Limewash over emulsion paint 

I’ve been wondering how best to  redecorate a room whose surfaces are finished in old materials – clay and plaster – but which have been painted with modern emulsion. That may be creating a breathability problem.

Very modern emulsions are impermeable to moisture. Cruder basic emulsions have a certain amount of breathability, but it’s hard to tell which it is, looking at the result at least 30 years after it was last painted. So should the emulsion be removed?

That would be the ideal solution. The walls could then be distempered, which I have read is the usual treatment for inside walls in Suffolk. Distemper can still be bought.

My answer is no to stripping, because emulsion paint is pretty tightly bound to the surface and my attempts to get it off with chemical strippers or wire brushes caused significant damage. The best I could do with a wire brush was to remove loose paint flakes.

So after a year or two of procrastination, I’ve decided to limewash over it, which should not make the breathability too much worse, even if it doesn’t improve it. I ruled out distemper on top of emulsion because I don’t think it will stick very well and it is also quite prone to leave dusty traces on clothes when touched.

But will limewash stick to emulsion? My favourite paint maker, Ingilby of Glemsford, says its limewash with linseed oil can be indeed be used over emulsion, and I’ve tried it in the past on a smaller area and it’s very effective. So I’m using it to completely redecorate  two rooms. One of the rooms  is the most complete in the house, in the sense that it has walls of unplastered plain clay daub (with straw showing) and a lath and plaster ceiling.

Ingilby’s limewashes can be bought in custom colours so I took along a sample of a creamy off white that we liked, which they copied exactly (and it’s now in their list as a colour with our name on it).IMG_5858

Limewash that I made up myself using instructions from books on lime was very thin and watery, and it was not surprising that the recommendation was five or six coats. Ted Ingilby’s lifelong professional experience of paint technology – not just lime, and including marine paints  – has produced a product that fully covers even a contrasting surface in only three coats, the first diluted 2:1 and the second 1:1.

It is still very liquid on the brush and so quite messy to apply, though nothing like as runny as my trial of home made stuff. The floor must be very well covered and exposed oak timbers masked with sticky paper tape, though I’ve found (see photo above) that only the horizontal timbers need masking – Ingilby limewash is controllable enough on the brush to avoid splashing the verticals when painting the panels between them.

This has now transformed decorations which previously were very obviously modern emulsion  into a beautiful matt surface with the soft, subtle colouring of limewash. It’s not the purist’s answer – that would be to strip and distemper – but it’s the pragmatic one.

Rat assault 

Just discovered a disadvantage of shallow brick footings: a large rat managed to tunnel right under them and up inside where the water and waste pipes enter and leave the building. 

Caught the rat with an electronic trap and filled the hole down to the bottom of the footings with a couple of buckets of limecrete, using lime, sand and gravel.  

Had to do it twice more because another rat twice bypassed the limecrete by digging a longer tunnel, partially removing the still soft limecrete as it went.

  This photo shows the top of the pad of limecrete, between the brick flowerbed retaining wall and the house wall. The limecrete is smeared with dark brown Jeyes Fluid in the hope that the foul smell will finally put off the rats.

I suppose the rats are in their natural habitat – we live after all in an old farmyard!

Old fashioned paint

It has been a positive pleasure painting the windows in the last few days using custom-mixed linseed oil paint from Ingilby’s of Glemsford. The window frames were stripped and repaired, and much of the putty renewed, some time ago, and they have since been given several coats of pure linseed oil to prepare and waterproof them. But for various reasons, I didn’t get round to completing the job.

Now I’m painting, I can immediately see the benefits of using linseed oil on old wood. Where I’ve left a leading edge of paint on bare wood, the oil spreads out of the paint into the wood, which seems to lap it up, almost pickling the timber in oil. That can only be good for long term preservation, especially as absorption is helped by the slowness with which the paint dries.

Slow drying is also the main drawback. It takes a good 24 hours to become touch dry and a week or more to harden, so if there are dust or insects in the air the surfaces will end up a little messy. It is not surprising that linseed oil paints were abandoned as soon as quick-drying solvent based paints became available. I’d certainly not bother with it on a modern building.

But on old wood on an old building, the paint is ideal. It is not cheap – £60 for 2.5 litres mixed to the RAL standard grey I specified – but it has excellent coverage and brushes out very easily. One coat covers completely, though two is recommended for a long lasting finish.The smell is of raw natural oil, which is not to everybody’s liking, but I find it rather pleasant, especially compared with solvent-based paints.

Woodpecker damage to clay daub

Went on holiday for 3 weeks, leaving the clay daub repairs till afterwards (see earlier post)  – and this is what the woodpeckers have done to the bare patch, digging for insects while the house was empty. One of the holes is six inches deep. Won’t leave repairs so long next time… 

 

Back to daub

005 (3)

Exposed clay daub

After focusing on  a new timber framed extension for so long, we’re now back to repairing the old building. Weak bits of of a rough cement render have been dropping off  the front wall, as the picture shows. The cement is just a thin layer skimmed a few decades ago onto clay daub.

We’ve been expecting this, but took a decision early on to let it happen by natural weathering, and then patch repair bit by bit.  Looking at the evidence of what’s there, we think patching is the way the walls have been maintained for a very long time – any weakness, then they just slapped on a bit of clay and limewashed it, and owners only recently took to adding cement as well.

In 2009, we repaired a lot of bare clay patches on both the front and back walls which were being ravaged by woodpeckers in search of insects. After attending an Essex County Council course at Braintree, we did all the repairs with a render of lime and sand.

Since then, as set out in previous posts, we’ve discovered that sand and lime render is not the authentic Suffolk method at all.  The local specialists use a plaster of chalk, lime and fibre on outside walls instead. Having done a gable wall that way, we’re now absolutely convinced it is best. Once limewashed, it is weatherpoof, it sticks much better than render to any substrate and when dry it is very strong yet still flexible. So no more clay, sand and lime render.

Instead, the plan is to use a new daub of clay and straw with a little sand to repair holes in the old daub. We have saved a few barrels of the pale grey boulder clay that lies under the house and pond for just this purpose. Its colour when dry exactly matches the clay exposed on the wall.

One option would be simply to limewash the clay, which as far as we can tell is the only protection it used to have. There’s a whole wall on the farmhouse next door that is clay protected only by limewash.

But because the exposed clay is rather fragile and crumbly, we think it is best protected with a thin layer of plaster to replace the cement render that has just dropped off. This will protect the old clay indefinitely.

How do I reconcile this with the decision to  completely re-render a big gable end wall with Savalit boards and lime plaster (see earlier posts)? The answer is that the gable had been rendered with concrete on wire netting, and all the concrete had to be removed. There was no clay at ground floor level, so patch repairs were not an option. The clay higher up would be protected by the boards. We’ll probably want to use the same method for the other gable. But there is enough clay in the front and back walls to make it worthwhile patching in the traditional way.

Other commitments will delay the work till later in the summer, but by then it is almost certain more will have fallen off. Below is a temporay solution to keep another large patch together. A quick application of lime plaster with chalk and fibre to the cracks is strong enough to stitch these pieces together for a few months longer.021

 

 

 

 

 

Adding an oak terrace

We needed something to connect the big glass doors with the pond, so we built an oak terrace cum patio.

002

The finished terrace and step

It was to be an irregular shape because of the pond, so we drew it exactly, levelled the ground and dug nine post holes, into which were inserted short lengths of 100×100 mm fence posts, bedded in concrete.

