The timber frame fires

Building carries a disturbing report of yet another fire in a half-built multi-storey timber frame apartment block. This one took place in Hatfield on Saturday (11/8/07) and Phil Clark has uncovered a short video showing the event. It follows hard on the heels of similar fires in Newcastle (April 07), Willenhall (Mar 07) and Colindale in North London (July 06). YouTube carries video clips of Colindale, Newcastle and now Hatfield.

I don’t know how many timber frame apartment blocks are under construction at any one time in this country, but it’s still a building technique that is in its infancy so I would be surprised if there are more than a few hundred on the go. The superstructure is often only left exposed for a few days — this is one of the rationales behind using timber frame — yet we’ve just had four burn to the ground in a similar manner. And amazingly, no one has yet been hurt, let alone killed.

Yes, it’s deeply worrying. But it’s also beginning to look a little suspicious.

Offsite 2007

Four and a half hours scampering around the BRE car park in Watford today brought on a familiar feeling of having been here before. New housing exhibitions opening, like Offsite 2007, are always accompanied by a feeling of excitement, anticipation, even euphoria, but I come away a little let down and rather bemused. “What was all that about?” I ask myself. It was rather like Offsite 2005 – hardly surprising because it was largely the same people on the same site. But it also reminded me of the Milton Keynes Energy World exhibition in 1986. Twenty years on, what has really changed? On the one hand, you want to congratulate everyone for the phenomenal amount of work undertaken to get everything up and running for the opening event. On the other, you can’t help wondering who this is intended to impress, especially as I understand the public don’t get access to the BRE site.

The star of the show was the Kingspan Lighthouse, the first house in the country to achieve Level 6 of the Code for Sustainable Homes – i.e. it’s zero carbon. It was a peculiar looking affair with lots of chestnut siding wrapped around a tall thin three-storey structure. It didn’t really look like a house, but I think that was the point. In fact, none of the new exhibits looked like houses that we would instantly recognise as houses. Tear up the old, bring in the new. You get the message: this is about re-engineering our lives just as much as re-engineering our homes.

The guys in Stewart Milne’s Sigma showhouse seemed just a little miffed. Their house only got to Level 5 and that thanks to three roof-mounted wind turbines which, of course, weren’t even spinning, let alone producing any electricity. “We’d get a six if we built them in a courtyard setting,” explained their minder. It’s all got very competitive, very quickly.

The Hanson Eco-Oast House made the nicest space, making use of one large room upstairs room, lit and ventilated by the huge vaulted roof. It was the only one with any wow factor, the sort of thing to excite the middle classes. Everyone had balconies, lots of balconies and there was timber in every shade and species you could think of.

So well done everyone who has chipped into this event. But is this the roadmap for tomorrow’s housing? We’ll have to wait and see.

Low energy housing in Sweden, Denmark & Germany

On Friday, I attended the debriefing session run by the DTI Global Watch Mission, which sent a small party to Sweden, Denmark and Germany in November. It was similar, in many ways, to the PassivHaus study tour, which I went on in February, but it was a longer tour looking at a wider range of low energy housing projects.

The general consensus was also pretty similar to that drawn by the PassivHaus study tour. In all three countries, they were looking at the same sort of things: a small number of exemplar projects, all with the emphasis is on pretty much the same features: massive insulation levels, air tightness, triple glazing, mechanical ventilation with heat recovery and a little booster heating to get things up to par. It’s a recurring theme: you’d be tempted to think that maybe they have the solution to the conundrum of how best to construct low energy housing.

However, there are subtle differences between the three national approaches. In no particular order, here are some of the notes and calculations I made, as a result of attending the session.

• 80% of new homes in Sweden are heated with electric heat pumps, of which over 60% are ground source heat pumps. Remember, Sweden has loads of hydro-electricity, so electric heating makes sense for them in ways it doesn’t in other territories. They use to prefer the cheaper air-to-air heat pumps but people found them noisy and the growth in the market is now happening with ground source heat pumps.

• In Denmark, 60% of homes have a supply of hot water pumped into the house from a district heating system.

• The Germans pay 50 cents (33p) per kWh for renewably generated electricity sold on to the grid. No other nation does this. Consequently, German roofs are covered in PV arrays.

• Micro Combined Heat and Power plant (CHP), fired by the Stirling engine, throw off one unit of electricity for every seven units of heat. You end up with far too much heat for optimal use. However, the next generation of fuel cell-based CHP, which should be commercially available by around 2010, should produce roughly equal amounts of heat and electricity. This should address the output issue, but at the moment fuel cell CHP is anything but micro: it requires a dedicated plant room.

