Tactical retreat in the face of overwhelming force

Well, we're no longer in the Summer Palace. Of our twenty days there, it rained on nineteen and blew a gale on five. Last night, the BBC was forecasting gusts of 82 miles per hour - literally hurricane force.

By dawn it was obvious that I couldn't really depend on the roof surviving; even if it did, sheeting rain was falling and fine spray was blowing through the Summer Palace, making everything wet. The shipping forecast was more soberly predicting force ten, and the BBC's 6am domestic forecast had dropped it's prediction to only 70 mph. If I'd just been myself I'd probably have tried to hold out, but the idea of trying to catch the cats after the roof had gone didn't appeal, and I decided to abandon ship while I could. I started to make things as secure as I could.

Neadless to say I didn't have the cat's transport box down at the summer palace. Ivan, who'd slept cuddled in with me all night, was still under the downie, so I emptied the cooking box and unceremoniously bundled him into it. I wheelbarrowed him over the hill to the farm; at the top I could barely stand.

With Ivan in the car I took the transport box back. I was wearing my waterproof sailing jacket with the hood pulled right down as tight as it would go; I'd deliberately left my glasses off for fear they blow away. The rain on my face was like buckshot. Needless to say, the cat  box blew clean out of the barrow.

I'd expected Penny to be hard to catch, but she'd been sleeping in her usual place in the structure of the roof, which was by now moving quite a lot, and when I called to her she climbed up through the roof structure to where she could see me. She was distressed and clearly couldn't get down by herself, so I hauled her out and bundled her into the box. Then I took down the one wall I have at the palace because it was acting as a sail and clearly making matters worse, bundled my bedding into plastic bags and hauled all my furniture into the centre of the platform, covering it with a strongly lashed tarpaulin before leaving. The roof may go but I don't think the platform will.

We're staying with friends for a couple of days while the weather abates and I get stuff dry, but then we'll go back. But after this a more permanent structure becomes more urgent - it won't wait for autumn, or for planning permission.

The Summer Palace

Tonight is our third night in the summer palace. The experience is throwing up problems I didn't expect, as well as ones I did. I have an urgent need to find somewhere for a midden for food waste - far enough away that it doesn't attract rats to the summer palace (although the cats would deal with them), near enough that it's practical to use.

Clothes don't dry in the wood - there isn't enough air movement and there isn't enough sunlight. So I'm going to have to put a clothesline out in the meadow somewhere. That also has implications for my living economy: if clothes do not dry, I must be very careful about getting wet. Fortunately, I haven't yet had a problem with rain blowing into the palace - despite two very wet windy nights; so I think I'm probably OK there.

Although I do love the palace's airiness and sense of openness to the elements, I think I will have to make it solid walls sooner or later. Just sitting, it's pretty cold. Of course, I don't yet have a chimney for my wood stove, so I can't yet use that; but even if I could, with no walls the warmth would just blow away. I'm as concerned about the cats being uncomfortable as myself - if they don't like it here, they could just leave me. And I'd hate that.

Fortunately they're settling reasonably well. Ivan is a bit clingy, and they both follow me about wherever I go in the wood (but, interestingly, not out in the pasture). But they seem quite calm and spend a lot of time playing together among the trees.

We've had our first visitors. Alice and Meg came over just to see the place, and Finn came to borrow my wheelbarrow, and stayed for a coffee.

Unexpected plusses, I think we have badgers in the wood. We've found a couple of gaps in the bottom of the fence, and the cats have been very interested in the smell of them. By one of them there was a large and well-clawed fresh paw print. I've not yet found evidence of a set though, and am surprised they don't find the wood too wet.

I've been cooking on my little camping gas stove, and am pleasantly surprised how well that's going; I'm having appetising meals from fresh ingredients with little compromise on what I would be eating if I had a more conventional cooker. Of course, I have to cook quick things to economise on gas, so no broths... but I will have the woodstove going soon.

Interestingly my nearest neighbour is also someone who is in the formal sense of the term homeless - he is living in a small caravan in the neighbouring landowner's woods. I've always known there was more rural homelessness than most people guess, but I'm starting to realise that there's far more than even I had guessed. Of course, I am in a sense voluntarily homeless, and hope it will be short term. But if I don't get planning permission to build a legal house, creating a comfortable winter palace in the woods does not feel impossible.

Batteries for both phone and computer are limited resources, so my Internet use is far less than I'm used to, but so far I'm not missing it much.

So all in all so far so good. I'll post occasional updates as time goes on.

Going straight

The singlespace roof has a slight twist, and I love it. The inner triangle is three and three quarter degrees off square from the middle hexagon, which again is three and three quarter degrees off square from the outer ring. It's that subtle twist that makes the roof so uneuropean, so quirky.

