Wednesday 17 August 2011


When someone approaches my building, the first thing - and the dominant thing - they're likely to see is the fenestration. That's because pretty much everything else is buried in the hillside; the only face of the structure that's exposed is the south face, which is primarily glass. My intention has always been to use commercial patio door units, because they're mass produced they're far cheaper than any custom unit I can make myself, while having well engineered latches and reasonable insulation.

In the 'sousterrain' design, the shape of the fenestration was to some extent masked by the eyebrow lintels and flying butresses. All the swoopy curves of the exposed concrete structure relieve the industrial regularity of the windows. In my first 'singlespace' drawings, I set the window units back behind the pillars, and thus behind the ring-beam and its braces, and this too masks the shape a bit.

But in reconsidering the front joinery I've considered first that the elements of the frontage are extremely visible, and second that they're exposed to weather. While making the rest of the structure in softwood seems acceptable to me, using oak for the front pillars and lintels will both reduce maintenance and add elegance. Given that it's only five pillars and four lintels, the increased cost is not going to be high. And it you're going to do that, the relatively lightweight 200x50 mm lintels of the front ring beam as originally planned are going to look wrong (and in any case require a lot of additional joinery to make a weatherproof space). Better, I thought, to make them true lintels, with a section of 200x200 mm or possibly even more. Yes, more expensive and not required from an engineering point of view, but the fenestration can now be fitted directly into the aperture between pillars, sill and lintels, completely solving the weatherproofing problem.

The trouble with that is, if the pillars are still roundwood, the only place the windows can now go is between the centres of the pillars, or else you're back to bodging the weatherproofing. Which means you can't easily get well engineered braces in front of them. Mind you, there's no longer an engineering justification for braces - all they're doing is relieving the shape of the aperture. If the windows are to be set futher back into the aperture, the pillars must be sawn wood - but they can be sawn to the angles I want, so that can be a feature. And I can have braces.

And the windows themselves?

OK, let's start with the ideal. If I were a plutocrat James Bond villain, the windows would drop vertically into wells in the floor to open them, and would be closed by hydraulic lifts or something similar. In summer, when the weather is warm and pleasant, I'd like them just not to be there at all. If not a plutocrat but merely very rich, I would have a custom made fold-and-slide door, where there was one continuous track and all the leaves of the left-hand two apertures folded and slid left while all those on the right-hand two folded and slid right. That would achieve almost the same end result - in summer, a clear space open to the outside - although the folded leaves would now intrude into the living space on either side of the single opening.

The whole point of this design, however, is that I'm neither a plutocrat nor rich. I'm poor. As I said to the woman from the planning department, the whole purpose of this proposal is cheap hooses for puir fowk. A single two metre by two metre fold and slide unit in oak veneer - not even real oak - costs of the order of £1,200 ex VAT, so the four of them cost £4,800 ex VAT. Solid oak costs more. That's in a structure which, even with my oak front joinery, has a materials cost for everything else except the front windows of under £7,000.

So, here's the rub. I can halve that cost by using uPVC units. uPVC units come in at £600 per unit, or £2400 for the whole set. But can I really live with uPVC?

Let's go back to the matter of doing things on the cheap. Makers of sealed glazing units often sell units that have been made but which weren't used in the project for which they were made, cheap. This is because once made they can't be refashioned, and if they aren't a standard size then they're pretty much scrap. So you can get cheap units but you're unlikely to get the size you want. Poor self builders often start by buying these cheap units and then making their window apertures to fit the units they've got. I could, and probably should, do this; if I can get units which are almost the right size it might even be worth respacing the pillars a bit to fit them.

However, that means spending out on the fenestration right at the start of the project. I don't want to do that because if I'm going to run out of cash at any stage I want to run out of cash with a weatherproof structure. Window apertures can be made good cheaply with weatherboard (at the cost of no daylight). A bloody great hole in the roof can't be made good cheaply.

But also, I don't, in the end, know how much I've got to spend. The financing of the farm is complicated and not relevant here, but it's possible that in due course I'll get some money back - possibly as much as £20,000, which makes a difference. If I don't get any of it back, then I'm going to have to see what cheap fenestration I can get. If I get a little, then four uPVC units are not a lot of money. If I get more, then I can have oak. But if I get a lot of it back, a custom fold and slide isn't utterly impossible... and it would make a significant difference to the space.

One question that I haven't answered in all this, and which will have to be answered, is... where will the cat-flap go? Penny and Ivan are going to be seriously unamused if there isn't a cat-flap, and I'm going to be seriously unamused if I have to keep getting up and going to open the door for them.