• There is yet another low energy standard that was mentioned that I hadn’t come across before. The 3-litre house. This refers to the amount of heating oil required to provide space heating for each square meter of a house each year. 3-litres is reckoned to be the bees knees for renovations. 7-litre houses are something close to the building regs standard in Germany: in contrast, the PassivHaus standard is lower still, probably around 1.5-lts. This fascinating new take on a performance standard sent me scurrying off to my Excel spreadsheet which analyses our home usage. On this basis, I reckon ours is an 8-litre house! For a house built 15 years ago, that’s not too bad but I am not sure I should boast about it.

• The 2006 Part L building regulation for England & Wales indicates a space heating requirement or around 40kWh/m2/annum. I reckon that could be termed as a 4-litre house. Or maybe it’s nearer 5.

• Don’t say it too loudly, but we use another 4 litres per annum per m2 just to heat our domestic hot water. These extra litres are somehow overlooked from the standard.

• The Swedes, the Danes, the Germans and the Brits all seem to design homes with a 60-year lifespan. Quite where this 60-year figure came from, I have no idea. Why not 50? Or 75? How come everyone settled on 60?

• Having said that, only around 15,000 homes are demolished every year in the UK. That is one tiny number when compared to the 180,000 new ones built and the 25 million homes that currently exist. At that rate, it will take 1667 years before we manage to replace all the existing homes we have built, which is precisely 1607 years more than their design life. Does anyone foresee any problems building up here?

• The NHBC reckon that currently 47% of new homes being built in the UK are apartments. How long are they going to have to last?

Spluttering SIPS: Chapter 2

Structural Insulated Panels or SIPS: quick to build, fantastic performance, building system of the future. Or so we hope. But so far the UK experience of SIPS has been checkered, and that is being very polite. There are some very successful SIPS developments, for sure: I visited one in September where everything had gone splendidly. But a good number have run into problems.

Here is an affordable housing scheme currently underway in Bugle in Cornwall. The builders have thoughtfully placed a webcam there so we can monitor progress. Not that there has been a lot. This image was taken off the webcam today (29 November). These 23 apartments have been going up since March and the roof still isn’t on. And the OSB is only just being covered up now. OSB isn’t designed to be left outside to withstand the elements: with SIPS, OSB is an integral part of the structure. Apparently there have been a lot of wranglings going on behind the scenes with building control and warranty providers. It doesn’t look pretty, does it?

My advice remains the same. I don’t want to put people off SIPS but if you do want to go down this route don’t try and do it on the cheap and make sure that your design and erection team know exactly what they are doing. Check their track record and speak to previous clients.

Timber frame wins on airtightness

Housebuilder magazine (Oct 2006) has published a short article by Graeme Owen of BSRIA about how new homes are doing in the airtightness stakes. The tests were carried out before the new requirement for airtightness testing came into effect but nevertheless the results are dramatic.

The pass rate, set out in Part L1A, is 10m3 per hour per m2 of heated envelope at 50 Pascals of pressure, which is such a mouthful that it is now being reduced to 10 q50. Energy wonks don’t regard this as very demanding: the best practice low energy homes score as low as 1.5 q50. Nevertheless, many of our new homes struggled to make it through the new hurdle.

All but 3% of timber frame homes achieved a pass and as many as 73% of timber frame homes tested achieve a score of 5 q50 or less. In contrast, 28% of masonry homes failed the new standard and only 6% met a score of 5 or less.

Why the huge difference? Well there is a clue given in the article. Apparently, the masonry build data was further broken down into houses which were wet-plastered and those which were dry-lined, using plasterboard stuck on with dot and dab. The wet-plastered houses achieved a pass rate comparable with timber frame: the dry-lined homes most certainly didn’t. To build airtight with dry-lined walls, you need to seal around all the edges, something that has very rarely been done till now.


If you want to download BSRIA’s guide to airtight building and the new testing regime, click right
here

Baufritz, the biological housebuilders

Last week I went on a rare expenses paid jolly to Germany. I was guest of Bavarian housebuilder Baufritz who, like many others in Germany, have watched Huf Haus blaze a trail into the UK selfbuild market and would like to do the same themselves.

There are something like a hundred fertighaus (factory house) companies in Germany, in the Huf Haus/Baufritz mould. They typically build a few hundred homes each year, similar in size to what several timber frame housebuilders do over here. The German businesses are also usually timber frame but it’s not timber frame as we know it in the UK. Here, we tend to supply just the timber skeleton, the superstructure, which then has to be finished on site. The Germans, by contrast, assemble most of the house, including internal and external wall finishes, in what is sometimes referred to as a closed-panel system.

