Annexe
The plans called for an annexe, a kind of large bed-sit at the western end of the house, and this was to be the starting point. It had a deep cellar, mainly to practice the techniques needed for the larger cellars in the main house, and so an oval hole 6m deep was dug, and 1500 tonnes of broken rock carted off site. Because of the poor access, this was done, at some length, in about 400 small truck loads, and took a week of eleven hour days to excavate the hole removing around 1500 tonnes of rock to a location provided by a helpful local farmer (who now has driveways and hard standings to rival any posh city folk).
At that time, all excited by the size of it, we used to spend long moments standing and gazing upon it. Home. The general comment about our project during these first years was “Why do you want to live in a hole”? What could they mean?
The Malverns are seriously old hills. You can tell this because the rock is not at all supple. It is in fact pretty hard, but luckily brittle and full of fractures, and it is generally possible to break it up with sturdy excavating kit. It also drains better than I expected. The rock walls were deemed self supporting, which was a good thing (a bit of judicious deeming here and there can make a big difference in construction). Nevertheless, a thick wall seemed like a good thing.
The STRUCTURE BELOW GROUND consisted of a 100 mm lightweight block skin, with 100mm of extruded polystyrene outside. The gap between the rock face and the insulation was backfilled with no fines concrete (gravel and cement paste), to form a self supporting fill so as not to load up the wall, which would also allow free drainage to an outside sump cut in the rock wall (if you try this at home, you will learn after a couple of times not to backfill too deep at once. After you get over the wall falling on you, you have the fun of shovelling up several tons of concrete mixed with broken blocks before it sets. It is only self supporting after it sets).
The block wall was rendered, and tanked with torch on modified bituminous membrane. The first lift of this, up to ground level, was one of the few jobs contracted out to professionals. It leaked. Inside this was a 440mm wall built from lightweight (29 kg each!) foundation blocks, laid end on 100mm from the tanking, the gap being backfilled with mortar, and the occasional ventilation pipe. At the same time, a spiral staircase was cast into the wall, one step at a time, using a steel mould fabricated for the purpose. At ground floor/cellar ceiling level, a steel supported block and beam floor was installed.
Around this time, the weather deteriorated and I decided to build a temporary roof over the annexe area. It seemed like a good idea, but it took a long time to build out of hop poles and lorry curtains (passing aircraft must have puzzled for a year or two as to why there was a large advert for the Aston Brewery high up in the Malvern Hills). A removable section was made to allow building materials to be lowered in. As it turned out, the temporary roof proved very useful, allowing work to continue through some pretty horrible weather. It also confounded the predictions of some sceptics who were convinced it would blow away during gales. Hop wire stays proved, to borrow a phrase, efficacious in every way. It was quite sad when the time came to dismantle it.
As the walls went up, the mezzanine bedroom floor went in, with similar construction to the ground floor. Now out of the ground, the insulation had to be bolted to the 100mm skin prior to rendering. In keeping with the free form shapes of the building, the doorways are sort of oval arches, narrower at the ground than at shoulder height (there must be a name for this). It is difficult to fit doors to a doorway like this (more correctly, it is difficult to fit a door that opens), and I had in mind a vaguely formed plan to make some tricky hinges so that the doors moved directly forward a short distance before swinging. However, idly flicking through a book about Gaudi's architecture one day I noticed that he used, in similarly shaped doorways, timber infill to allow a straight section between the hinges. So I may well do that. A bodge, but it is only an annexe, after all.
The walls went on up, and quite a serious ROOF went on. Designed to be capable of carrying over 1000mm of earth over an unsupported 9m span, it is basically a beefed up version of the internal floors. The reinforced concrete beams sit on the lower flanges of the steel supporting beams, to avoid deep internal intrusions (this caused a lot of grief for the concrete beam manufacturer, who doubted their ability to maintain sufficiently accurate tolerances for beam length. In the event they did an excellent job) Over the beam and blocks is a 300mm layer of foundation blocks (which are good thermal insulators) over which is 100mm of extruded polystyrene. The blocks are mainly there to bring the level up to the top of the steel beams, avoiding high external intrusions (extrusions?). I doubt that many roofs are made from foundation blocks. Over this is 100mm of reinforced concrete, which is tanked with the same torch on membrane. Over that (call that a roof?) is 75mm fibre reinforced screed, covered with drainage sheet, which should keep the screed reasonably dry while also providing a water reservoir for the green roof. A large roof light protrudes through the roof, which will eventually house the observatory (having a triple glazed glass floor to allow light into the rest of the building, but to cut down on convection currents which would upset observations). Outside, part of the building is to be covered with, er, hill, and the part which is exposed is faced with rock saved from the excavations. However, first the whole building was skinned in concrete blocks, to keep the insulation snug, and to avoid worry about the stability of the rock wall. The parts to be faced with rock had a 225mm block wall, sloping 300mm out from top to bottom (although not built in that order), and the rest was protected by a vertical 150mm wall.