Holes for the posts, lined up with the first joist The stakes mark later holes.

Holes for the posts, lined up with the first joist The stakes mark later holes.

The four showing at the front were oak, and the rest were standard treated timber. When all the posts were in, we screwed the outside joists onto them, using heavy duty exterior screws. This was easy for the softwood, but the oak posts at the front had to be drilled out very carefully. All joists were 150x50mm treated timber. The longest one by the brickwork was 3.6 metres.

First joist fixed, so others can be measured from it.

First joist fixed, so others can be measured from it.

The books advise maximum spacing of posts of 1200 mm and of  joist centres 400mm, and we stuck to that (in fact less in this case for the joists because of the shape of the terrace). Before the inner joists were fixed, we put two pairs of bearers across from front to back to spread their weight, maintaining the same 1200 mm spacing.

Frame ready, with two pairs of supports for the inner joists

Frame ready, with two pairs of supports for the inner joists

Once the whole frame was in and fixed we could start contemplating the oak planking. This we bought from an excellent supplier in Corby, Northants, called UK Oak , though for small quantities their fixed minimum delivery charge is quite high so we went 70 miles to collect with an estate car. (Their service is good – there was a small problem with the order when we arrived, so they ended up delivering it to us free a couple of days later). We went for smooth rather than ribbed, since oak has a good natural non-slip surface if it is scrubbed regularly.

The planks were fixed with stainless steel screws, also supplied by UK Oak, and then extra planks were used round the outside to cover up the grain ends where the planks were cut and to hide the cheap softwood frame.

The finished terrace, with step and oak facings

The finished terrace, with step and oak facings

Finally, a small (1800x500x120 mm) frame was built for a step by the window, and planked in a similar way. Gaps between the planks for drainage were 4.5 mm (calculated to avoid having to cut a plank lengthwise, and controlled using a thin template) which is just enough to lose something small and valuable through them. But the design and the use of stainless screws allows planks to be removed again, just in case! The terrace is too low to get underneath to retrieve things.

 

 

Finishing touches

We’ve done a few small jobs to finish off round the extension over the last few months.

The old footings, with Japanese anemones growing from them

The old footings, with Japanese anemones growing from them

To show a bit of the the history of the site, we decided to keep some of the footings of the original building, which fell down decades ago (and whose previous existence played a big part in getting permission to rebuild). Part of the footings had been cut through for the main drain, and other parts had been damaged by machinery, so we ended up repairing them, using lime mortar.

The extension floor was dropped down a foot below the old ground level, so keeping the old footings meant creating a narrow trench along the side of the building, the bottom of which was lined with anti-weed fabric covered in gravel. Some of the rampant Japanese anemones that used to grow on the site survived, their roots lodged in the cracks between the old bricks, so we left them where they were, and they are flourishing again. As the photo shows, the old footings are at an angle to the new building, because we slightly changed the orientation, to look straight down the pond.

Steps, and more of the old footings

Steps, and more of the old footings

Next, we remade the path from the parking area, re-using York stone slabs, and made steps to the new front and back doors with 900×600 mm Bradstone natural sandstone slabs from Travis Perkins. They were labelled autumn green, but are more of a grey, and when using them previously we found they weathered quickly to a shade not very different from York stone, at a quarter of the price.

Using the same sandstone and new bricks, we then made steps up one side of the extension from pond level to the back garden. Building even a modest flight of steps up a steep slope requires some careful planning, so we measured the site exactly and drew out the plans on paper before cutting the ground. The risers were made from bricks standing on edge, which created exactly the right height.

Next: creating an oak platform by the pond.

How to stop worrying and relax a little

Sometimes it’s quite hard to relax when you live in an old house, because the lazily wandering eye will all too often spot something not quite right: a beam that suddenly seems to have too little support, or a pile of dust that might betray active wood pests. Nine times out of ten the risks spotted in that casual glance prove, on reflection, to be exaggerated, perhaps a function of the pessimistic frame of mind that develops when repairing an old house: the deeper you examine the building, the more problems you have to solve. Here are pictures of three things we did after staring rather a lot at potential problems. Probably only one of them was really necessary, but doing the other two made us feel better.

IMG_3335

Wrought iron bracket

The first picture is of a wrought iron bracket we had made to reinforce the end of the biggest beam in the house, which does not quite meet the wall (a result of the repair of the chimney and fireplace which were leaning into the room but are now straight, which moved the bearing surface in the wall away from the timber). The beam end since then has rested not on a post but on a one inch thick steel slab inserted in the wall whose other end bears the weight of a beam and studs above. Solid and permanent, we were assured, but sitting in an armchair underneath, somehow it didn’t look right. The bracket bought piece of mind, and looks nice, too.

IMG_3339Next, the only repair of the three that was definitely  necessary, and so we commissioned it early on. A large section of the massive chestnut beam in the kitchen had been weakened by old rot. The beam deflected worryingly when walking across the room above. A deep slot was cut in the beam from above and a T-section steel “flitch” beam was cut to drop into it, bedded in resin. The flat top of the beam rests on top of the wood. You can just see the steel in the photograph through one of the cavities left by the rot, which  long ago disappeared, leaving crumbly wood around the holes.

Oak knee

Oak knee

The third picture is of an oak knee, a precautionary reinforcement of the end of another main beam, which does rest partially on a post, but after staring at it over several years, mainly from an armchair, it was decided that there wasn’t a big enough contact surface between the two. The knee is a naturally grown shape, bolted onto the other timbers. You can see from the photo that we were probably worrying too much, but then we definitely feel more relaxed when we look at it, especially as that wall had a significant risk of movement when the extension was being built behind it. Starting on the extension was what prompted us to do something, just in case.

PS on lime

I gather that there is questioning among some East Anglian lime specialists of the claim reported in the previous post that traditional hair mixes are significantly prone to failure.

As I mentioned, this is not about the well known deterioration of hair if it is left too long in a wet mix before being used, but about intrinsic faults in the hair, especially if it is imported.

Another lime specialist tells me that there has been little peer to peer discussion of this issue among experts, so it is not clear how extensive the problem really is. There’s a view around that it is really to do with persuading people to adopt artificial fibre mixes for commercial reasons.

I have to say that even if the problems with animal hair turn out to be too rare to bother about, I’m still a convert to using fibres and will continue to use Anglia Lime’s Fibrechalk or a similar mix for the next stage of repairs ( we have two more outside walls to do). Anglia’s product was easier to use than the natural hair mix I tried before and doesn’t deteriorate if it is not used soon after mixing.

Maybe that plasterer who put the wind up me last summer was just pushing his own fibre mixture! (It wasn’t Anglia’s).

My main point stands either way – things have moved on from the purist days and it is OK to mix old and new.

Lime developments

I have to admit now that I have been a bit nervous about our new chalk lime plaster since last August, though I haven’t confessed it so far to the blog: we used imported pigs hair for the plaster in 2013, and more than a year after it was done I was speaking to another lime specialist who gave a deep intake of breath, frowned, and asked: has it fallen off yet? He seemed to relish recounting the story of another Suffolk house where the pig hair had disintegrated and the plaster coat had failed in the first year, on his version of the story because it had a foreign bug that had eaten it.

I’m now much more confident that our pig hair plaster is OK. There are no signs of failure, and from time to time over the autumn I pulled out a bit of hair near the surface and checked it for strength. I tried again this morning and the hair was as strong as ever.