To a visitor from the UK, the gleaming efficiency of the production lines is a site to behold. It’s reminiscent of car factories with wall and roof panels moving through assembly lines with a minimum of human intervention. If you want to read up more on it, there is a very funny description on the But She’s a Girl blogsite. Baufritz produce around 250 houses a year with a staff of 260, of which around 80 work in the factory and a further 60 work on site as erection crews. Work it out: it’s roughly one house per person per year. Say 1800 hours input. In the UK, you would expect to see between 3000 and 5000 hours work go into a traditionally-built house of similar size, and not that much less for a timber-frame one.

The reasons for this different approach are several.
• German’s pay high tax on overtime so there is an incentive to increase production from regular working hours.
• Germany hasn’t experienced the housing booms and busts that we have had in the UK. Consequently, German builders are confident enough to invest in manufacturing facilities.
• Unlike in the UK, Germans rarely seem to sell family companies and they tend to invest for the long term. Baufritz is a good example: it is owned and managed by the fourth generation of the Fritz family.
• The basic German house shape is simpler and less variable than its British equivalent. It lends itself to prefabrication.
• Planning permission for homes in rural districts is much easier to obtain in Germany and the selfbuild market is much larger, maybe ten times the size of the UK
• Without wishing to resort to oversimplistic stereotypes, Germans seem to be naturally good at organisation, teamwork and manufacturing. In contrast, Brits are more inventive, more willing to experiment and quicker to embrace change. All these factors combine to explain why prefabrication has taken off in Germany but struggled in the UK, for it’s expensive to change horses when you have invested huge amounts of money on one particular method.

And there are elements of German housebuilding that are surprisingly conservative. Various innovations which have been widely taken up in British housebuilding have seemingly been ignored in Germany. Trussed roofs, engineered timber beams, precast concrete flooring: no sign of these in Germany, as far as I can work out. In fact, most fertighaus builders lay a wet concrete screed within the intermediate floor, just because that’s the way it has always been done, despite the fact that it stops the house erection process stone dead whilst the concrete dries.

Consequently, German housebuilders tend to be looked at as being much of a muchness. Very, very good at what they do, but a rather limited range. Huf Haus stand out because they have gone down a most unusual design route and have decided to build these iconic black wood and glass houses, designer conservatories for the minamalistically inclined. But how does the competion differentiate itself?

Well, Baufritz specialise in what they call the biological approach to homebuilding, as promoted by the IBN (Institut fur Baubiologie) in Neubeuern. If I can summarize just what it is Baufritz’s take on a healthy house is, it seems to be:
• avoid all products from the petro-chemical industry
• use only untreated timber (and lots of it)
• insulate with compressed wood shavings treated with whey (for fireproofing) and soda (as an insect repellent)
• rely on natural ventilation– avoiding mechanical ventilation systems
• shield against electro-magnetic radiation
• use radiant heating systems — both underfloor and in-wall

Their take on mechanical ventilation is certainly unconventional and puts them at odds with the Passive Haus movement, who promote ultra low energy housing in Germany. Not that a Baufritz house is a high energy burner: on the contrary, they seem to be building some of the best insulated homes I have ever seen. The biological builders approach is simply to monitor indoor air quality via a carbon dioxide detector and then to open a window if things look to be a little struffy. They also don’t insist on extractor fans nor trickle vents, two of the banes of UK building regs. A Baufritz home doesn’t suffer from condensation - it’s simply too warm and well insulated.

The other really revealing technological aspect of their homebuilding is their using a specially-prepared plasterboard with a carbon skin which attracts and earths electro magnetic radiation. I have never seen nor heard of such a thing before. They take electrosmog very seriously and insist on getting radiation levels right down to trace levels, and have equipment on hand to prove it. Having written somewhat dismissively about the threat of electrosmog on this blog in June, I must admit that I am beginning to start getting worried about it!

The healthy house is a difficult call from a marketing point of view. Whilst no one would possibly object to living in a healthy house, there is just a slight fear that paying good money for levels of protection not available in normal housing is verging on being paranoid. People may worry about microwave ovens and mobile phones, but not enough to stop using them and so they may well feel that buying a biological home is a bit weird. I asked the heavily-pregnant owner/manager, Dagmar Fritz, pictured here, how many of their sales resulted specifically from customers demanding healthy homes and she admitted the number wasn’t that high. The reason they build biologically is because of a family commitment to the issue and Baufritz have established a unique reputation amongst German factory house builders and are now poised to see what the UK makes of their offerings.

