The Concrete Chronicles – Part 1

As some of you who are closer to me may know, or those of you who read this blog consistently, I am a really big fan of Christopher Alexander, and especially of his book A Pattern Language. The book can teach you how to build practically anything, from a country to a doorknob. However, once it gets into the idea of actual construction, it focuses a lot on the use of lightweight concrete, which has about the weight and density of wood, but is many times stronger. In pattern 207 (Good Materials) of A Pattern Language, Alexander encourages the use of concrete “whose densities lie in the range Of 40 to 60 pounds per cubic foot and which develop some 600 to 1000 psi in compression…a range of mixed lightweight aggregates, containing vermiculite, perlite, pumice, and expanded shale in different proportions, can easily generate 40-60 pound, 600 psi concretes anywhere in the world. We have had very good luck with a mix of 1-2-3: cement-kylite-vermiculite.” Since this pattern is so fundamental to the basic construction of nearly anything in A Pattern Language, I decided to experiment with lightweight concrete.


The first problem I encountered was that there appears to be no such thing as kylite, one of Alexander’s recommended aggregates. All I could find on the internet were other people quoting Alexander and adds for skylights (Put an “s” on the front and misspell it, you’ll get there). So I went looking around for other lightweight concrete formulas, and found out for the most part that Alexander’s formula still stands if you just replace kylite with sand. I also found that vermiculite and perlite are more or less interchangeable in these formulas, so I wanted to get both so that I could compare their properties. Based on a few different sources, I came up with a recipe of two parts cement, four parts sand, six parts perlite/vermiculite, and three parts water.

Perlite is used as a garden aerator and is actually fairly easy to find at any garden store. Vermiculite, on the other hand, was not so easy to find. According to Home Depot’s website, there are two stores out of the dozen or so in the Philadelphia area that sell vermiculite, neither of which are the ones that are reasonably close/transit accessible. So the next step for me, since my license has expired and I am therefore not legal to drive or use PhillyCarShare, was to find someone willing to drive me to the Upper Darby Home Depot to pick everything up. Eventually Brad Packer from the ward volunteered, and this morning we set out for the store.

At the store, we went and got a 2×4 cut into 30-inch sections with a six inch block left over. We found three cabinet hinges and two clasps. These all would be used to make the concrete form. We then got a bag of sand and a box of cement. I would like to point out that I am quite happy with my purchase of Cement All, partially because they were one of the few manufacturers that sold it in packages that were less than 50 pounds, and they also have a carrying strap and a resealable package, which is great for non-contractors like myself. We headed back to the garden section and…no vermiculite. Apparently Home Depot’s website is full of lies. But they had plenty of perlite, so I grabbed a bag of it and we loaded up the car and took it home.


I first cut the last six-inch block into four inch and two inch sections. I then took the four-inch and two of the 30-inch sections and attached them with hinges to the third 30-inch section. This is so that when the concrete is dry, the sides can fold down for easy release. I left the two-inch block unattached so that it can be moved to create sections of different lengths. I put the clasps on the end of the assembly to hold the sides up while the concrete dries.

The concrete form.

By the way, I pretty much did all of this in my kitchen. More on that later.

I then rinsed out a gallon milk jug, quickly made a funnel from a piece of printer paper, and poured my cement and sand into the jug. I then got out a real funnel and added the water, because one of the columns I had read said that it was important to mix the other ingredients first and then add the perlite. Perlite is a really interesting material. It weighs next to nothing and looks like something between corn snow and Styrofoam balls. I added this into the jug, put the lid on, and mixed vigorously for two minutes or so.

My cement mixer.

I started pouring the mixture into the form, and after I got impatient cut the top off the milk jug to speed up the process. The cement came out in thick, heavy globs. The concrete was so thick that it wouldn’t slide down into the form on its own, so I quickly squished my paper funnel into a paper trowel to level it.

This is a real professional job here.

All in all I filled a space about 27 inches long. And then I waited.

Letting the concrete set.


The cement I used was “rapid set,” and was supposed to be set in 15 minutes and cured in an hour (a really exceptionally short time since most concrete can take days or weeks to fully cure). After about three hours, I busted open the form.

See, that’s why I put hinges on it. Pretty cool, huh?

I started tapping the concrete with a hammer to loosen it, something that any wood would stand up to, and a section about eight inches broke off at the bottom. This didn’t bode well. I had two tests in mind to examine the density of the material: the nail test and the saw test. In both cases, the material was, if anything, considerably less dense than wood. The nail went in quite easily.

Simple drywall nail in concrete. Sorry for the bad image.

The saw in particular was very telling. My saw is fairly dull, and it took me more effort than I would have wanted to cut the six-inch wood block into two. My saw went through the block like it was cardboard before it split after I got through about two thirds of it.

The broken block. The solid gray part is where it broke without the saw touching it.

It was around this point that I read the warnings on my box of cement which said something along the lines of “do this in a well-ventilated space with a lot of safety gear or else you’ll get cancer,” so I took the remainder of the operation out onto my deck. The main portion of the block was stuck fast to the form, and as I hammered it to get it to come off, it was like hitting a well-built sandcastle. This stuff was nowhere near hard enough to build a house out of. When it finally came off, I took it in one hand and gave it a good solid whack with the hammer. It broke right in half. No two-inch wooden board would break that easily.

What a mess.

Lessons for next time

  1. Go outside!
  2. I feel that I may have been impatient in only giving it three hours to cure, despite the assurances on the box. Next time I will try giving it a solid 24 hours.
  3. Later research into Cement All showed that their product already includes some sand. This could be why the concrete initially came out so globby. I may try replacing the four cups of sand with four more cups of Cement All and see what happens.

That’s all for now, but don’t be surprised to see another edition of The Concrete Chronicles very soon!


About Dave Munson
This blog is about architecture, cities, and myself.

4 Responses to The Concrete Chronicles – Part 1

  1. Dave, Sept. 5, 2011

    Sounds like your Concrete Chronicle was easier read than done. I remember from the Concrete Lab I took in my college days the the various concrete mixtures we students had to mix, test for slump and compression according to ASTM standards, and then compare the different mixture results in graph form. It was mighty interesting!

    I’m thinking if you make contact with the Civil Engineering Dept. at Penn they’ll be happy to let a graduate student use their lab with concrete cylinder forms, slump test cones, and hopefully even their compression test equipment, so they can add the above lightweight concrete mix results to their data base. This would give you valuable comparative information with the most common concrete mixtures used in construction today and familiarize you with standard concrete field testing methods. This might save you the travel and expense mentioned above, especially if the CE Dept. has the materials on hand and will let you use them. They may even have test results on the mix above already in their data base.

    If you’re already familiar with the standard concrete test procedures from your undergraduate days, and are headed in a different direction with your home tests, please disregard all the above. Ha!

    Uncle Bob

    • Dave Munson says:

      Wow, that is very helpful, I will have to see if I can use those facilities here at Penn. For now though, part of this is seeing how much I can do on my own (a big part of Alexander’s book is that all his stuff can be done by laypeople, and I am a very lay person). But after I find a mix I like I should take it to campus and test to see how close I actually am to the compression I want.

  2. K says:

    Very interesting. Sounds like you’ve taken the first step – more to be done.

    Karen Hill (Heidi’s aunt, Lars’ sister)

  3. Pingback: Limited Urban Space Inspires Versatile, Completely Adjustable Apartments – PSFK « Munson's City

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: