Month: April 2020

In this post I’m going to describe how the walls were built. By this, I mean the timber frame and the hempcrete infill. The exterior weatherboarding (siding) and interior lime/hemp plaster will be dealt with in separate articles.

Building the timber frame

The building uses a simple stud wall system fixed on top of the previously constructed base with an OSB floor.

Pictured below are the walls in the early stages of construction.

If you study the picture closely, you can get a good idea of how the walls were tackled.

The image below was from a French architect’s site (Amnios) and shows the structure very clearly.

This page on their site no longer exists, so I can’t link to it. Nevertheless, the same architect helpfully mentioned that the studs are normally 120mm x 45mm, spaced 60cm apart. So that’s exactly what I chose for the tiny house. I addition, my walls would have 50mm of lime/hemp plaster on the inside and an exterior weatherboarding thickness of 60mm (Including an air gap). This brought the overall thickness of the walls to around 230mm.

Going back to my tiny house photo above, you can also clearly see the 3 rafter ties on top of the walls. These were added to stop the walls from spreading under the load from the tiled roof. Some leftover joist hangers were perfect for fixing them to the top plates of the walls.

As the framing progressed, horizontal noggins and diagonal braces were added, as can be seen below. Scraps of timber were also fixed to the sides of the studs to act as a mechanical key to lock the cast hempcrete in place.

I later learned that, such is the rigidity of set hempcrete, noggins and diagonal braces are unnecessary. Furthermore, they also make it much more awkward to put the hempcrete in place. If I were to construct using cast hempcrete again, I would omit the noggins entirely and try to devise a system of temporary diagonal braces that could be removed as the work progressed.

It’s probably worth pointing out that the timber used for the walls was untreated douglas fir. All the nailing was done with a Hikoki gas nailer using 65 and 95mm galvanised nails. Having a good nail gun really helped speed the work up and I’d recommend it to anyone undertaking a similar project.

I should also give credit to my ex-colleague and master carpenter Robert (aka magic Rob!) who pretty much took charge of the framing. Having his experience on site was invaluable. I could possibly have muddled through on my own, but I would have been much slower and more error prone.

Adding the hempcrete

You can read in this blog post how I cast the hempcrete around the timber frame. Basically, sheets of OSB are screwed either side of the timber frame and wet hempcrete placed into the void. The next day the shuttering is removed and reattached to form the next ‘lift’. In my case the shuttering on the inside of the wall was spaced from the frame by 5cm to allow the framework to be completely covered by the hempcrete.

It’s not a complicated process but it seemed to take forever. This was largely because I did it on my own: the shuttering, the mixing and the placing of the wet hempcrete. For prospective self-builders I would recommend trying to get a few people along at this stage of the process, otherwise it’s a real slog.

Another thing that didn’t help was the extreme heat. Doing hard work in 35deg C is tough. Especially for someone more used to the Scottish climate. Such were the time pressures of the project, I couldn’t afford to wait until it cooled down. Indeed, I think the only time I took a day off was due to torrential rain!

Anyway, to help cope with the sun and heat I often wore a ‘legionnaires’ cap (protects the back of the neck). They can be had easily on eBay for a few pounds. I also wore an old, light cotton shirt. On the hottest days I had a routine whereby I would regularly soak myself with the garden hose. The water evaporating from the cotton shirt and cap really helped cool me down, especially if there was a breeze.

Another trick I used was setting up a makeshift shade from a plastic tarpaulin. This was used over the concrete mixer to give some respite from the sun while making the hempcrete. 4 wooden ‘poles’ (from roofing battens or similar) were used at the corners and the whole thing braced using camping guy ropes. This arrangement can been seen on the left of the photo below.

Getting back to the building …

As the hempcrete casting progressed, I found the shuttering boards getting progressively smaller as they were cut up to do the fiddly bits around the windows and doors.

Little by little, the gaps in the framework were filled in with grey porridge, producing some very solid walls.

