Thursday, September 30, 2010

Q & A with John Cheah of the University of Auckland

I am very grateful to Mr. Cheah for taking time to share his information with this blog. 

If you'd like to keep up with more earth construction news in New Zealand , be sure to subscribe to the Earth Building Association of N Z newsletter.



What kind of forming system do you use? Any pitfalls or improvements?
We build modular rammed earth panels. This allows us to build walls up to 2.2 meters long and at variable thickness from about 350mm to 150mm. The height of walls that we can ram is limited to 2.4 meters. I've attached a photo of the original red formwork. This formwork is designed to be very flexible in terms of wall thickness.

We have since designed a complimentary set of black formwork with less flexibility (best for ramming 200 mm thick walls) which is about a fifth or less of the weight of the red forms and much easier to set up.We use both and have a few more ideas of how to make it easier to build certain difficult geometries including T sections and L sections.

There aren't really any pitfalls with this system that I've found so far. Single standalone walls are very easy to build using this modular formwork. Because we reuse the formwork we only need enough formwork for one panel to build a house. Building joining walls is a little more complex but with familiarity and experience should take the same amount of time to take down from the previous wall and set up for the next (which is about 45 minutes). Making sure everything is wedged up tight and level is important and prevents issues arising later on.

We sometimes put a strop around the whole system to add stability and safety (in the place of a guard rail at higher heights). The speed at which we have been ramming the Ahipara house has been about 5 panels a week. We can build 2 wall panels a day but it is very physical work so we aimed to do one wall a day and spend the rest of the day preparing for the next day and resting. A 2m long, full height wall takes about 3-4 hours to ram with a 6 person team.

We needed 6 people because we decided to do the Ahipara house without a bobcat and had a large archimedian screw sort of machine instead.

We would typically have:
- 1 person mixing in the right amount and proportion of sand and clay in the screw drive
- 1 person mixing in the right amount of NZ Flax fibres and cement in the screw drive
- 1 person looking after the water addition in the screw drive, making sure it's about right and ensuring everything is getting mixed well (quality control)
- 1 person transporting the mixed soil to the wall panel being rammed (also does some quality control)
- 2 people on the formwork ramming - alternating with the hand rammer and the pneumatic rammer.

What type of footing/foundation?
We have a reinforced concrete foundation around the perimeter of the rammed earth walls/house. The floor in Ahipara will be of stabilised rammed earth - the same as is used for the walls.
It should be noted that for each structural wall panel, there were two D16 vertical reinforcing bars placed 150mm from both edges of the wall panel.

How did you finish the top of the walls and attach your roof? 

Top of wall panels have a reinforced concrete ring beam into which the vertical reinforcing going through the wall panels was bent into. In NZ we have to design for seismic loads and this is the reason why we have steel in our walls and a reinforced concrete bond beam.

How did you attach your doors and windows?
We embedded pieces of wood into the edges of the wall panels when ramming which could later be screwed into.
In Ahipara we rammed the wall panels next to doors and windows flat (with no embedded wood) and will fix a thin piece of timber to the wall panels next to openings and then fix the doors and windows onto those pieces of wood.

What type of construction equipment did you use?
We have used bobcats in the past. For the Ahipara project we decided to use the screw drive. Interestingly we found near the end of the project that hand mixing worked quite well and was fast - but tiring. After working on site for a while though, the guys I were working with were more fit and are generally big and strong so this worked well.
We used one pneumatic rammer and one hand rammer.

How long did it take to build the walls?
There were about 28 walls in the house - this included short walls, window walls and non structural walls. In terms of ramming days it took about 25 days as we were able to build two walls in one day on several occasions. In terms of time, we built the first wall on 23rd April, 2010. We built the last wall on the 17th of June. Once we had reached the physical construction stage, delays took the form of gathering resources, rain, family and community events (including trying to save and then later euthanising a beached pilot whale), occasional disputes and equipment failure.
The Ahipara uku walls are 200mm thick and the Rotoiti UKU walls are 150mm thick.

Any horizontal re-bar? I am surprised at how little reinforcement you are using considering your seismic zone 

No. We have done lots of testing at a small level and at full size in the labs and on the field and did a specific design for the house.

We are fortunate that Ahipara (in the far North of New Zealand) is in the lowest seismic zone in New Zealand. I'm currently doing shear tests in the UK to better understand the shear performance and characterisation of earth so we can more confidently design in more earthquake prone areas.

No concerns regarding moisture ingress with flax fibres?
The flax fibres are a naturally abundant resource in many areas of New Zealand. We include them for:
- cultural reasons - traditional use of flax fibres in many aspects of lifestyle however not in it's current function as reinforcement for rammed earth walls.
- It does improve the thermal insulative properties significantly albeit from quite a low base so it is not significant enough to be a reason to include it just for thermal advantages.
- It increases the tensile strength of the material so cracks form under larger forces and after cracks haved formed the material has a much larger residual strength instead of being prone to cracking further with no reliable strength.
- It reduces shrinkage in the walls and increases volume stability.
- It helps with durability - it acts as a kind of stabiliser binding everything together - in parallel with the compaction and cement and clay actions.
- I am currently doing research on the shear performance and seeing what benefit there is on that property and to what extent. Shear strength is important particularly in earthquake prone areas of which NZ is one.
The cement stabilised walls are more susceptible to long period of exposed water. E.g. through a wet winter and no sun to dry the walls we have had one instance of water getting through the wall. The fibres do not seem to be creating any issues with moisture. Eaves are important and the Ahipara house will likely have eaves out to 2 meters. Sun exposure is also very important.