Plan B

Of course Plan A is to get planning permission. Of course Plan A is to live in a proper, legitimate house, pay council tax and have my rubbish collected in bins. But that may not happen. It, actually, may not happen for three reasons. Firstly, and very probably, planning permission may simply be refused. That is actually pretty likely - I probably have less than 50% chance of getting planning permission anyway.

Secondly, planning permission may be granted but it may be granted too late. I must have something considerably more comfortable and weatherproof than the Summer Palace before winter; neither my mental nor my physical health would stand up to a winter spent literally out of doors. So if I haven't got planning permission by mid-July I have to go to plan B anyway.

Thirdly, planning permission may be granted but I may not be able to afford to build to the quality required. This is complicated. In theory one third of what I invested in the farm is loan, to be repaid. It was never likely that it would all be repaid, but I was hoping that half of it would be. This now looks unlikely. If I had got half back, I would have £16,000 to pay for materials for my house, which may sound ludicrously little but actually as self-builders of houses with planning permission can reclaim VAT it's effectively £20,000 and it is - just, barely - enough. But unless I get some of the loan back I have £6,000, and that just isn't.

Of course I could get a mortgage - I do own the croft - but I'd have to get a steady job to repay it, and I really don't want to do that.

So there has to be a plan B, and plan B is about counting assets.

I have several hundred harvestable sitka trees. I have a growing barley crop, and shall have straw by autumn. There is also wool, but asking around I find that unless you treat wool with some fairly nasty insecticide it doesn't make good household insulation.

However, a log cabin is possible using materials I don't have to pay for. If I build a roundwood pole structure with straw bale walls, I need to pay for the render on the walls... at least, in theory I could make it from raw materials, but that's too complex and I don't have time. But the render is affordable. Unless I thatch the roof - which takes skill I don't have, and longer, better quality straw than I'm likely to get - roof insulation also needs to be paid for. In any case, windows, hardware, electrical installation and drainage need to be paid for. It may be, for the first winter, I have to do without electricity all together. It may be, for the first winter, that I have to do without glass.

But a habitable house for £6,000? Yes, I think it can be done.

Wednesday 10 August 2011

Building plan for the Winter Palace

If the bothy I'm wryly refering to as the Winter Palace is to be built in one weekend - hopefully in one day - a lot of preparation has to be done first. Let's start with the basics: materials.

The winter palace comprises straw bales, timber, render, insulation and a few other things. It would be elegant if the straw bales came from barley crop and, in theory, they might. But I'm not confident that my barley will be ready to harvest in time, so it would be prudent to get bales from another source. Similarly there would be a lot to be said for using timber from my own wood; in terms of energy and the environment it would be the best solution. But I'm not at all confident we can mill enough timber in time and even if we do it will be horribly unseasoned. So, again, it may be prudent to buy the timber from the builders' merchant.

On the matter of render, the choices are clay or lime. I'll need between two and three cubic metres. Clay will cost the labour of digging it and hauling it on site; lime will cost real money. Lime is the preferred option but may simply be unaffordable.

Tasks to do in advance

Level the rails

The bothy will be built on three oak rails lying on a stone foundation. The foundation is in place now, but not perfectly level. Some components of the rails are cut, but most are not. Getting the rails to site will need, probably, Alex's Unimog. The rails will then need to be assembled on site and levelled. That's a job for probably two people over two days, in dry weather.

Cut hazel rods

The bale walls will effectively be nailed together with hazel rods; I need about one metre of rod per bale, or 130 metres of rods. I could actually do that myself in a couple of days and fetch them back in my car, but it would be easier with two people and a trailer; again, this is a dry weather job. (Update: I've already cut 100 metres of rods)

Dig clay

If I can't afford lime render I'm going to have to dig clay. Three cubic metres is a lot of digging. Also, I can't really haul that back in my car, I'll need to borrow a trailer and something to pull it. Two people, one day, and again dry weather would make it enormously more pleasant. On sober reflection, three cubic metres of water is (obviously) three tons, and clay is substantially heavier than water; so that's quite a large trailer and it probably needs to be a tractor!

Prefabricate sub-assemblies

One of the issues with a straw bale structure is that although bales from the same batch - baled with the same baler on the same day with the same tension settings of straw from the same field - are likely to be consistent in length, you don't know until they're delivered exacly what that length will be. The walls sit on a sill plate and are capped with a wall plate, each of which will be essentially a box structure. The box structures can be prefabricated, but need to be constructed to the maximum likely length and designed to be easily shortened on the day.

The roof is supported by four main trusses. Again, the trusses can be prefabricated. The front and rear trusses will probably eventually have windows in them, and it's possible that the frames for these windows can be prefabricated at the same time.

There will probably be a back door and there will certainly be a front opening. Each of these openings needs to be framed so as to support the walls. The frames can be prefabricated.

All this prefabrication can take place in the Void, so can happen on wet days; but it can't happen until I have timber and I'd estimate that it's at least a week's work.