In the picture above you can also see the boards that were used as window liners. These were wide (and unseasoned) so they inevitably cracked a little and were not very flat. In hindsight it would perhaps have been better to have made these from exterior grade plywood.

The picture above shows as far as I got during the first summer of construction. In fact, I ran out of time and so had to leave the triangular area of the gable end wall until the next visit. The exterior face of the wall had been weatherboarded and the black surface which is visible is the fabric rain screen. So no water got into the building, despite the ‘hole in the wall’

On my next building visit, this triangular shape was filled in using a ‘lath and plaster’ type technique. This was done (again!) due to time constraints. I simply screwed some strips of scrap timber horizontally over the studs and pushed wet hempcrete into the gaps.

Admittedly, the overall hempcrete thickness was somewhat reduced by this method but it was quicker and easier than trying to cast hempcrete in this awkward space.

So, that was it as regards the main ‘body’ of the hempcrete walls. The next step was to apply the 5cm thick hemp/lime plaster. You can read about that here.

In this blog post I describe how I plastered the tiny house walls with lime and hemp and then limewashed them.

Mistakes were made, as they say, and I’ll own up to mine during this phase of the build!

*Note: I use the term ‘plastering’ in this article to mean the application of a finish coat of lime and hemp. I didn’t use any gypsum plaster.*

The basic idea

The basic idea was … to coat the inside of my cast hempcrete walls with some more lime and hemp. It needed be fairly smooth as it would become the finished interior surface. When the lime and hemp had dried, it would be limewashed i.e. painted with watered down lime. The resulting wall should have a soft, matt white finish.

I didn’t want this lime/hemp layer to be too thick. In fact, the thinner the better. The reason being was that I didn’t want to surface of the wall to protrude too much from the electrical socket and light switch boxes. Luckily, I wasn’t too bothered about the flatness of the walls, so I could taper the finish coat in towards the electrical boxes if need be. I was aiming for a thickness of 1 or 2 cm.

You might wonder “why plaster with lime and hemp, instead of normal plaster?” There are various reasons for this. Firstly, I had lots of leftover lime and hemp that could be used. Secondly, I quite liked the texture of lime and hemp when smoothed out. Thirdly, normal gypsum plaster is not as breathable as lime and might compromise the performance of the wall. Lastly, limewash is not recommended for use on gypsum plaster.

I could have plastered with lime and sand but that wouldn’t have given the look that I wanted and besides, I still had a big pile of hemp that I wanted to use up! The fibrous nature of lime and hemp plaster is also said to make it more resistant to cracking. In this instance, it didn’t work out quite like that, as we will see later.

Previous experience

I had, some years previously, used lime/hemp to apply a thick insulating layer over a stone wall. I had also used it over brick both during a training course and on a real building site.

At that time, I used a lime called Tradical combined with course grade hemp. While there was certainly a knack to applying it, it was possible to get quite a nice, smooth finish (with the odd hemp fibre sticking out). It couldn’t be trowelled flat like gypsum plaster but that didn’t really matter. The image above (from the Tradical website) gives some idea what it looked like after application.

Getting it on the walls

OK, getting back up to the tiny house, I had just over a week to do all of the plastering. It was during the October school holidays and the completed hempcrete walls were completely dry (they had been done in August).

One complication, as I mentioned in a previous post, was that I hadn’t had time to cast hempcrete around the gable areas of the walls. So these ’triangular bits’ had to be filled in somehow before the lime/hemp plaster could be applied.

I solved this problem by using a lath and plaster method. Strips of scrap wood were fixed horizontally to the studs and hempcrete was forced in between. The finished hempcrete didn’t have the normal thickness but it was the best I could do, given the limited time I had.

With the gables filled in, it was nearly time to start applying the lime and hemp plaster. Before doing so, I dampened the walls with water from a garden sprayer. This was to avoid the dry walls sucking moisture from the new plaster too quickly.

I used the mix ratio suggested by St. Astier. This worked out as 1/2 sack of lime : 25 Litres of hemp : 15 litres of water.

