fabricate in the field. Drilling structural steel on the job site is not fun. Commercial projects go up like kits with all the steel structural pieces pre made (holes punched, brackets welded, beams cut) from a steel fabricator so they just need to be bolted together at the job site.
Steel does loose strength in a fire, but at temperatures way above those when a wood structure would be on the ground in a pile of ash. Fireproofing helps too (a layer of concrete or what I call “smurf barf”). Steel is far and away more fireproof than wood without question.
Newer lumber tends to be less dense than lumber from days of old. This is mainly due to tree rants using species selected for fast growth. The load tables allow for this (codes allow shorter spans than they used to for the same sizes joists, for example), but the new stuff is more prone to splitting in my experience, and also seems more prone to twisting and cupping. It’s not all bad wood though, you just have to be more careful. I’ve also found that real lumberyards seem to consistently have better lumber than the box stores. Using SYP instead of SPF also gets you denser and stronger boards. If I’m doing something picking like maximizing floor stiffness in a minimum size joist, I’ll use MSR graded (machine stress rated) lumber and not the usual visually graded (#1, #2, etc) stuff. You pay more for MSR lumber though.
BTW, masonry is a crummy insulator because of its density and atomic structure. Thermal transmission by conduction relies on certain atomic properties of materials just like electrical conductivity does. Lightweight insulating materials consist mostly of air, which is a very poor thermal conductor. The best thermal insulator is a vacuum (by definition, a void of nothing at all), since there is nothing to conduct thermal energy. A vacuum has to be enclosed though, which makes it difficult to use in normal applications.
Bill
Hi Bill, I don't follow your concerns about steel being a hassle. People build steel-framed homes all the time in America, Australia, etc. Though it's far less common than wood frame, there are 330 million people here so even rarer choices add up to big markets with established practices.
You don't usually see structural steel in steel-framed homes. That's a commercial grade. In homes they normally use light gauge steel, which is differentiated from structural steel. The studs are open C-channels that are the same shape and size as wood studs (except that they're partly open).
It's all standardized, with span lengths and all that sorted out based on the gauge of the studs, joists, etc. I think it's all pre-ordered to the appropriate lengths and so forth. Putting up a steel frame on site is supposed to be a lot faster and easier than a wood frame.
The Blue Sky bolted moment frame actually uses a bit of structural steel, which is unusual. They use Hollow Structural Sections for the corner columns, which I think are key to their design. All the other steel, for cross beams and so forth, is light gauge steel.
I'm building a house right now with some steel in it. The structural engineer specified steel when wood wasn't strong enough to fit in the space provided by the architect. At least around here steel is substantially more expensive than wood, both for the materials and for the cost of assembly.
I didn't mean it was a hassle, I actually prefer working with steel -- everything is straight and square, every time! All the dimensions are what they're supposed to be, every time! Nothing splits when you drive a fastener, etc. I do mostly commercial work, and I design mostly with steel on those projects. I run my drawings, send them to the fabricator (who nearly always redraws everything for their shop regardless), and get my stuff with a usually pretty quick turn around. Nice. Fabricating stuff in the field does suck though.
Residential contractors just tend to not be familiar with steel. They're used to wood. I've even run into residential architects who shy away from steel. I had to run all the load calcs once for a steel beam in my Dad's cottage because the local architect didn't want to use steel, and wood would have resulted in a too-low clearance under the beam. To the architects credit, he did tell me he wasn't familiar with steel because he doesn't work with it enough, so he preferred to play it safe. I just did the engineering on that one beam for the project and spec'ed the material. The contractor was surprised how cheap steel is when done the way the commercial guys are used to doing it.
I think that's the usual reason for issues with steel in residential projects -- neither the design team or the contractors are as familiar with it. The crews are much more used to a big LVL beam than a girder. In my experience, steel is usually cheaper for a given beam than the equivalent wood beam would be, and the steel beam will be dimensionally smaller for the same stiffness and load carrying ability.