It will be fascinating to find out. Oliver Rehm, a friend of Dagmar’s since student days, and a UK resident has gone into partnership with Baufritz to sell their houses into the British market. They have established an office in Cambridge are currently working through various technical approval issues. They don’t have any brochures or pattern books; everything is bespoke. Prices have yet to be finalised but expect a Baufritz house to cost something similar to a Huf Haus, that is to say around the £1500 per sq metre mark, making a four- bedroom, 2,000 sq m house £300,000. That puts it firmly at the upper end of the cost spectrum, and probably aims it at the prosperous South East corner of the country. Still, I can see them going well with people who admire the quality of German factory housebuilders and yet don’t want something quite as severe as Huf Haus.

House leaves £60k competitors standing

I have selected a new benchmark house for the 7th edition of The Housebuilder's Bible, which is due out in September 2006. It’s the smallest house I have yet worked with and in many ways it’s a back-to-basics structure, a four-square box with very little embellishment, finished to a pretty basic standard. It’s builder, Mike Capp, is working up some of the costings figures for me over the next few weeks but he assures me it cost between £60k and £70k to build.

I’ll be writing at length about this house in the next edition so I don’t want to go into too much detail here. But it makes an interesting case study in itself in the year that the government has sponsored a House for £60k competition. To qualify, the house needs to be at least 76.5m2. My benchmark house is 110m2 internal floor area. It appears, cost-wise, that it would cream it!

How long will my new house last?

It’s a good question. And it’s the wrong one. I’ll tell you why.

The moment a new house is completed, it starts to wear out. The weather is at work on the components before the house is even finished. The ground around the foundations will be shifting. All the moving parts inside the house will start to disintegrate. The glazed units will be beginning to break down. And the water in the pipes will be looking for ways to escape other than via the taps. All that is a given.

Eventually something will go wrong. It will have to be repaired or replaced. As long as this ongoing maintenance is kept up, the house will remain serviceable indefinitely. It makes no difference if the house is masonry, timber frame, steel, concrete or any other building system, the principle remains the same.

So the question you should be asking is this. Which type of house is cheapest and easiest to maintain? Or put another way, what are the whole life costs of a house over, say, a 100-year period? Now that’s an interesting one to which there doesn’t appear to be a clear-cut answer. But it does raise some interesting additional questions. Like…

Windows: plastic v timber
The initial costs are pretty similar. However timber needs redecorating every five years or so, if it is to last. But well-maintained timber windows will last for centuries. Not so plastic, which has to be ripped out and replaced every 20 or 30 years. Such is the high cost of labour in this country that it’s actually cheaper to rip out and replace plastic windows three times a century than it is to install timber ones once and then repaint them 19 times. But will that relationship between low material costs and high labour costs persist for a century?

Timber frame v blockwork
Surely blockwork will last much longer than timber? Intuitively, yes. But timber, if looked after, will last indefinitely. In this case, unlike the timber windows, the frame doesn’t need maintenance to last the course, it just needs to be kept dry so the weatherproof cladding around it must be kept in good condition. So really it’s more a question about what external claddings last the pace.

What external claddings last the pace?
You’d think brickwork would be No 1. Sometimes it is, but it also has an alarming failure rate. It tends not to be very watertight and once the water gets through the brickwork it starts to cause all manner of problems behind it in the cavity. Soggy insulation, rusting wall ties, damp patches inside the house and, if timber frame, potential rotting away of the frame. The success of brickwork depends on a number of factors, the principal ones being how well it is built, how well the cavity behind it is detailed and built and what conditions it subsequently gets exposed to. If it works, then brickwork should be pretty much maintenance-free. But it’s a big IF. Of the main alternatives, render is prone to cracking and hence much the same problems as bad brickwork; tiles are prone to wind damage and cracking; timber boarding performs pretty well but can be prone to rot and needs attention on the decoration front if it is to continue to look good. In short, they are all less likely to suffer initial failure than brickwork but they are also much more likely to require routine maintenance, if not full replacement, before the century is up.

Caspar: a Halloween nightmare

A flagship off-site construction project in Leeds is being evacuated because of fears it will blow down in high winds. Caspar (pictured here) was built in Leeds six years ago by the Joseph Rowntree Foundation (JRF): it was one of the first of the new wave of prefabricated housing projects which the government has been so keen to promote. Oh dear. Oh very, very, dear.

Where did it all go wrong, John?

One of the key selling points for using off-site construction methods is that the amount of snagging is minimised. The aim is to hand over the new building “defect-free.” No doubt when Japanese contractors Kajima handed over the 45 flats that make up Caspar to the JRF for them to rent out as affordable homes for swinging Leeds singletons, much was made of the advantages of all this.