The resulting mix was fairly sticky i.e. it adhered well enough to the wall with no problems of slumping or falling off.

It was, however, not a great deal of fun to apply. It just didn’t flow very well and it needed a fair bit of muscle to get the walls covered. The fact that I was using course hemp for a thin coat probably didn’t help.

Not being an avid fan of cement grey, I can’t say that the walls looked great at this stage. But at least the patchwork of cast hempcrete began to disappear under something more homogenous. I’m fairly sure I only applied one coat, but it was nearly 2 years ago, so my memory may be playing tricks on me.

In terms of tools, I used a normal plasterer’s finishing trowel (stainless steel) to apply the lime/hemp plaster. It’s also good to have a variety of plastic floats in order to help smooth the plaster as it dries. I suggest one large and one small rectangular float as well as one with a pointed shape These can be used in a circular motion on the wall with progressively more force as the plaster becomes firmer. Small amounts of lime/hemp, placed of a float, can be rubbed onto the wall to fill in holes or low spots. The pointed float is handy for getting into corners and for ‘slicing off’ high spots of material.

The photo below shows the trowel and floats that I used. They are resting on my home made ‘hop up’ which I used to reach the tops of the walls. Making one of these is highly recommended as it can be exactly the height required and is much more convenient than a stepladder. I made mine from leftover timber and OSB shuttering. A smaller hop up can be seen to the left.

I got the plastering finished within the week and had to head home to Scotland. Ideally, I would have been able to leave the windows open to make sure the house was properly ventilated. Instead, I had to shut all the windows and shutters and leave the house until the following summer.

When I returned, the plaster had dried but it was obvious that the house had been quite damp inside due to the water released from the drying plaster. Some clothes that I had hung up were mouldy. I had hoped that the breathing construction of the building might have allowed enough water vapour to pass through. Clearly this had not been the case.

Limewash

By this stage it had been 12 months since the walls were built and 8 months since they were plastered. My first job was to limewash the walls before I could get on with anything else.

To make the limewash I mixed lime putty with water, using a paint whisk with a cordless drill. Lime putty is made from very pure lime that sets in contact with the air. As such, it comes in a tub and is covered with water. I got my lime putty in France from a supplier of eco building materials. You can get it from specialists in the UK too, such as Mike Wye.

Limewash is mixed to a thin consistency. One French artisan on-line suggested it should be like “waffle batter”. More helpfully, he said the ratio is 1 part lime putty to 2 parts water. I just mixed it using my judgement as to what seemed ok.

Then I applied it to the hemp plaster using a wide brush. Being so thin, the limewash was quite ‘splashy’. When it goes on at first it is transparent but gradually becomes white as it dries. Several thin coats are required and it’s recommended to leave 24 hours between coats. I found an excellent guide to using limewash on this UK site.

In fact it took 4 coats to get the walls looking white (or white enough). During this time, I ran out of lime putty. The supplier was a long drive away and I didn’t want to waste precious days waiting on a delivery, so I switched to using powdered lime instead, which I could get locally.

The lime I used was called in French “chaux aérienne” (air lime). In the UK, it’s called non-hydraulic lime. The sacks are commonly marked CL90. It’s the same stuff as the putty, but a bit less convenient to use. The mix ratio is similar. One online source suggests a mix of 1 part powdered lime to 2.5 parts water.

Results

I was happy with the finished walls. I wasn’t looking for perfection (and certainly didn’t get it!) but the result was in keeping with the rustic vibe of the place. The walls were not so wonky that I couldn’t hang a shelf or a cabinet. There were smooth bits and less smooth bits and the whitewashed finish came out just as I had hoped.

The only problem was the appearance of several cracks in the days following the application of the whitewash. Some were tiny and some were larger, perhaps 5mm wide. They went deeper than just the whitewash.

I was puzzled as to what had caused them to appear. Perhaps it was because the house was now exposed to the hot summer air flowing through it and it had caused a final drying (and shrinkage?) of the lime/hemp plaster.