I’ve now discovered more about the pig hair issue in a new briefing on lime which has arrived with the latest issue of the magazine of the Society for the Protection of Ancient Buildings, of which we are members. An article by Joe Orsi confirms that three years ago it was noticed that there was a pattern of render failure caused by degradation of animal hair. Samples were examined by a member of the Building Limes Forum, and it was found that some of the hair imported from outside the EU had been aggressively treated with acids to kill anthrax spores; the hair had been severely weakened because the keratin that binds it together was damaged.

More alarmingly, further tests with some locally sourced, untreated hair have found similar patterns of failure. Mr Orsi also says that there is evidence of hair failure in the past, but other hair has been unaffected and it is not clear why. (In the context, I assume that this is about hair that starts in a damaged state, rather than the well known problems arising when unused hair and lime mixes are allowed to stand for too long, which dissolves the hair). He goes on to discuss alternatives to hair such as artificial fibres, which we subsequently used for indoor plastering.

As I have mentioned in earlier blogs, when we first got involved with repairing a very old building 8 years ago there was great enthusiasm for complete authenticity, which meant we were encouraged to use ancient techniques wherever possible. But even at the time, revisionism was setting in, with a growing appreciation that there could be drawbacks to old methods and materials that in some circumstances make it better to mix ancient and modern. The discovery that animal hair can fail suddenly highlights this.   I’m sure that if lime had not fallen out of use and then been revived after a long gap, there would have been craftsmen warning many years ago about the various ways hair mixes can fail.

The importance of lime remains absolutely clear, as does the advice against using cement in any building originally constructed using lime. But the more we learn about older methods, the more it seems we can find modern techniques to overcome some of the drawbacks.

I recommend the SPAB lime briefing for getting up to date on what the specialists are saying, including modern lime products such as Hempcrete and Limecrete.

Archaeologists find an ancient invoice

A rather large invoice arrived for work by Suffolk County Council archaeologists six years ago. It was a condition of planning permission that we paid for an archaology report.

They established that we owned what had once been a brewhouse, and among other things they painstakingly troweled through layers of soil and took away bricks from the Brewhouse hearth to date them. Interested to learn as much as possible about the building, we chased their report a number of times over the following year, were twice promised it but nothing arrived, and we eventually gave up and forgot about it.

Then out of the blue an invoice for more than £4,500 landed in the postbox, and still no archaeology report. Six years before we would not have grumbled, because we had set aside the cash and would have liked to find out more about the building’s history, but after this length of time it’s a blow. When we pointed out that there was still no report, the council withdrew the invoice, and said the report would be available within the next two months. It wasn’t afte two months, four months …. we wrote again, but no response. They must have lost all the paperwork. After six years going on seven, I really don’t think they can legally revive that invoice.

The Old Brewhouse saved

Nearly eight years after buying The Old Brewhouse, and six years after the start of work, we’ve now got the home we imagined when we first came across it in 2007 in a crumbling but idyllic state. There were roses climbing round the doors and windows, ivy smothering the walls and chimney and a lovely but overgrown pond and meadow for a garden.096

The roses will soon be back to their former glory, though not the deadly ivy (the quickest way to destroy an old clay wall), the pond has been restored and the meadow preserved just as it was, for the sake of its orchids and fritillaries. So this is a good moment to recap what we have done.

Looking down the pond

Looking down the pond

The previous owners had begun in the early 1980s to convert the building next door to their farmhouse into a dwelling, but their project stalled; we bought the property a quarter century later on the basis that their planning permission had lapsed and we would have to apply again from scratch. The fact that it looked like a house rather than a barn – and was probably a dwelling centuries ago – made no difference. In planning terms it had similar legal status to a barn.

The pond from the bedroom

The pond from the bedroom

003 (2)

The pond from the old building

Our initial meeting with the council’s conservation officer to discuss a new planning application was nevertheless a shock: her first reaction was to say that there had been too many conversions in the neighbourhood and they had decided not to allow any more. It began to seem as if we had bought a very expensive shed.

While we waited for our planning consultants and architects, Hollins of Framlingham, to resolve our problems, we spent days at a time camping upstairs, dining on fish and chips and warm white wine, thinking through how we would turn the building into a home and worrying whether we would ever get permission to do what we wanted.

Hollins eventually solved the problem by finding a way to get the original 1983 planning permission reinstated. If work is not started within three years, a permission expires, but if the job can be shown to have been properly under way, permission lasts indefinitely. The council had no records of the project starting, but using builders’ receipts that the previous owners turned out of old files for us and the visual evidence of work that had actually been done, such as the rebuilding and strengthening of the chimney, it was eventually agreed  that the original 1983 permission could be reinstated.

The whole process, including submitting detailed drawings for the new work we wanted to do, going through the planning and listed building processes, and finding a builder and getting his firm on site, took 18 months from the date we completed on the purchase.

010

Spiral stairs

Phase One of the building work was to repair the structure, install water, electricity and sewage, lower the ground floor to improve on the 1.7 metres headroom, repair the oak frame and rebuild the brick footings on which it rested. This proved to be a far from simple job. As shown in earlier posts, the building had to be jacked up section by section (a timber frame is very flexible), the rotten oak sole plates removed and replaced and the brick footings built back up underneath them. The floor was another big exercise, because the planners rightly banned concrete and insisted on a breathable substitute using lime.

We hired Robert Norman Construction  of Framlingham to do the work, but to cut costs to meet our then strictly limited budget we left a long list of jobs to complete later. So while the builders did an excellent job finishing the main conversion in the summer of 2009, it took another four years until late 2013 to finish the work for which we had planning and listed building consent. This included repairs to the walls, inside and out, which involved lime plaster and wattle and daub, (where we learned the skills and did a lot of the work ourselves); commissioning Felix Oliver, who builds new oak framed buildings (Suffolk Timber Frame Buildings), to make the new stairs, and working with members of Suffolk Building Conservation Associates: William Clement Smith to continue with the oak frame repairs and Alan Wilkins and Rory Sumerling to plaster one big section of the outside wall.

The bedroom, up the spiral stairs

The bedroom

Looking down the stairs

Looking down the stairs

As set out in the blog pages on our approach to the work, the basic rule was to repair wherever possible, and not to replace anything unnecessarily. For example, we have kept as much as we possibly can of the remaining old clay in the walls, and were delighted to receive written  praise from Paul Harrison, the Mid Suffolk conservation officer, for our efforts. As we have seen so often when we have looked at modernisations of old Suffolk houses, the first instinct of many owners is to rip out any remaining clay and replace it. There’s a particularly egregious example within walking distance of us.

While we tackled the remaining jobs piece by piece over four years, we took time out to restore the old farm pond, which is the main feature of the house’s setting. (See the pond blog  for details and costs). We also decided that it would be a mistake to build the tiny extension, the largest we could persuade the conservation officer to accept in 2007, and went back to the drawing board with Hollins to convince the council that we should be allowed to rebuild the part of The Old Brewhouse that had fallen down a few decades ago.

083

Across the meadow garden

Since we could prove that it had been half as big again in the fairly recent past, we had a good case. We had a  productive pre-application meeting with the planning officers and Paul Harrison (whose predecessor in 2007 had been so restrictive), and their views were fed back into the design; we took the trouble to brief the Parish Council, and won their support. All this smoothed the way to the planning permission and listed building consent we won in early autumn 2012 for a much bigger two-storey extension, designed by Hollins.