But, according to a report in this week’s Building, something’s obviously been amiss there for a long time because Arup, our premier building consultants, were hired to look into things, and Arup have said “Yipes – get out of there before the next storm.” What they’ve actually said is that there is a 2% chance that the whole building will collapse in high winds. JRF have even admitted that, if the cost of repairing the fault is excessive, they will consider demolishing the whole structure.

Is this the beginning of the end for modern methods of construction, just as the collapse of Ronan Point in 1968 marked the death-knell of system-build in the UK 40 years ago? It’s impossible to say. Caspar (stands for City-centre Apartments for Single People at Affordable Rents (neat or what!)) is just one of many such projects and, unlike Ronan Point, no one has been killed or even hurt. But the failure of Caspar does pose some very uncomfortable questions. Such as…

Why are so many of these schemes prototypes?

Why the curved roof? Doesn’t it look suspiciously like a wing!

Why the curved anything? I thought this was meant to be affordable housing.

Is this stuff really so very different to the 60s tower blocks?

Is it a design failure or is it, yet again, a workmanship issue?

This project was actually conceived as half-modular, half-flat pack. It was designed by Levitt Bernstein, an architects’ practice at the forefront of the new social housing, and factory-built by Volumetric in Bedfordshire. The superstructure was erected in less than three weeks. Each flat is 51m2 and cost £68.6k to build. That’s a typical outcome for these schemes. In other words, they are rather more expensive than conventional builds. They would become cheaper if we built thousands of them but we don’t. Every scheme is different. Every architect has a different spin to put on the concept of off-site construction.

And now here, in the brave new world of modular construction, we have an award-winning scheme about to be condemned.

Why timber frame isn't always quicker

One of the biggest questions every would-be selfbuilder has to address is whether to go brick and block or timber frame. For me, it’s something of a hoary old chestnut, having built using both methods and having been writing articles around this issue for many years. You’d think I’d have pretty set views by now but I don’t. Partly this reflects my tendency towards hopeless prevarication in all things, but it’s also undoubtedly because it’s a question with no easy answers.

However, a site visit this week to a timber frame house in Cambridgeshire gave me pause to think afresh about the issue. This was a site in a conservation area and that meant that there were some quite prescriptive design guidelines: basically it had to match up to the house next door, which was a detached vaguely-Georgian style brick house under a slate roof.

However the new house design was L-shaped, a stepped L-shape no less. What I mean by that is that it consisted of a main rectangular-shaped box (or aisle) of two storeys, with a single storey rear extension. My attempt at the relevant 3D house design is pictured here (blogger.com permitting). A common enough house layout, you’d think. And you’d be right.

The junction of the two parts of the house is what I wish to draw your attention to. It’s referred to as an abutment. However you do it, it causes problems because there are waterproofing details to be made between the wall of the main aisle and the roof of the extension. Lead flashings have to be fitted over the junction and the wall cavity has to have an effective barrier laid across it in order to drain any cavity water out onto the extension roof. If not, you risk the cavity draining directly down into the house below the junction. Messy.

The subject of cavity barriers can wait for another article. What I am driving at here is the disruptive effect this all has on the construction of a brick-clad timber framed house. On a simple rectangular-shaped house, or even a more complex one, as long as it has a single eaves level all the way around, the brick cladding is taken off the critical path and can be completed at leisure after the framers have departed. Indeed it can go on simultaneously with the roof covering and the fit out inside.

However, split the roof levels like this and the brick wall of the main aisle has to be completed before the extension roof can be covered over. Indeed, on the site I was on, space had to be left for the brickies to have access to the wall, which was being built-up off a steel beam over the opening. Not even the extension roof carpentry could be completed.

Result? Weeks had been lost waiting on the brickies. The roofers had been in and finished the main aisle but had to return for another visit to cover the extension. The scaffolding was gently racking up hire charges. The house could not be effectively waterproofed, let alone secured. All the supposed speed of construction advantages, which timber frame sells itself on, had been lost. In fact it would probably have been quicker to use brick and block on this house.

Had it been such, you would not have noticed that there was a problem here because the roof carpentry wouldn’t have started on either roof until the brickies had got to eaves level all around. And had the external skin been anything other than brick (or stone), the wall above the extension roof could have been finished off at leisure off some form of adapted scaffolding. Boarding, render on mesh, hung tiles, all fine: they are hung off the timber frame and wouldn’t disrupt the critical path. But bricks? They have to sit on something, be it a steel or a foundation: they can’t be ‘hung’ off the background timber frame and they can’t be built-up off the roof cover. So bricks have to be in place before the extension roof cover can be laid.