The cracks were more annoying than worrying. The timber frame was doing the structural work and I’m sure the cracks did not go all the way through the cast hempcrete walls. In the end, I filled most of the cracks in with decorator’s filler, which seemed to do the trick.

Lessons learned

If had to do it again I would use different types of lime and hemp for the plaster coat.

Regarding the lime, I would tend towards a weakly hydraulic lime (NHL 2). St Astier do one called Téréchaux, which they say can be used with hemp for finish coats. This should be relatively pleasant to apply and possibly easier to smooth out. It would also look nicer and, being lighter in colour, cover better with limewash.

The lime which I did use (Batichanvre) was really more suited to cast hempcrete and thicker insulating layers. A plaster layer is more decorative and doesn’t need the same mechanical strength.

As for the hemp, I would use the fine grade for a hemp plaster top coat . I’m sure this would make it much easier to get a nice finish based on a thickness of 1-2cm.

As it was, I had loads of ‘less than ideal’ lime and hemp left over, so I used that and made it work. The final result was ok but getting there was harder than it needed to be.

In this post, I’ll describe the steps I took to install untreated, Douglas Fir board and batten vertical cladding on my tiny house.

It seems that there are number of terms that could be used instead of cladding, such as weatherboarding or siding. For the sake of simplicity I’ll stick to cladding.

Step 1: Choose the cladding style and material

I wanted the tiny house cladding to resemble that of a typical tobacco drying barn that can still be seen dotted around the Dordogne valley.

(Although the industry is in decline, there were still nearly 500 tonnes of tobacco grown here in 2019)

If I could achieve something similar to this, then the tiny house would be fairly discrete and look ‘at home’ among the local agricultural architecture.

Evidently, the cladding would have to be vertical. I like the slightly irregular appearance of batten on board cladding (as opposed to machined tongue in groove boards). From what I had read, this is also a pretty good system in terms of shedding rainwater.

I also wanted the wood to age and darken naturally, as these barns have done. Furthermore, I didn’t want the hassle of treating the timber with preservative every few years. I already knew that Douglas Fir was a moderately durable softwood and could be used for cladding without treatment. What’s more, it weathers to a nice (in my opinion) silvery grey. Perfect!

I had already decided to use Douglas Fir for the structural timber, so I went with it for the cladding too.

Regarding durability, I learned that completely untreated Douglas Fir has a life expectancy of 10 to 15 years in outdoor use. This is good enough for me and I suspect that the more sheltered walls will last even longer than this.

Step 2: Determine the cladding dimensions

I spent quite a bit of time reading French self-build forums before arriving at what seemed to be sensible dimensions for my cladding.

The cladding is composed of 4 components

1. Vertical battens (70mm x 40mm). These are nailed to the timber frame studs.
2. Horizontal counter battens (70mm x 40mm). These are nailed to the vertical battens.
3. Cladding Boards (140mm x 21mm), These are nailed vertically to the counter battens, leaving approximately 10mm gap in between.
4. ‘Cover-joint’ Battens (70* x 21mm). These are nailed vertically to cover the joints in the boards.

*Note: This size was originally supposed to be 54mm. The sawmill boss advised me to make it larger, to reduce splitting, and I agreed. I can’t remember what they suggested but I’d guess it’s 60-70mm. I’ll check on my next visit.

The sketch below will hopefully help illustrate this system.

Another explanatory sketch, from the top of the wall looking down, is shown below.

*Note: Cover-joint battens is probably not an industry standard term. It makes sense to me though as it is the literal translation of “couvre-joints”, which is the French term I am used to using. Also, I can’t remember why the boards are 21mm and not 20mm thick. It’s not like the sawmill is going to be that accurate anyway.

The chunky battens and counter battens combine to give an air gap of 80mm. I have read that the air gap (1) helps keep the house cool in summer (2) helps water vapour leave the wall and (3) helps the boards dry more quickly after periods of rain.