We started Phase Two, the extension, in January 2014. It is timber framed, in keeping with the spirit of the old building, even if technically very different. It combines a shape borrowed from the old building with modern detail such as aluminium doors and windows to prevent it looking like a pastiche.

There were some tense discussions over the foundations, which we were told would have to be piled, because it was to be built on clay next to a pond. (The old structure on whose site we were building had lasted quite a few hundred years, but rules are rules). Piling is very expensive, but DJE Construction of Attleborough, with Stroud Associates of Harkstead as our engineers, kept the cost very close to their original estimate, for which we were grateful, because the unexpected can be particularly costly where piling is concerned, and the design had to be changed several times.

A sideways glimpse of the pond

A sideways glimpse of the pond

We were lucky to find that Terry Booty of Booty Builders of Laxfield, along with Andrew and Tim, his two excellent craftsmen, were able to fit our extension into their schedule in the spring, and they completed it in the early autumn. They were multi-talented, with all the building skills between the three of them. (They only subcontracted plumbing and electrics, lime plastering the outside walls and the specialist metal roof on the single story structure that joins old to new). Terry was happy for us to take on some of the sourcing and ordering, notably the aluminium windows and doors (Kloeber) , the spiral staircase (British Spirals and Castings, formerly Cottage Craft Spirals)  and the slates (Glendyne from Cembrit) , and we also researched and specified the oak front door (Jonathan Read, Malcom Neeve Joinery), the tiles and – with advice from lime specialists – the external plaster and paint, which was limewash from Ingilby’s of Glemsford.

This all meant some interesting weeks visiting suppliers in East Anglia, London and as far afield as the Peak District (for the stairs), comparing specifications and costs and double and treble checking every measurement. It was a relief to find that  our paperwork combined with Terry’s accurate sizing of window, door and stair apertures led to an exact fit for everything we had ordered. Terry had earlier shown his measurement skills by accurately pegging out the foundations, which was particularly tricky because the extension is at a lower level than the old building, and at a slight angle to it, so that it looks straight down the middle of the pond.

Our project has taken a long time, but we have no regrets about the delays, which gave time for reflection: as a result, we were able to change our minds for the better on a large number of issues, most notably the extension, where the original proposal just wasn’t ambitious enough. If we had charged ahead and finished the project in one go, it would have produced a much less satisfactory result.

047

Grey and pale yellow ochre colour scheme

Costs

Our revisions of the plan to add a much bigger extension meant that the project was much more expensive than we foresaw when we first took it on. The two big hits were the extra expense of the foundations and the loss of zero VAT rating for extensions to listed buildings in the 2012 budget, which affected phase 2 of the work. Our early  difficulties getting planning permission were also expensive and time consuming to resolve through the consultants, but it was worth every penny, or we could have had  an unusable farm service building on our hands. Overall, we spent about £1,275 a square meter in building costs before VAT for the conversion and extension taken together, which is in line with many of the examples given in Homebuilding and Renovating magazine. Including VAT at full rate on the later work, the average for the project was £1,450, which is on the expensive side, but then the product is a quality extension combined with a fascinating mediaeval building.

Here are some more photos.

And here is a link to a post showing what it looked like when we arrived.

Ready for landscaping

The finished extension after limewashing

The finished extension reflected in the pond, after limewashing

We have nearly completed our seven year project to convert and extend an old farm service building. We’ve made the minimum possible changes to the old building so we can preserve the history of its fabric, including the clay in its walls, while turning it into a comfortable home.

With a modern extension built on the footprint of the section of the farm building which fell down a few decades ago,  the shape and bulk of the ancient farmyard has now been restored. The 14th – 15th century farmhouse is along the north side and our conversion and extension along the west.

The fence in the picture below is a temporary one to prevent toddlers wandering, but normally the boundary will be left open to give a unified view down the meadow that runs away from the houses. There will be a stepping-stone path through the meadow where the boards (laid to take machinery) are now visible.

Our home on the left, the ancient farmhouse to the right.

Our home on the left, the farmhouse stretches to the right behind the tree with just the kitchen visible. Note how the new slate roof tones with the farmhouse thatch.

Narrow windows

Narrow windows are a feature of the extension

The texture of the wall of the old building, which is still mainly clay and straw, is in striking contrast to the smooth finish of the new section but the limewash on both helps to unify them. There is a lot more repair and preservation work to do to the old clay walls, but they’ve lasted for centuries, so we’re not planning to rush it.

The architects were Hollins of Framlingham, the extension was built and finished inside and out by Booty Builders of Laxfield, the piled foundations were by DJE Construction of Attleborough, and the engineers were Stroud Associates of Harkstead. We have been very pleased with the quality of the build and the design.

071

A single story hallway joins the old and the new

051

Photographed from the edge of the meadow.

Finished

Terry Booty’s last day on site, last job – the front step, below – the debris cleared, except for the last skip awaiting collection. A toast is most certainly required this evening.

IMG_2090.JPG
Plenty of interesting landscaping yet to come, not to mention plastering the other gable end of the old house and…. no, let’s not think about all that till next year!

Pros and cons of a breathable alternative to limewash – Keim

One of the conditions for listed building consent was that the colour of the extension must match the yellow ochre limewash of the old building, so the simplest way to do that would be to use the same limewash mix again. Hollins, the architects, instead specified Keim in the listed building consent application, a mineral paint which has been on the market for well over 100 years, and was developed in Bavaria as a more durable substitute for limewash. I’d never heard of it before, but a web search shows it is widely known.

Brushing on Ingleby's 3-coat tallow limewash

Brushing on Ingleby’s 3-coat tallow limewash

Keim is claimed to last 15-20 years at least, and the company says there are some examples 100 years old. The web site describes it as a liquid silicate paint using a potassium silicate binder with inorganic fillers (feldspar) and natural earth oxide colour pigments. Applied onto a mineral substrate, the binder is absorbed and forms a micro-crystalline silicate structure. This crystalline structure is said to allow the substrate to breathe but stops rain being driven in.

With a much bigger building, now including four gable ends, the idea of a paint that won’t need recoating every few years is appealing, as long as we can get the colour match. (I’m happy to carry on limewashing the old part, for authenticity). So we looked in detail into using Keim.

In the end, we went back to limewash. Here’s why:

The first three free samples from Keim, ordered from their colour chart, were close but not quite close enough. They then offered their free colour matching service if we could send samples. So we carefully chipped off a few square inches from the surface of the old building, wrapped them in plastic and sent them in a parcel to Keim in Telford, receiving a new free sample a few days later. This was a closer colour match to the sample, although its texture is not exactly the same as limewash and may have affected the apparent colour, making it look significantly darker and browner, and slightly less yellow than the limewash on a newly replastered wall on the old building. Keim suggested it might be because we had put the samples on without primer, so kindly sent us a sample of the fixer as well. We tried the paint after leaving the primer to dry for a day, but the colour was still too dark compared with the limewash on the old building. Curiously, when painted onto a completely impermeable surface such as plastic, the paint dried to a pretty exact colour match, but we couldn’t get it to do the same on lime plaster.

Throughout all this, Keim was tremendously helpful and speedy in its response, so I’d have no hesitation in recommending them from that point of view. The paint itself is easy to apply, and behaves on the brush rather like an emulsion. But in the end we went back to our regular limewash supplier, Ingleby of Glemsford, Suffolk.