There’s a lesson in here somewhere. To simplify it right down to basics, despite what the salespeople say, bricks really do work best with blocks behind them. And timber frame is really seen at its best when it’s clad in something other than brick or stone.

Importing Homes from N America

I'm looking at importing a house from America. Has anyone out there done this? Do you have any company sites or contacts that export to the UK?

You will find that many custom build companies in both continental Europe and N America are more than willing to build a house in the UK or Ireland.

Having said that, the ones that build the most are the ones that have local contacts on the ground over here. The Americans have proven rather poor at this - they build when they are asked but they don't put any effort into export or marketing. The Canadians are much more proactive - they took a large chunk of floorspace at Interbuild in 2004 for instance - and many of their custom builders are excellent. I have seen the results of two; Allouette who are building in West Malling in Kent for Sunley Homes, and Interhabs from Nova Scotia who have been building in Co Mayo and around Inverness (pictured here). Interhabs, I know for sure, are looking for individual custom builds. Find out more from www.super-e.com. Plus the exchange rate on the Canadian dollar is such that they are almost certainly going to offer better value than their neighbours to the south.

Alternatively, there are a number of UK builders who make a thing of building in the N American styles. Tim Crump of TJ Crump Oakwrights is an enthusiastic student of N American (and German) housebuilding methods and has recently completed a few N American style homes.

Spluttering SIPS

I have been writing about SIPS (Structural Insulated Panel Systems) building panels for five years now. In that time I have been on six sites where they have been being used. Each time there has been a tremendous amount of enthusiasm for what’s being, or has just been, built. And yet each time there also seems to have been a catalogue of mishaps and delays. These are somehow never attributed to the SIPS building system itself, but seem to be caused by the shortcomings of the designs or the supplier.

This time last year I visited a site of six terraced houses being built by Nick Robinson in Hampshire. Nick is a professional housebuilder with a keen interest in new construction systems. I had previously seen a site he had built out using steel frame, which had worked pretty well. But when he enquired about using steel frame again, Corus, the manufacturer, told him they were no longer interested in supplying small sites so he had to look elsewhere. Nick chose Kingspan’s TekHaus, a polyurethane SIP panel made in Germany, and the main player in the nascent UK SIP market.

I had seen Nick a few weeks before visiting his site. Back then, he had already laid the foundations and he told me then that he hoped to have the structures of all six houses up for my visit at the beginning of October. The reality couldn’t have been more different. Two slabs were untouched and on the four where a start had been made, none had a roof panel on. Kingspan were using a middleman, known as a process partner, and he had screwed up badly. Deliveries were late, were in the wrong order, were sometimes damaged and some panels were cut wrong. A lot of time was wasted on site trying out to sort this mess out, and often rebuilding the panels. Brain Clark of IPC Erecting, who ran the site crew there, told me that they eventually finished the contract at Christmas. What should have taken six weeks took nearly five months.

Last week I was in Wiltshire visiting another Tek Haus (pictured) in the process of erection. Who should I run into here but Brian Clark. And what was Brian’s erection crew up to? Mostly snagging again, it would seem. This contract, a single house of just over 300m2 floor area, had started on August 1st and was booked for a three-week erection. Here we were, in Week 8, and the roof panels were yet to go on. The story behind the delays was familiar. Unresolved issues with structural design, panels being supplied that weren’t quite right, glulam beams being supplied which weren’t nearly right, crew waiting for cranes. All little niggly things, all very disruptive. If these two jobs were exceptions, all well and good, but the truth seems to be that they are not exceptions at all and that delays and overruns are in fact the norm with small SIPs projects.

I was first introduced to SIPS building in Britain by Tim Crump, proprietor of TJ Crump Oakwrights, a Hereford-based green oak housebuilder. Tim was an early adopter and had understood how and why SIPS were beginning to take-off in North America where they are frequently used to wrap-around custom homes. Tim built several houses incorporating SIPs but by 2003 he had fallen out of love with the whole concept. At the time Tim told me that he thought there was a big gap between theory and practice and that they always seem to spend far too much time sorting out snags on site that should have been designed away in the office, before the panels ever left the factory. Tim ended up losing money on a contract to erect two houses for an architect because none of the pre-cut panels actually fitted together as designed. After that, he has steered well clear of factory-made panels and carried out this type of work in timber frame with his own site crew.