There's a reason, of course. The reason is that I couldn't get rafters long enough to span the ten metre diameter internal space that I wanted; and I didn't want to have to make a very complex joint at the top of each pillar. But I've been spending the last week working very hard on working out how to make my dwelling simpler to build, lower carbon and, ideally, cheaper; and one of the questions I've asked myself is how big a single space could I build with the rafters I can get.

The answer is that I can get 4800mm rafters at 200 x 50mm cheaply - just as cheap per metre run as 3600mm rafters. Given that the rafters cannot go right to the peak of the roof and that the gradient is shallow, two 4800mm rafters will actually span almost ten metres. But that's the full span of the roof. The walls come inside that span. If I'm going to use straw bale - which I'm now thinking of very seriously - each wall is 600mm thick, and allowing 150mm for eaves that takes 1500mm - or 15% - off the inner diameter, and consequently off the floor space; down from 78.5 square metres to 56.75 square metres.

It means, sadly, that I lose the twist. In a sense it doesn't matter; if I'm insulating the roof with sheeps wool - which again I'm now considering very seriously - there needs to be an inner skin on the roof, a ceiling, to hold the wool up. Which means the rafters are hidden, so the twist would show only in the alignment of the pillars - a bit subtle. But this process of refining the design has been one of losing one little elegant touch after another, and I mourn the passing of each and every one.

You'd think that losing 15% off the internal volume would also be a pretty big issue. Surprisingly, it doesn't seem to be. The picture shows exactly the same furniture (at exactly the same size) as in all the other 'furnished' drawings, and it all fits with adequate circulation space. The two sofas are no longer at right angles to one another - but they are still fully two metres long. The office area is now quite cozy and definitely wouldn't be comfortable for two people. And there's some awkwardness between the kitchen counter and the dining area which needs a little more thought.

Fifteen percent off the diameter is actually slightly more than 15% off the cost, because the longer rafters are more efficient of timber. And the smaller volume will be warmer for the same heat input. All this may be academic; it now seems likely that we'll be able to mill our own timber on site by the time I'm ready to build. And reducing what was already a small dwelling needs serious thought. But... simpler is quicker, and already next winter is snapping at my heels.

Wool gathering

Singlespace, as described in the essays up until this one, is a building with a concrete floor and wall, and extruded polystyrene as insulation. Both of these materials involve a lot of embodied energy, and hence are far from carbon neutral. They're also not local - they don't occur naturally on site, but need to be transported in. I need to use these materials because I've embedded the house into the hillside, and it's a damp hillside.

But, as Pete pointed out, I don't need to earth-shelter the walls. I could have a walkway round the back of the house. Then it could be drained much more conventionally, and the wall wouldn't have to resist the pressure of either earth or water. So the wall could be much lighter. If there were problems with it, access to repair it would be easy.

More significantly, it could be timber - and timber does occur on site (although if I plan to build this year I won't have seasoned timber of my own and will have to use 'imported' timber). Even 'imported' timber has a far lower energy cost than concrete.

But even more radically, if I use a timber wall with air outside it, I don't need the polystyrene insulation whose primary virtue is it works when buried in damp earth. Instead, I can stuff the wall, between two skins of timber, with wool. Ordinary sheep wool.

Now that makes a huge difference. We now have fifty ewes (and a hundred lambs) grazing the farm. We have wool. We don't have enough wool - my estimate is I'd need about 400 fleeces - but Galloway produces masses of wool, and under current market conditions it has little value. Farmers burn fleeces to get rid of them. I can get enough fleeces for little cost.

There are problems with this solution. Concrete is such amazingly cheap stuff that, even though the polystyrene insulation is enormously more expensive than wool, using a wooden floor and wooden walls adds at least 10% to the total cost of the structure. Second, there are problems with invertebrate infestation in wool insulation - there are solutions to this, but I need to research them. Third, extruded polystyrene, as well as providing insulation, protects the waterproof membrane from mechanical damage. Without it, something else would be needed - but old carpet would do.

But finally and most seriously, there is the problem of condensation in the roof. Extruded polystyrene can be put outside the waterproof membrane; it works when wet. Wool insulation doesn't, so it must be inside the membrane. This, too, can be worked around. If there's an airgap - 50mm would do - between the insulation and the waterproof membrane, with a controlled draught, it should clear condensation. But this makes the roof much more complicated to build.

You'll gather from the tone of this essay that I'm not yet persuaded of this solution. I love the idea of an almost zero carbon dwelling. I love the idea of sourcing my materials in my native landscape. Romantically and ecologically, it is undoubtedly the right thing to do.

But two things make me cautious. The first is cost. I really am very short of money to build anyway. Adding 10% really is a big deal. It means I'm going to have to cut corners in other areas. But the other is reliability. Provided the membrane is not breached, the concrete structure will need no maintenance in my lifetime. Of course, if the membrane is breached, then that's a disaster, but it's a reasonably unlikely disaster. By contrast the wooden structure will likely - almost inevitably, in fact - require regular maintenance as I age. The maintenance will be relatively easy to do, but it will need to be done.