Are the houses plastered inside and out?

No plastering inside or out.

How did you determine your soil mix?
Soil mix follows commonly accepted limits for rammed earth construction. I use Houben and Guillaud's recommendations. The sand content is around 60%, clay between 10-15%, silts and gravels vary in between this. We do use  around 7.5% portland cement I also do quite a number of other tests included Atterberg limits tests to find the liquid limit and plastic limit and a Proctor Density Compaction test and shrinkage box tests. As a research student at university it is much easier for me to do these tests and there would be considerable cost to get these tests done in New Zealand using an independant testing laboratory.
We have mainly used two different soils so far. The compressive strengths do vary quite a bit but our soil strength must be at least 1.3 MPa to be acceptable for structural use and we have measured strengths up to an average compressive strength of 7.5 MPa. 

Are you concerned about variability in your mixes?
This is more an experience and a process and quality control issue. We minimize variability as much as we can. We got quite consistently high quality mixes and wall panels throughout the project. If any panel was poorly rammed we knock it down. We did it once on the Ahipara project. Due to the large conservatism built into the mix design and house design a bit of mix variation was not a concern. For me, the use of cement was helpful in this sense as it does give a reliable base level material performance. We were using people from the community that were inexperienced initially in earth building and were building in a largely variable climate (hot and sunny / scorching, very windy, and sometimes ramming in heavy rain). Looking at the finished house I am happy with the result and quality.

How many houses have been built since the start of the project?
We have built 4 UKU structures. The first two were single room dwellings, built in 2004.  The first house was built in Rotoiti, the next in Ahipara.

What has been learned with each successive build?

Many things! We've been extensively monitoring the first UKU house thermally and with 150mm thick walls in a area that went down to -7 C in 2009, it performed really well. We've learnt a lot from a construction and practical perspective.

I've personally learned a lot about dealing with council and the engineering design side of the project. The project has been quite a holistic one and I have enjoyed it all the more because of this. Strangely, although it takes more time to get to know the community I'm working with, things actually happen quicker because after a decision is made, everyone knows what is happening and people and resources come together just when we need it.

It requires a little bit of faith and a good portion of optimism but they're rich in resources and expertise up there, and very resourceful such that we have rarely been delayed because of materials and labour. It's actually been council and gaining council consent that created the most issues for us (and this is on-going). Still, we are building bridges with them also and plan to build more of these earth houses in the near future. The second house should be much smoother to work through council as they will be more familiar with the building technique and we will be more familiar with what they want to see and know.

Working in rural Maori communities also has many different dynamics that need to be understood and that can greatly benefit the project. Including things like following their ancient calendar called Maramataka. This specifies that there are good days to work, and bad days to work, days to rest - not based on 7 day week. So we won't have a meeting with council on a particular simply because the maramataka is bad.

The house construction also followed traditional Maori tikanga/protocols and one of those is that females are not allowed on the work site until it is completed. A female walking on site would traditionally be part of the officiating/commissioning process. Heeni, the wife, and a few other female friends and relatives did help out with some of the flax fibre processing and were allowed on site to have a look after hours. All the physical construction was done by males. Heeni did a lot of work on raising sponsorship, community awareness, dealing with council, general communication, keeping track of project, media control, resources and funding.

Maori have such an impressive carving culture, it would be wonderful to see these designs transferred to walls, especially if there are standardized housing plans in practice. Any experimentation with artistic techniques? 
Not yet - although we have managed to get beautiful impressions from the plywood formwork. I have thought about it and several people working on the project and visiting the project have asked about this. I can definitely see this is a valuable thing to explore. Earth walls are built from the soil their ancestors lived on and that provided for them so it is very valuable and meaningful  (and beautiful) already but with some art / symbols in them would become so much more valuable.

Do the houses have a heating system?
Rotoiti was meant to have a wet back fire place but it hasn't yet been installed. They used a gas heater through the worst of winter. Generally speaking the energy usage was still very low.

I was impressed by the community's insistence on houses that last six generations
On this point - Maori are connected with their land. They have no intention of moving or selling. Everything is inherited by their children. They don't actually see themselves as owning the land either. They are guardians, this is their role and responsibility. Houses that last 6 generations was decided as a good target lifespan to towards which to work.


Weyimi Lori said...

WOW!!! love your work and extensive research. I love the idea of the house been able to last for 6 generations and its relatively cheap and fast to build. How can i get more information on this subject and be properly informed and ready to build one.

Signal1 said...

Hi there and greetings from Australia. Great article and interview. I would really like to know more about the construction of the rammed earth floor. Is there anywhere I can look up how this was installed and the process involved? Feel free to email me on

Look forward to hearing from you.