I don't use light gauge steel much except for non-load bearing commercial walls. I work with primarily heavy industrial type projects though in the telecom world, so all our stuff is way to heavy for the light gauge stuff. We used big columns and girders with concrete over floor pan. I know of one project that put a 750kw generator on top of an approx 30 story building even. Fun stuff -- they used a sky crane service (big helicopter) to do the lift. This was right off of I75 and I remember the crews having some concern that there would be a wreck on the highway from people watching the aerial crane do it's work.
Bill
I need you to work on my house!
I do wish the light gauge steel was stronger. Most of the gauges are only 33 ksi yield. Some of the suppliers like Clark Dietrich, Steeler, etc. offer studs made from stronger steel, but in every instance they seize the opportunity to make the steel *thinner*. Any increase in strength they just throw away by making these super-thin studs.
I don't understand the mentality of the construction industry, of these engineers and so forth. There's no progress. Any improvement in strength is just an opportunity to use less material. They're just treading water, never producing a net improvement in a home's strength or rigidity.
Or they go backward, like with the "Advanced Framing" that builders are promoting. It's 2 × 6 wood frame using as little wood as possible, 24 inch centers and various other steps to strip as much material from the structure as possible. Yeah, so "advanced".
I'd love to see my uniroof idea at some point, where we'd have a one-piece high strength fiberglass or carbon fiber roof. Forget all these sheets of plywood or OSB – you would have one seamless structure, factory made. It would be so much stronger, and weatherproof, would last forever. Ideally, solar would be almost built-in. We'd just have thin, flush-mounted panels with some sort of toolless install, maybe just snap into place, with factory-made openings in the roof under the center of each panel. It would be super watertight and incredibly strong. The panels would be the roofing material, instead of shingles or tile.
Because engineering is trade-offs. If you can get stronger steel, then you should use less of it to achieve the same result. If you need more strength specify a bigger stud or more of them.
If you build to code, your structur will be more than strong enough. Using more steel just for the sake of using more steel is the opposite of good design.
Most materials used in houses are stiffness not strength limited. The reason your floor joist are a certain size is to limit deflection, not because they will break.
This means that even if you use fancy high strength steels, you can't make the stud/beam/column any lighter as it will deflect too much (all steel is the same stiffness no matter the alloy or heat treatment). There is very little benefit in most residential applications for higher spec steel.
As a side note. I'm currently building something that has both CFS stud and wood construction. The CFS stuff cost about 50% more in materials and take about twice as long. Almost every connection requires a special fastener which adds a fair bit of additional logistics to a build. You can frame an entire house with pretty much two boxes of nails, you are definitely not doing that with CFS.
It is nice to have all the material come to the exact size and dead straight. However, if you are paying for that time, it quickly adds up. So make sure there is a sound reason for picking a construction method.
BlueSolar,
Isn't that one of the chief goals of engineering - to optimize the use of materials? The failures buildings experience are almost never caused by the lack of strength of their structural materials. They are due to those materials being compromised by damage (usually moisture related), or being subject to forces beyond the remit of the building method chosen (tornados, earthquakes, etc.). What are the structural components you want to be stronger and why?
So I subscribe to car magazines, and for years I've read about how a new generation of a car model has X% greater body rigidity or stiffness. It's a recurrent theme of automotive engineering – every time they redesign a car (typically every 6 years or so), they shoot for even greater stiffness and strength. They also tout improved noise, vibration, and harshness (NVH) characteristics, which can be influenced by body stiffness.
The auto industry is always introducing stronger and stronger steels, now called Advanced High Strength Steel (AHSS). And they experiment with other materials. They do tend to try to reduce the amount of steel they use as the steel gets stronger, but they still steadily improve body stiffness, crash worthiness, etc. They don't stand still.