Lay the floor

The floor can't be laid very long before the main build since it won't benefit from getting wet, but it doesn't need to be laid on the very same day as the rest of the build, provided it can be protected by tarpaulins. Laying the floor is neither hard nor especially time consuming; two people could easily do it in one day and I could easily do it by myself in two.

All that adds up to a fortnight of my time and a few days of other people's time, over a period of six weeks; it shouldn't be impossible.

On the day

On the day I propose to have four teams, Carpentry, Catering, Transport and Walls.


The job of the Transport team is obviously to make sure that everything is on site when needed. If we have a spell of settled weather this isn't an issue since everything can be stacked on site in advance. Render can be stacked on site in advance anyway. But otherwise I'll need a tractor or 4x4 with trailer, and one or two people to help load/unload. Until the bale walls are built the Carpentry team can help Transport.


The job of the catering team is to feed everyone.


Walls team starts the day by assembling the bale walls onto the sill plates, which should already be in place, and staking the bales with the hazel rods. Apart from the front opening and the rear door there are no holes or other openings in the walls, so this should be very fast - an hour at most. Obviously it's important that they be vertical!

Once the walls are up and the wall plates have been fitted, walls team's job is to apply render. About 2.5cm of render has to be applied to both inside and outside of the walls, all over. However, as rendering is going on the carpentry team will be working on the roof, so some co-operation will be needed, and generally the carpentry team's needs will take priority.


The Carpentry team don't start work until the bale walls are up. Until then, they can help transport team bringing materials to site. Once the walls are up, they install and fit first the wall plates, then the main trusses and their cross-braces, then the inner vapour barrier, then the purlins, then the insulation and the outer vapour barrier. At this point the roof is more or less weatherproof and the walls are protected; if this is as far as we can get that's OK.

However, I'm hoping we can get further. If we have enough carpenters to divide into two teams, one team will be set to cladding while the other is set to lining; if not, cladding comes first, lining happens later, and, indeed, need not happen as part of the build weekend.

If there's time it would be nice to dismantle the summer palace, and to use its wood to construct the bed-loft mezzanine. It would also be nice to have the flue fitted.


The object of the weekend is to get main structure up and weatherproof. It isn't intended or expected that to have a completely finished building. Render will have to dry before the interior walls can be painted, and that must be done before furniture can be moved in.

I don't envisage fitting the back door, or panelling or glazing the front opening, as part of the build weekend. This is partly because I will need through draught to dry the clay render, and partly because frankly I haven't a clue yet how I'm going to fill the front opening (although, obviously, as much glazing as possible is essential). The rear gable opening is also intended to be eventually partly glazed, but it may be partly panelled before being erected.

While I shall almost certainly sleep in the new structure on the Sunday night (because the summer palace will very likely no longer exist) I don't expect to be fully moved in until 17th October.

Towards a farming plan for the West Croft

My objective for farming the West Croft is to essentially make as large a contribution to self sufficiency as possible, while minimising cash inputs. I'm not significantly concerned with maximise cash profit, although I can't afford for it to become a major cash drain.

I intent to continue to farm the croft as organic, not so much because I believe in the benefits of organic food but because I want to maximise biodiversity and minimise long-term environmental impact. It's no great advantage to me for the croft to be registered organic, since the difference in profit between certified organic production and uncertified production is unlikely to cover the cost of registration, but if the croft continues to be covered by the Standingstone registration that is in my interests.

These objectives need to be balanced against constraints.

Cropping the croft without fertilising it will lead to impoverishment in nitrates. Also, regular cropping of the same land will lead to a decline in meadow herbage, which is not what I want. Organic fertiliser is expensive, and, from the point of view of a registered organic holding, not permissible as the major source of fertiliser anyway. So I need to maintain pasture under grass/clover with some shit-producing animals as a significant part of what I do. It could easily be all I do...

During years under grass/clover, and particularly with animals, nitrates build up in the soil. During years under crop, the nitrates deplete. Scottish Agricultural College recommends an organic cycle of three years grass/clover, one year cereal, one year roots, one year cereal. This doesn't appeal to me as if I plough up my land three years in succession every six, I'm going to seriously degrade the meadow herbage.

At the same time modern tractors and combines need a lot of turning room. Therefore, economically, I ought to treat my whole croft as one field. But if I do that there's no reservoir on my land for herbage species to recolonise from (there is of course the commons). Also, treating the whole croft as one field means that in any one year I have only one sort of produce, so I'm forced to trade most of it in the cash economy since I'll have more of that product than I need, and will be short of other products which I also need.