So is the SIPS bubble about to burst? Far from it. Adam Holmes of Kingspan told me this weekend that the TekHaus system is making great progress in the social housing market and that the volumes they are doing are now enough to justify setting up a manufacturing plant in England. SIPs score highly on the EcoHomes ratings and housing associations have to take this into account when placing orders for new housing. It seems that with all the money the government is throwing at affordable housing, the future of SIPS is looking pretty rosy. Or at least it is for Kingspan’s TekHaus, which is out and away the most important of the UK suppliers.

However there must be a question mark over the use of SIPs in the selfbuild and one-off developer sector. The way Kingspan have organised this, the responsibility for design and erection falls onto a number of regional process partners who effectively quote against each other to supply essentially the same product. This creates some surprising anomalies; for instance, the house I visited in Wiltshire was being supplied from Northumberland, because the quote was 25% cheaper than the local supplier. The Northumberland supplier, SIPHome, might just have made money on the project if, as they had planned, it had taken three weeks to erect and had been all delivered in six loads. But of course, the job had overrun spectacularly. This of course was SIP Home’s problem but it doesn’t instill confidence in a delivery system that is so wayward and disorganised.

It’s not as though these SIPS systems are cheap. The Wiltshire contract for erection of watertight shell was valued at £68k. It’s hard to see the equivalent in blockwork, or a basic Taylor Lane-style timber frame, costing more than £50k, including all the insulation. Whilst the resulting home should be extremely energy efficient — there is no central heating system, as such, just a mechanical ventilation system with heat recovery — it’s still an awful lot to pay for a finish standard, which could easily be achieved by other means.

The question is this. Is there something inherent in SIPS building methods that makes them prone to delay and cost overruns? Or is it just that the assemblers haven’t really worked out how to get the best out of the system. For what is clear is that the goodwill towards what promised to be a new, green building method is running low and that if someone doesn’t sort it out soon, it is in danger of evaporating altogether. Many selfbuilders I meet are very enthusiastic about using SIPS for their projects but it seems that they are not really getting the service they pay well for and hence deserve.

It may be that in the short term, SIPS are best suited towards very simple structures such as you would find in most social housing projects and maybe commercial sheds, but not for the more complex designs met in the selfbuild arena. In theory, the design skills needed for SIPS construction should be very similar to timber frame and there really isn't any reason to expect problems, but in practice there seems to be a skills gap somewhere in the SIP production stream which is proving costly and frustrating. What should be a simple, quick and efficient system is currently anything but.

If you can only see this single article on screen please click on the 'Housebuilder's Update' heading (screen top left) to view the rest of this month's items and have access to our archives.

What's Belgium got to teach us about housebuilding?

To British students of modern manufacturing methods, Danilith are known (if they are known at all) as the Belgian housebuilders with the robot bricklayers. But that is only a small part of the story. For Danilith is a vertically integrated housebuilder of the type that just doesn’t exist in the UK. They are a family business that has been running since the 1920s and they own a large plant in Wortegem, which they use to build as much of their houses as they possibly can. Not only do they prefabricate wall, floor and roof panels, but they make their own joinery as well. They employ 260 people, use very few subcontractors and they undertake around 250 projects a year, mostly individual homes in Belgium and Holland, mostly fully finished.

Let’s just consider this last sentence. 250 houses built each year by 260 people. That, in itself, is remarkable. No one at Danilith works more than an 1800hour year if they can help it — Belgian overtime is very heavily taxed so there is an incentive to stick to the contracted hours. Of these 260 employees, just 30 work in the factory, another 100 on site erecting and finishing the houses and the remainder in various admin roles. So essentially, 130 workers are building over 200 houses each year. That’s just under 1200 hours labour going into each house. In comparison, the standard British site-built home takes 3000 hours to construct. Danilith workers are thus over twice as productive as their UK equivalents.

This is reflected in the prices charged for Danilith’s homes. Their brochure shows that the selling price is almost always under €1,000 per sq m (£690/m2). But in Belgium, new homes attract VAT at 21%, which is included in this figure. It also includes 4% taken for design fees and a mandatory 1% fee payable to a private security firm to oversee the construction sites. The net sales price is in fact just €800 per sq m (£550/m2). It’s difficult to see something of equivalent quality in the UK costing less than £900 per sq m. In fact most MMC advocates say that you can’t build at less than £1,000 per sq m unless you are constructing identikit houses in long production runs. But Danilith’s homes are all different, although the elements used to put them together are modular.