Firstly, Ingleby charges £80 ex VAT for 20 litres of limewash made with an old Suffolk tallow* recipe to our own colour, Rodgers Flint, making an exact match. A 25 kilogram tub of Keim’s Soldalit to cover the same 60 square metre wall area is £383 including carriage, ex VAT, with extra for the fixative which has to go on first, and there is also the colour match to consider. The fixative for Soldalit is about £55 ex-Vat for a 5 kg tub, so with VAT the total would have been more than £500.

The price premium would surely be well worth paying in many situations because of the durability, which could delay the next big painting job for many years.  It would be particularly attractive if an exact colour match is not an issue (eg when the whole building is being painted).

Inglebys reckons that its tallow limewash should be good for at least 5 years and maybe 8 or longer if we don’t mind seeing it fade a bit (which with limewash is attractive). Certainly, their limewash has proved very durable over the last 5 years on our old building, apart from one area where we made the mistake of applying a couple of coats on hot, sunny days. Ingleby’s recipe produces a limewash that covers in only three coats, rather than the minimum of five traditionally recommended (and 8 to 10 according to some purists). Three coats are also required for Keim if the primer coat is included. The amount of work involved is not very different.

So with thanks to Keim for its help and advice, we’ve reverted to a traditional Suffolk paint for the extension.

* Tallow is very water repellent, too much so for some experts in building conservation, some of whom are suspicious even of linseed oil and prefer a simple mix using casein as a binder, because it is more breathable. This seems to be an area of lively discussion among the experts. We started with Ingleby’s tallow mix (called Traditional) on one wall of the old building, then switched to the linseed oil version for the rest, and have switched back to Traditional on the new building. It all looks great. See this link  for more on the subject.

Back to tradition – lime plaster

We’re using lime plaster to finish the outside walls. Lime is a fairly recent revival as a building material, and people are re-learning the old trades as they go, so there are still a lot of disagreements about the best way to do things. It’s not yet Lime Wars, but there are regular skirmishes.

Lime plastering the outside

Lime plastering a gable

As we’ve noted before, the Essex way (as taught on a course we attended near Braintree) was to use lime and sand as a hard render on the outside of a wattle and daub house. We repaired a wall like that, only to be told off a year later by a local Suffolk expert: the true vernacular coating in Suffolk was a plaster made from chalk, hair and lime, which is tougher and more flexible.

We couldn’t find much support at the time for this in publications of the Society for the Protection of Ancient Buildings, or indeed at a seminar of theirs we attended. The focus was on plasters inside and sand renders outside. But we found later that chalk lime plaster was very much backed as an external material by one of our local suppliers, Anglia Lime, when we went to see them at Belchamp Walter, which happens to be on the Esssex side of the border (but is not much more than a rifle shot from Suffolk).  Anglia Lime makes strong claims for the durability of their FibreChalk Mix (using artificial fibres) and Haired Chalk Mix (using animal fibres) on outside walls. We used their products for ceiling and internal wall plastering in the old building, and a neighbour says he is very happy with their fibre mix on his outside walls.

Starting to plaster a gable end

Starting to plaster the other gable end

When we removed the cement render on the gable end of the old part of our house, Alan Wilkins and Rory Sumerling used a chalk lime plaster with pig hair, mixed on site, to finish it (see earlier posts). Alan first suggested artificial fibres, but we wanted to stick to animal hair on the old building.

Alan couldn’t take on the new job, so Terry Booty introduced us to Keith Southey, another Suffolk lime plaster specialist.  He supplies a chalk mix using artificial fibres called Fibrelime, made from lime produced by the Lincolshire company Singleton Birch.  Suffice it to say that Keith didn’t see eye to eye with Anglia Lime on fibre and chalk mixes or with Alan Wilkins on pig hair, let alone with our one-time tutors in Essex. But we let Keith’s team get on with it (which they did very speedily), and the end result was a good, clean lime coating on the walls.

Why, you may wonder, did we opt for a traditional breathable lime finish on a modern watertight building? Surely a more durable modern cement-based render would be better, and more waterproof?

In fact, traditional lime makes a lot of sense on a timber-frame building, which is bound to show some movement. Lime plaster is much more flexible than a hard render, so less likely to crack. It also makes for a much more interesting surface texture that will be closer to that of the old building.

I have one more reason for preferring it: in ordinary British weather it is almost impossible to complete a building in the dry, and to the extent that some moisture will be trapped in the fabric, a coating of breathable  lime on breathable Savolit boards will allow it to escape. After the building is finished, lime will absorb more moisture than cement would, but it will also expel it again quickly in dry weather, rather than trap it (which is what happens when moisture gets through microscopic cracks in cement render).

A new roof over our heads

Tim and Andrew laying the Glendyne slates

Tim and Andrew laying the Glendyne slates

After the frame was finished and most of the insulation installed, there was a delay getting our Canadian slates on site, because the UK supplier had run out of stock. (For an explanation of why they had to come so far, see Continental Drift and the Art of Choosing Slates). Once the slates arrived, the shell of the building was quickly made weathertight, which included fitting the conservation rooflights from The Rooflight Company. These are solidly built, nearly flush and much favoured by conservation officers.

We took charge of ordering the other windows and doors. After extensive research over three or four months, including visits to showrooms and factories, we did detailed price and specification comparisons and settled on windows supplied by Kloeber – a British company based in Cambridgeshire with a showroom in London.

Slates done, Conservation Roof Lights installed

Slates done, Conservation Roof Lights installed

Ordering the windows proved a rather demanding process because there are so many detailed variations of specification that have to be thought through as precisely as possible – top, bottom or side opening, inwards or outwards, tolerances in the size of the openings in the walls, number of trickle vents, the precise colour (there turn out to be an awful lot of shades of grey), whether to have aluminium wood composite with aluminium on the outside, or all aluminium, and whether to have the manufacturer or the builder install. We chose wood and aluminium composite for the windows and all aluminium for the doors. (There was a a special offer of triple glazing for the price of double on the doors, so we snapped it up). The builders installed them all.

Roof done, barge boards painted, Kloeber windows and doors installed, more Celotex insulation fixed to frame.

Roof done, barge boards painted, Kloeber windows and doors installed, waiting for more Celotex insulation to be fixed over the frame.

With the windows and doors in, the rest of the insulation was fixed, and then the Savolit board on top. Savolit is very similar to the better known Heraklith (which is more expensive). It was fixed to the timbers using large plastic washers to prevent the screws pulling through. Both types of board are made from thin shavings of wood soaked in a cement mix which turns them into a hard material with many air voids. (You can blow through a Savolit board). They make a tremendous key for plastering.

Doesn't look much different - but the walls have now been clad in Savolit woodwool board, ready for plastering

Doesn’t look much different – but the walls have now been clad in Savolit woodwool board, ready for plastering

Savolit can be cut to shape easily. Where the outside edges of the boards cannot be screwed direct to the frame, the makers recommend gluing along the edges with a solvent free adhesive, but Tim and Andrew managed to fit all the board edges expertly to the frame without using glue.

A clser look at the Savolit boards - excellent alternative to laths for plastering and also an insulator,

A closer look at the Savolit boards – excellent alternative to laths for plastering and also an insulator

By this stage, the electrician and plumber had been called in to do the first fix, of wiring and piping. This of course has to be done before plasterboarding the internal walls and ceilings and screeding the ground floor. (The screed was laid on top of a layer of Celotex insulation).