I guess I was assuming that housing could progress similarly if not for the culture of laziness that dominates the industry. In terms of specifics, I was thinking of resilience to extreme weather, and overall stiffness as contributing to reduced noise and rattling. Also, seismic resilience, though I don't think that would apply to a Las Vegas build.
I also assumed that there's a lot of room for improvement in roof strength, since I read that lots of roofs can't support common loads of solar water heaters and even some of the heavier types of roofing.
With walls, Nichiha in particular had higher requirements than James Hardie for fiber cement board siding, such that 20 gauge (33 mil) structural studs won't do. My default approach to things like this is to comfortably exceed the minimum requirements at the time of construction. So I want to have strong steel studs, and an extremely strong roof that could support any foreseeable load, etc.
I might be misguided in thinking that construction could or should be like the car industry in steadily advancing its materials and overall structural strength. Maybe there's not much room to improve tangible outcomes. It seems like there is at least some opportunity to make sure you have robust, resilient wall and roof structures.
BlueSolar,
Sure, I'm all for engineering improvements to address real problems. When cars get stronger they get safer, and perhaps perform better. But you don't see similar gains from beefing up the structure of a house. You can easily improve the structural quality of a house by doing things like changing fastener schedules on sheathing, and adding some Simpson hardware. But the only reason to do so would be to address a perceived problem. What makes a house perform well and improves it's longevity, are almost entirely related to the control layers of the assemblies. So that's where you do see constant improvements in the industry - and that's what you want to concentrate your efforts on. Not over-building the structure without any identifiable aim.
Engineering is all about tradeoff's, as Malcolm mentioned. An example is when I design a battery plant for a customer, they can pay more now and have a longer life before replacement, or pay less now and replace the battery string sooner. Overall, the longer life is cheaper, but sometimes the customer has other budget concerns. It is the job of the engineer to satisfy the project performance specifications at minimum cost, and to design to the customer's needs. Sometimes that means design for lowest installed cost. Sometimes the goal is lowest lifecycle costs. If I was designing spacecraft, it would be all about minimizing weight. Different projects, different goals, and not just "treading water".
The amazing ability of manufacturers to design things that last only days past the expiration of the warranty is a testament to the quality of materials. When the quality is very high, and very consistent, parameters can be known well enough to project out device lifetimes very accurately. Customers might not always see things that way, but it's true! If materials were lower quality and less consistent, it wouldn't be possible to accurately predict how long something would last.
Note that when specifying steel studs, you can specify strength AND GAUGE. If you want stronger, get a stud made from the same steel alloy but in a thicker gauge. Commercial suppliers should have no problem getting you whatever you need.
Bill
"It's a recurrent theme of automotive engineering – every time they redesign a car (typically every 6 years or so), they shoot for even greater stiffness and strength."
Stiffness and strength improvements are a response to a design imperative to use less material. Less material=less weight=improved fuel economy, itself a response to tightening CAFE standards. ICE have pretty much tapped out any efficiency gains to be had any other way.
"They also tout improved noise, vibration, and harshness (NVH) characteristics, which can be influenced by body stiffness."
That's true, but not in the way you seem to imply. Increased weight dampens vibration. Reducing vehicle weight comes at the expense of increased NVH, so it must be addressed. In terms of NVH, it's also useful to consider what the baseline was. If, for instance, GM brags about improved NVH on Silverado trucks, that's not an improvement; they just corrected a huge problem.
"I guess I was assuming that housing could progress similarly..."
It's worth noting that efficiency and (especially) safety improvements in the automotive industry have come about largely because industry was forced to incorporate them into every vehicle.
If the construction industry has anything to learn from the automotive industry, it would be how to market performance levels they are obliged to meet.
Look at straw bale walls. It can be a post & beam structure or load bearing. Good insulation and earth, lime ore cement plaster.
How thick are they usually? How do people manage the windows and doors with super-thick walls? It seems like all the standard components are sized for 2 × 6 walls.
The same way you do with double stud walls or exterior insulation. You add bucks or jamb extensions to the inside or outside.