In so doing, Danilith produce a very high quality house. We are used to seeing timber and steel framed houses built in factories but Danilith work mostly with brick and concrete, reflecting the prevailing preferences in the Benelux countries and France. The 10metre-long wall panels passing through the Danilith plant are not unique: in Germany, basements are commonly constructed with prefabricated walls. But Danilith have added to this a robotic brick laying machine which first slices the bricks in two lengthwise and then lays them face down on a flat bed, against some retardant paper. Mortar is subsequently brushed into the joints from above before three more layers, concrete, polyisocyanurate insulation board and lightweight concrete are added to the beds, the whole process taking around four days to cure and the final wall thickness being around 250mm. Finally, the walls are hoisted vertically, to be finished by hand on the last of the production lines. Windows, doors, electrical channels and plumbing runs are all set into the walls as they are being constructed. You can view the 15 stage process on the website www.milbank-danilith.co.uk. Floor panels are constructed elsewhere in the plant. Roughly a house a day passes through the plant: on a traditional British building site this would represent the efforts of around 600 hours work. In the Danilith factory, it’s 25 workers on an 8-hour shift: 200 hours. Plus, of course, a lot of investment in machinery and forty years worth of know-how.

The technology used by Danilith is far from hi-tech. It’s the sort of thing that is used on production lines across the globe. Parts of the process are still carried out by hand and compared to the latest robot car builders it all looks a little primitive. And there are many other smallish housebuilding businesses in Germany and Scandinavia using similar finished-panel systems. However, in the UK, the nearest we come to factory production in housing are the numerous timber frame companies who semi-fabricate wall and floor panels, sometimes referred to as open panel building. This requires insulation, services and finishing on site, both internally and externally.

Danilith see themselves as custom homebuilders first and foremost. Although they make windows and staircases, as well as the wall panels, they don’t sell them to third parties; they only supply their own projects. However, they do undertake a little spec. building from time to time but usually to fill gaps in the production schedule that would otherwise have them laying off staff. This is quite revealing in itself. It’s frankly hard to imagine a British firm adopting such an attitude. Indeed, it’s just the sort of action that shows the gulf between Anglo-Saxon and Continental business practice. Whilst we are busy cutting costs and growing market share, the Belgians are putting the interests of their workforce first and foremost. Whilst many British economists would highlight Danilith’s working practices as being a prime example of Eurozone featherbedding, they would also have to admit that it’s resulted in productivity levels which we in the UK can only aspire to some time in the not-so-near future. The Belgians are clearly doing something right.

This remarkable company would like to transfer some of its know-how to Britain. They have found a willing partner in Milbank, who, like Danilith, area a family-run company with a keen interest in masonry prefabrication. Milbank are big in precast flooring in South East England and have slowly grown their business to embrace haulage and joinery. But they are not housebuilders. Both Milbank and Danilith are keen to start building homes in the UK but there are significant obstacles to overcome in order to establish a British version of Danilith.

Perhaps the biggest problem is that there simply is no tradition of custom home building in the UK. On most of the continent, custom home building is the principal route by which detached houses are built. Individuals buy plots — relatively cheaply — and commission a new house to be built on them, without spending a great deal of time working on the project themselves. In Britain, the new homes market is dominated by speculative builders who build solely for sale: there are around 15,000 individual homes built each year but most of these are taken on on a semi-DIY basis with the selfbuilders acting as their own contractors. The burgeoning timber frame sector prefers to offer a water-tight shell option and lets the selfbuilders finish the houses off in their own time. One or two Continental companies offer a full-build service, notably Huf Haus and the Swedish House Company, but they only account for a tiny number of new homes each year, probably under 100 in total.

Typical of Continental custom homebuilders, Danilith is set up to build one-offs. Everything in the Danilith system is modular but provided the design fits the modules, almost any design can be built. They don’t aim to achieve economies of scale by mass production: their goal is to simplify everything down to a set number of options and to avoid prototyping. That way they avoid mistakes and they keep costs to a minimum, without becoming too repetitive.

Quite why the major British housebuilders have failed to take up factory-building Danilith-style is a mystery. Their pattern book approach to housebuilding, combined with the guaranteed volumes, ought to lend itself to prefabrication. But for some reason vertical integration has never appealed to British builders and no one has ever seen fit to try. The only plc housebuilder to currently use a significant input of prefabrication is Westbury, who have built the largest timber frame plant in Europe (Space 4) to supply some of their new homes. But Space 4 is a long way short of what the likes of Danilith and Huf Haus do.

The big question is this. Can the cost savings shown by Danilith be achieved simply by copying the technology or is the different business culture an essential part of the mix? Whilst it ought to be easy to introduce robot bricklayers, it will be much harder to integrate the Continental work ethic needed to keep the robots ticking over. Mark Brinkley