We also had to pay a great deal of attention to the detail of the spiral staircase opening, because of the building regulation demand for at least 2 metres of headroom at every point on the stairs. This is not so easy when you come up under a sloping roof, as shown in the photo below. We in fact got a dispensation from our building inspector who said he would allow a lower clearance. But with the help of the staircase supplier and a lot of repeat drawings, in the end we managed to juggle the position and design of the spiral stairs so that they would fit with full headroom. It involved slightly changing the size of the opening after it had been made, by 150 mm.

Staircase opening on the first floor showing Celotex between studs

Staircase opening on the first floor showing Celotex between studs

Andrew screeding the floor

Andrew screeding the floor

The stairs were bought from  Cottage Craft Spirals  of Chapel-en-le-Frith, in the Derbyshire Peak District. This followed a long search among suppliers of spiral staircases, a large majority of whom were Italian companies exporting kits to the UK. We looked at at least 20 of these stairs, didn’t like any of the designs and thought they were also rather flimsy.

The steel and cast aluminium spiral stairs made by Cottage Craft Spirals just installed - the post has yet to be painted and the screwholes on the handrail will be filled.

The steel and cast aluminium spiral stairs made by Cottage Craft Spirals just installed – the post has yet to be painted and the screwholes on the handrail will be filled.

Cottage Craft Spirals makes everything in England, including the cast aluminium stair treads. They invited us to their factory, collecting us from Macclesfield station and taking us for a scenic half hour drive through the Peak District, so it was a very pleasant day out. Their stairs come in a variety of designs, can be tailored easily to individual specifications, look good and have a tremendously solid feel. They cost considerably more than the cheapest Italian kits, but no more than some of the allegedly bespoke upmarket Italian spirals we looked at, which didn’t look bespoke at all when we actually inspected examples.

Modern plasterboard and plaster skim inside

Modern plasterboard and plaster skim inside. This shows the shower room door.

With the first fix done, the internal walls and ceilings finished and plastered, the builders were ready to move on to the finishing stages. This included tiling and plumbing the shower room (actually a wet room as we dispensed with a shower enclosure and completely tiled it), painting, finishing the electrics, installing a third Conservation Rooflight in the hallway near the new front door; and finally hanging a lovely oak front door made of wide, solid planks by Jonathan Read Ltd of Framlingham (trading as Malcolm Neeve).

Next: a traditional plaster for the outside of a modern building.

New materials, old ideas

The more I look at the way our modern extension is being built, the more I am convinced that it is a fitting companion to the original 16th century oak-framed house. There is not a great deal of difference between the basic structural concepts, and only the materials and methods of fixing are new.

The biggest difference is that an old oak frame is mortised and tenoned rigidly with a few very heavy timbers while a modern softwood frame uses more but lighter timbers fixed with nails and steel hangers, and plywood sheets to increase rigidity. A modern construction can also use some big timbers, eg our glulam roof beam which runs the length of the building.

The timber frame takes shape

The timber frame takes shape

The roof timbers are added, with a Glulam central beam

The roof timbers are added, with a Glulam beam

The panels are infilled with insulation, with plywood panelling behind

The panels are infilled with insulation, with plywood panelling behind

The extension is a conventional modern timber-frame building. First, a wood frame was constructed piece by piece by Terry Booty of Booty Builders and his team, Andrew and Tim, on a low brick base; then they filled the spaces between the timbers with blocks of Celotex insulation, today’s equivalent of the hazel sticks and clay daub that fill in the panels of the old house.

The new timber is treated softwood, which I suppose is a lot less likely than oak to be here in 500 years time, but it is still a pretty durable material. The insulation is a hi-tech foam that takes us well above the new building regulation minimum, and will be much warmer to live with than clay. (One of the advantages of timber frame construction compared with blockwork or bricks is that it much easier to stuff loads of insulation between the panels). The fastenings of steel may not last as long as wooden dowels but they are cheap and easy to use.

Putting all those changes aside, if you stripped the old house back to its frame it would look quite similar, from a distance, to these pictures of the new extension. Its original builders would have been on site fitting individual timbers together, using different methods but with an objective – constructing a wooden frame to be infilled on site – much the same as Terry’s expert team, who made the basic structure in an impressively short time. The detail has changed, of course, but not the underlying idea of how to make it stand up as a rigid structure that keeps out the weather.

I am convinced that modern materials using an old concept are just as authentic as new oak framing for building an extension to a mediaeval timber building. In fact, it is hard to stop an oak framed extension with exposed timbers looking like a poor pastiche of the original. It can be done, but there are very few examples around Suffolk that manage it.

The first layer ofroof insulation is added and protected with a plastic membrane

The first layer of roof insulation is added and protected with a plastic membrane

Moats and beams

An excellent visit to a mediaeval settlement, organised by the Suffolk branch of the Society for the Protection of Ancient Buildings, threw new light on a puzzle about The Old Brewhouse and the farmhouse next to it: why are they outside the arms of their moat when you’d expect them to be inside?

As set out in the blog about how we reclaimed our pond (see this link), there are the remains of an obvious U-shaped mediaeval moat next to us. But it encloses a flat, empty space, now part off the next door neighbour’s garden.

Last Saturday, we were taken on a tour of Westhall, a Suffolk village where the original settlement has largely disappeared, leaving many clues to its existence, including pottery, raised platforms, moats, ponds, tracks and surviving buildings.

Two points stood out from the talk by archaeologist Peter Warner, who studied the area in great detail early in his career: houses were often moved, as happened at Westhall to Stradbroke Town Farm, a well-preserved house dating to the 15th century, owned by Chris Higgins, who hosted the event. It is probably not on its original site.

Coincidentally, William Clement-Smith, an oak frame specialist working on The Old Brewhouse, told us a similar tale about Ivy Lodge Farm nearby: investigating the structure for a potential purchaser, he found it was two individual frames from different periods joined together in a way that made it seem – to a craftsman like him – that the newer beams and studs had been bought second hand, reassembled on the site and joined rather arbitrarily to the old building.

The second point made by Mr Warner was that sometimes the farmhouse would be outside a moat because another more important building, the manor house to which it belonged, was inside.

Our Spab group of a dozen people struggled up through brambles, (entertained to strident protests from a householder who disputes what we all thought was a very obvious right of way!) to a large almost hidden moat. This enclosed a substantial mediaeval farm that was abandoned in the 1940s because its land was taken for a wartime airfield. The ruined timber frame was visible in the 1970s, but there is now no obvious sign of the building above ground. So a house abandoned centuries ago might be much more completely obliterated.

This all seems like good circumstantial evidence for the suggestion we have heard that our own moat did once enclose a mediaeval building, which was moved or has disappeared. It seems it is quite easy to up sticks, literally, and cart the posts and beams of a 16th century house to the other side of a moat. One reason for that could be the slightly higher ground on our side.

If would be just as interesting if there was once a manor house on the other side, between the ams of the moat. Not sure, however, that the neighbours will be up for excavations to find out more!

Below : Peter Warner describing how the apparently featureless fields were full of pottery and other traces of the old Westhall. The track just visible through the wheat is a modern right of way to St Andrews, Westhall’s lovely village church.

20140714-141457.jpg

Foundations

After several redesigns, the engineers settled on a system of 12 piles and a reinforced slab. It was a bit of a saga getting to that point, because although we are next to a pond, there was nevertheless more water in the trial borehole than the experts expected.

Auger drilling an 11 metre test hole.

Auger drilling an 11 metre test hole.

The initial plan by the contractors was to auger the piles, to avoid using a pile driver near an old building. However, this was vulnerable to water inflows, which proved excessive,  which would have made it hard to be sure the concrete would set properly.