Step 3: Order the materials

The timber cladding was part of one big order placed with a local sawmill. This covered all the timber for the tiny house (except the OSB, Beech worktop and Chestnut floor). I placed the order while still in Scotland and paid a deposit. This gave the sawmill plenty of time to get it ready before I arrived in France. When I got to the site, they delivered it all and stacked it exactly where I wanted it. It was a good service for a fair price and I was happy to support a local business.

The rainscreen and anti-rodent grill were ordered from a local builder’s merchant. Again, this was part of a larger order (which included the lime, hemp and OSB for the floor and shuttering). It was great to have the stuff turn up in one big delivery and the driver was able to drop it off really close to the house (he drove the truck over the adjoining field!).

Step 4: Staple on the rainscreen

The breathable fabric rainscreen should help protect the hempcrete wall from water ingress, in the event that some gets past the timber cladding. It was stapled onto the timber studs and overlapped so that rain can’t run down inside the fabric.

Step 5: Nail on the battens and counter battens

This was simple enough. It was easy to find the studs by feeling through the rainscreen and then put a few nails in. However, this was the moment to remember the “anti-rodent grill”! I don’t know what the real name for this is in English. I have literally translated the French term “grille anti-rongeurs”. It is the perforated, right angled metal strip shown in the picture below. The purpose is to help prevent mice and similar creatures from getting into the cosy gap between the glassing and the wall.

The anti-rodent grill came as a coil of flat strip and is bent into the right angled shape when installing. You can see that it is fitted between the vertical and horizontal battens. It should be installed at both the top and bottom of the wall.

Step 6: Nail on the boards and top battens

In the photo below you can see Rob nailing on the cladding boards. Note that he’s putting the nail roughly in line with the centre of the board. The nails should be in a single vertical line. Using 2 nails, side by side, increases the chance of the wood splitting as it expands and contracts.

The boards were spaced roughly 10mm apart. In practice it was impossible to get a consistent spacing as the boards were not dead straight. Sometime there was no gap, other times it was bigger than 10mm.

In the foreground you can see the temporary stop that was set up and attached to the chop saw. This allowed all the boards for this wall to be cut to the same length without measuring every time.

A jigsaw was used to cut the tops of the boards to fit around the rafter ends.

The ‘cover joint’ battens were then nailed in place. The theory is that the nail should pass through the gap in the boards underneath. In practice, this is practically impossible. So, we made our best guess, but we probably caught the edges of more than a few boards.

Unfortunately the cover joint battens would sometimes split as the nail was fired in by the gun. If you have the time, it might be worth nailing these battens on by hand. I did so on one wall and it helped a great deal. The old trick of blunting the end of the nail slightly also helps avoid splits. If you really wanted to be super neat (and you had the time) you could drill a pilot hole before nailing. This might be taking it a bit too far though…

The final result is shown below, complete with DIY zinc window sill protection.

Further Information

Cost

The total materials cost of the timber cladding for was around 1600 Euros. This includes the rainscreen, the anti-rodent grill and all timber.

This works out as 45 Euros per square metre.

Interestingly, this website says that Douglas Fir cladding normally costs 30-45 Euros/m2 (materials only) and 60-140 Euros/m2 (including installation).

So my materials costs are seemingly within the normal range. I did, however, employ ‘magic’ Rob the carpenter for a few days. Including his labour (400 Euros) brings the total to 56 Euros/m2. Not bad.

I should note that I had enough left over cladding to build a tool chest. Waste not, want not!

Weathering

It’s now about 2 years since the cladding was fitted. The photos below shows how it has weathered in that time.

It’s interesting to note how the walls have weathered differently. The gable end receives the worst of the wind and rain (which is unfortunate, as that is where the door is!) and is much darker than the more sheltered side.

I’m sure the weathered look will not be to everyone’s taste. It’s certainly not a ‘bright and clean’ style but it is what I wanted. As it ages, I think the tiny house is already starting to look like it belongs there.