Three park strategy

These considerations make me feel that there's merit in a three park plan: to divide my current 2.9 hectares of arable into three roughly equal parcels each of slightly less than one hectare each, with more-or-less permanent fencing between. This means that in any one year I can have all three down to grass/clover, two down to grass/clover and one down to crop, or one down to grass/clover and two down to crop.

If two adjacent parks were down to the same crop in the same year it would be good to be able to remove the fencing between them, but this may be more work than it's worth. If I'm going to have crops, I need to be able to use a tractor from time to time, and given the park sizes it needs to be a small one. It could be mine, or communal, or shared, or contracted in (although contractors rarely have small tractors).


I like to eat meat; not a lot of it, but some. Keeping hardy livestock involves little work and helps fertilise the land. Livestock slaughtered and butchered for home consumption require remarkably little bureaucracy and hence, cost. Cattle, sheep and pigs are all possible. However, pigs are particularly hard to fence, and sheep (especially lambs) are fairly hard to fence.

Regardless of species, I could buy weaners in the spring, fatten through the summer and slaughter in the back end, or I could keep my own females and breed from them to produce animals to slaughter.

For cattle a stocking density of two cows (including suckling calves) per hectare is often recommended, with up to three stirks per hectare. Galloway cattle are particularly hardy and low maintenance, and they are, after all, bred for precisely our conditions. A viable strategy would be to have two parks as pasture at any one time; to have two cows, serviced by AI each year, and run the calves with their mothers for twenty-four months until slaughter. This would produce an average of two beef carcases per year at around 300Kg live weight, each yielding about 200Kg meat. That's more than twice the meat I need for own use, so one animal could be sold each year.

For shelter, access to part of the wood in winter is probably all Galloways would need, but a small cattle shed would be a good thing.

For sheep, stocking density is about six ewes per hectare, each producing on average two lambs per year, so probably eleven carcases per year at about 30Kg each live weight. This means only one park would need to be devoted to livestock in any one year. The advantage over cattle is more slaughters per year means less meat in storage at any one time, and meat is eaten fresher. Also, it would be possible from an economic and husbandry point of view not to produce much excess. The downside is that slaughter is inevitably a pretty grisly and horrible job, but doing it more often would get one accustomed, I suppose. If I kept just one park of sheep I could put one park down to hay each year, which could be a cash crop.

Sheep wouldn't strictly need shelter and definitely shouldn't be allowed into the wood, but a small shed would nevertheless be useful for lambing.


The SAC states that the average yield for oats in Scotland is 5.92 tons per hectare, but that's assuming conventional production; if I half that to around three tons/park (given I have organic production and that my parks are shy of one hectare) that's a reasonable target. Thus if I grow one park of oats every two years, that gives me

250Kg own use for two years (I currently consume about 0.5Kg of oats a week, but if I baked oaten bread or oatcakes could easily increase that)
175Kg reserved as seed for next planting
2500Kg available as cattle feed or to sell

To have oats in my rotation requires that I have a granary capable of holding at least 3 tons of oats (5 tons capacity would be better given that I may be being overly pessimistic on yield), and a means of hulling and rolling small quantities of oats at a time.

As to root crops, the obvious one is potatoes. The trouble is I don't actually much like potatoes. Also, mechanically harvesting one hectare of potatoes is almost certainly uneconomic, and manually harvesting a hectare of potatoes is a lot of work. Onions seem to be doing reasonably well for me this year despite neglect; the parsnips I planted this year seem entirely to have succumbed to slugs. If I add roots to my rotation, it seems to me best to divide a park into strips of early, main-crop and late potatoes, onions, and possibly brassica roots such as swedes or parsnips. This means that they won't all come ready for harvest at the same time, spreading the labour of manual harvest over several months. It would be possible to have strips of green brassicas (cabbage, chard, etc), courgettes and peas in the same park since potatoes and brassica root crops will need regular tending for slugs and weeds anyway, so the additional work would be low.

You can't reasonably store root crops over two years, and I can't afford to devote one park a year to roots, so I'd need to sell surplus production in producing years and buy in non-producing years - possibly in an arrangement with another Standingstone croft.


At present my feeling is that my best plan would be to divide my croft into three parks and to seek to buy two Galloway heifers in calf next spring. I would then run a rotation with each park doing two years pasture, one year oats, one year mixed roots. But I'd be very interested in other people's comments on this plan! This plan assumes that the commons remains permanent meadow (which is the current plan) and thus serves as a seed source and genetic reserve to keep repopulating my parks with meadow herbs ('weeds'); if the commons is not managed as permanent meadow I may need to set one of my parks as permanent meadow, in order to maintain biodiversity. This would mean either that I could not practically keep cattle, or that I could not grow crops.

Creative Commons Licence
The fool on the hill by Simon Brooke is licensed under a Creative Commons Attribution-ShareAlike 3.0 Unported License