So we were advised to switch to steel-cased piles, with the first three metres augered. With hammered piling starting three metres below ground, it was hoped that the vibrations in the old house would be reduced. This proved the case, though we checked the house carefully every hour or so to make sure there was no damage.

The best way to test the effect was to hold a hand lightly against a wall. In fact, we were rather more worried about the brick-lined well, which we also checked regularly, because it is less than 10 metres from the nearest pile. There was no sign of damage. In the end, most of the piles were less deep than we budgeted – under 7 metres instead of 10 metres – because they quickly went into the hard grey boulder clay, which is very strong and stable. One pile went to 11 metres.

Pile driver and steel piles, before they were filled with concrete and trimmed back.

Pile driver and steel piles, before they were filled with concrete and trimmed back.

Other technical complications meant constant revisions to the foundation plans, but in the end the work was signed off by the building inspector, ahead of pouring the concrete for the reinforced raft.

The contractor was DJE Construction of Attleborough and the consulting engineers were Stroud Associates of Harkstead, Ipswich.

Steelwork for the raft in place, ready for the concrete.

Steelwork for the raft in place, ready for the concrete.

Continental drift and the art of choosing slate

What’s the connection between Canadian slate mines and Snowdonia National Park in Wales? The answer is that 500 million years ago, before continental drift formed the Atlantic Ocean, they were in much the same place.

What has this got to do with extending a Grade II listed house, you may well ask? Quite a lot, it turned out: by last week we had obtained all the listed building consents required before breaking ground on the extension foundations, bar one; the remaining condition was the source of the roofing slate. The strong preference of our local authority was for Welsh slate, because that was what used to be used in East Anglia. Our request to use Spanish slate did not go down well.

Welsh slate is, however, hard to get and very expensive – five to six times the price of Standard Spanish slate, with long waiting times for delivery. Furthermore, a visit to a specialist supplier established that one of the very few easily available Welsh slates, from the Penrhyn quarry, has a blueish tinge. So the supplier, Cembrit, suggested a Canadian slate called Glendyne, which is approved for use by the Snowdonia National Park authority for use within its own borders as a substitute for the well known dark grey Ffestiniog slate, which is no longer mined. If it can be used in Snowdonia, why not Suffolk? It is pricey, but still less than half the cost of Welsh slate.

We delivered samples of Glendyne to the planners. We were taken aback by the reply: it was acceptable if we could confirm that it came from the same geological formations as Welsh slate.

Now it so happens that many years ago I did Part I of my science degree in geology and mineralogy, before switching disciplines. It was so long ago that our lecturers were the very same scientists who had finally proved the reality of continental drift, more than a century after the theory was first proposed (in 1858).

With a combination of geological and slate producer websites, not to forget an old university geology textbook that I had kept for decades on a bottom shelf, I was able to confirm that the Glendyne slate quarry, on the Quebec-New Brunswick border, now 3,000 miles from Wales, was from the same Ordovician formations as Ffestiniog. They drifted apart hundreds of millions of years ago. (If you’re after a substitute for Penrhyn, that turns out to be from the even older Cambrian era, and is a match for Newfoundland slate. But that does not seem to be sold in the UK).

The answer was given, the permission came, and the foundations were started.

Replacing the part of the house that collapsed.

20140121-163248.jpg

The cleared site ready for work to start

As described in an earlier post, we are now about to rebuild the end of the house that rotted and fell down 40 or more years ago. The first step, which took up much of the summer and autumn, has been to complete the main repairs to the old building, especially the gable end where the new structure will join.

The new one and a half storey building is not a copy or pastiche of what was there before, but it is roughly the same size and shape. That means the overall bulk will be as it used to be before one end collapsed. It adds 55 square metres of floor space to an existing house of just over 100 square metres. The architects and planning consultants are Hollins of Framlingham, who also oversaw the conversion of the Grade II listed building.

There used to be doors at ground and first floor level between the two parts of the building, and they were joined completely. We plan to connect the old and new only at ground floor level with a narrow, relatively lightweight, single storey entrance hall. By avoiding elaborate joins, the two structures will be able to move separately, which makes sense because the old building has shallow foundations and the new will have reinforced concrete piles as much as 8 metres deep. Piling is necessary because the site is perched on the edge of the pond.

The piles were initially planned to support reinforced concrete ground beams, but this was changed to a reinforced concrete raft supported by piles. This avoids filling in part of the pond, because it needs a smaller working area for the machinery.

The quotes are in for the piling. Time to make a decision!

20140121-162543.jpg

Hemp insulation

Early on, we filled panels where the clay had disappeared with new wattle and daub. This year we have switched   to a new plan: these panels, which had been filled in the last 30 years with Thermalite blocks and broken bricks, have been made into a breathable hemp sandwich.

We lined the outside of the building with Savalit woodwool boards (as described before), then plastered them with haired chalk and lime mix bought from Anglia Lime. We used battens to fix Savalit boards on the inside, recessed just enough to take the plaster, and gave them a thinner layer of chalk and lime plaster, also with hair. In fact, internally, Savalit boards can take an extremely thin, almost skim coat of haired lime plaster and still look good.

In the middle we put  two 75mm layers of natural hemp batts, which have exactly the same insulation rating as wool but absorb less moisture. The hemp is also grown in Suffolk, and processed in Wales by Black Mountain (though  they had a fire that stopped production in summer 2013).

We attended a course by the authors of The Old House Eco Handbook, published in collaboration with the Society for the Protection of Ancient Buildings. Marianne Suhr, co-author of the book, gave three very practical suggestions for the panels, in a private conversation: Pavaflex wood quilt, then oak laths and a coat of plaster, which she said would be thermally the best; next best would be a wood fibre board to fill the whole panel, which has the advantage that it can be plastered direct; finally hemcrete, which can be applied behind shuttering and left to set, and then directly plastered. Pavaflex sounds good, but the laths would be a fiddle.

However, when we looked in detail back home at prices and the amount of work involved, we opted for the local suggestion of Savalit boards and hemp insulation batts, which the conservation officer approved.

These pictures show the method in more detail, for those who want to try a hemp sandwich:

Some battens had to be fitted to irregular wall sections

Battens were fitted, some to irregular wall sections

Boards were cut in short sections across their length (easier to fit, reduces wastage) and hemp put behind

Boards were cut in short sections across their length (easier to fit, reduces wastage) and screwed on with large washers. Board edges were joined with solvent free adhesive. Hemp was put behind

Two 755mm layers of hemp fitted in without compressing it. The material was easy to shape.

Two 75 mm layers of hemp fitted in without compressing it. The material was easy to shape.

Similar method of finishing the new ceiling and joists (installed by previous owners). No need for breathability so used Rockwool slabs and compressed it by more than half by installing the Savalit very tight against it, to make a good acoustic insulation. Hemp was used near exterior walls.

Similar method of finishing the ceiling above the main living room (where new floor and joists installed by previous owners). No need for breathability so used Rockwool slabs and compressed it by more than half by installing the Savalit very tight against it, to make a good acoustic insulation (Thanks Georgia, for the acoustics advice). Hemp was used in ceiling abutting  exterior walls for breathability.

On left, old limewashed wattle and daub, centre Savalit boards ready for plastering, right, after plastering.

On left, old distempered wattle and daub, centre Savalit boards ready for plastering, right, after plastering.

x

Oak and other problems under the old render

All new plaster, limewash, oak pentice board, soffts and bargeboards.

All new plaster, limewash, oak pentice board, soffits and bargeboards.

When the cement was removed, we found some repairs were needed for the timber frame, including a new stud where an  old one had disappeared (judging by the empty mortice), completing another stud that had rotted away for the first foot above its mortice in the sole plate, a knee to reinforce the joint between a main internal beam and a post, and a  strengthening of one of the corners, where a sound main corner post did not seem very well tied in to the sole plate. It was reinforced with a flexible glass fibre rod through the timbers and a steel plate. All the timber work was carried out by William Clement Smith.

The corner post needed tying in better to the sole plate

The corner post needed tying in better to the sole plate

He also installed a beautiful oak pentice board above the gable wall window and another lower down on the chimney return wall, to deflect drips. The barge boards on the gable end wall were beginning to rot, so he replaced them with oak. The soffits had to come off anyway for the lime repairs, and new oak ones were also installed.

Other issues also came up, which always happens when you start removing things. A small side roof by the chimney (a couple of square metres) proved to be badly fixed to the wall when the cement render was removed, and once several rows of tiles had also been removed to fix this it became clear that it would be a good time to re-roof it, and also to put in some insulation and something to carry the water away from the back wall, where it is causing a damp patch. It was in any case on the repair list for the longer term because it leaks badly. A new oak knee, made from a naturally curved piece of timber, was installed to reinforce one of the main load bearing beams, which did not have a good joint where it met the wall. An existing rather poorly attached knee, which supports another main beam, is being replaced with a wrought iron bracket, and the front and back door frames and doorsteps, which are rotten and let in the rain, are also being repaired.

Limewash

We decided to use a simple limewash mix of lime putty and water (one to three, and one to six for the first coat ). This was intended to be a breathable base layer. It also has the advantage of being very cheap, and it can be stirred by hand with ease.

We put on four coats  with one or two days between to give an opportunity for the limewash to cure. We then used three coloured topcoats – the first coat more dilute – of Ingilby’s linseed oil limewash, making seven coats in all. If we had used 7 coats of linseed oil limewash we would have had a much less breathable coating. As the gable end wall needs scaffolding to paint, the hope is that the extra coats will make it last longer.

Removing cement render – 2013

…. in preparation for the extension

Repairs on a building this age are never really finished, but there is a list of essential jobs we want to do before starting on the extension, for which we won planning permission in late 2012. The first step is to finish the external repairs to the end of the old building where the extension will be built. This means removing a large amount of cement render.

  • Alan Wilkins, a lime specialist, has been commissioned to remove the cement and plaster the south gable wall and the return to the chimney. The plastering is being done in close cooperation with Paul Harrison, the conservation officer.
  • The first stage was to make four exploratory cuts in the old cement render to inspect the condition of the timbers and materials underneath, which was generally good.

    Four exploratory holes

    Four exploratory holes

  • The rest of the cement was cut off in small pieces with a grinder, using towers at first though later we used proper scaffolding, for safety. All but one of the ground floor panels proved to be filled with thermalite blocks, which were removed by Alan, who worked with another historic buildings specialist, Rory Sumerling. Four smaller panels on the first floor were similarly removed. Only the thermalite blocks filling the old first floor doorway were left in place because they support the window frame.

    Much of the old cement render is off, the clay on the upper level is being exposed, and the thermalite blocks on the ground floor have been removed.

    Much of the old cement render is off, the good clay on the upper level is being exposed, and the thermalite blocks on the ground floor have been removed. The first floor doorway into the demolished section of the house is now occupied by the window and thermalite blocks.

  • Lead flashings were also put in above the brick footings. They are needed because the wall leans inwards and water running down it has been running in over the top of the bricks.
  • Paul Harrison has agreed that we can use breathable Savalit boards (a similar alternative is Heraklith) made with ‘wood wool’ instead of oak laths. The argument for this is that the clay at first floor level is largely intact and covers the frame. It does not seem to have had laths fixed in the past and has a uniform almost smooth surface which would be better protected by boards than laths. The boards can be screwed on without as much damage. They also add a small amount of insulation. Plaster can be applied direct to the surface. And the clinching argument is that they are easily removable ie the process is reversible and does not permanently alter the historic material.

    Clay in good condition from first floor upwards.

    Clay in good condition from first floor upwards.

  • We have accepted Alan’s advice that chalk, hair and lime plaster is the best option for the external surface. He advises that it is more authentic locally than sand and lime render. The mix is three chalk to one hydraulic lime (NHL2, the softest type.) Although some say lime putty should always be used, because hydraulic lime contains some clay and is a bit less breathable, the argument for this recipe is that in practice most lime in our region would have had clay in it like hydraulic lime, it cures faster, and on a gable end wall the extra strength is worth having. It is also easier to mix.

    A clay and hazel panel, next to thermalite blocks still in situ

  • The new plaster was also limewashed with four coats of a simple and cheap mix of water and lime putty (three water to one lime), which is highly breathable. We then applied three coats of limewash with linseed oil, which is slightly less breathable, but more water repellent. It was mixed to our preferred pale yellow ochre (labelled Rodgers’ flint) by Ingilby’s of Glemsford. Alan said afterwards that three coats with linseed oil, even diluted, reduces breathability too much, so perhaps we should have stuck to two.

Moving on – chalk and woodwool boards

 Our thinking has moved on from when we started work in 2009 and 2010. One advantage of repairing a house very slowly is that it gives time to learn as you go along.

This applies particularly to the question of which materials to use when repairing walls and infill panels and when plastering and rendering, and to the question of energy efficiency.

About a third of the old clay in the building has disappeared and been replaced with modern materials such as thermalite blocks. We started in 2009 and 2010 by putting back new wattle and daub where there were gaps, and much enjoyed doing this. But more recently we have been encouraged to use modern breathable insulation made from hemp fibres in the body of the wall to improve the thermal performance of the house.

After attending a course in June 2013 on improving the energy performance of old buildings, I came away with the impression that the expert pundits are also looking at a wider range of techniques than they used to. They certainly sound less purist than a few years ago, when there was a strong encouragement to use entirely traditional material for repairs, mimicking what was there before, and not to trust anything new. Now the talk is much more of using thermally efficient substitutes that will improve the energy performance, though only where it can be done reversibly, and without short or long term damage to existing materials or structure.

In 2013 we applied this new thinking when we re-rendered a gable end wall and part of a side wall which were covered 40 or 50 years ago with thick cement on wire netting with a plastic sheet underneath. After conversations with Anglia Lime, with a neighbour who has re-rendered a similar cottage and finally with lime specialists Alan Wilkins and Rory Sumerling (formerly of Anglia Lime), we have used lime plaster with chalk and hair as our external render to protect the clay daub, instead of the lime and sand mortar we used in the first stage of repairs. And instead of laths (which don’t appear to have been used on our building anyway) we are using breathable Savalit boards, which can be plastered direct. The result is hemp insulation sandwiched between Savalit boards.

Chalked lime is softer and more pliable and seems to be more resistant to movement cracks than lime and sand render. Our neighbour made a number of test panels which showed that the plaster weathers well. It is considerably stickier than the clay daub  and lime plaster we were using before, and does not need quite such an effort to work it onto a clay surface, which is a major advantage. And there is evidence that in our part of Suffolk chalked lime plaster was historically the main material used, rather than lime and sand render. So we may well abandon sand and lime renders altogether.

Wonderful clay

The grey, chalky pond clay makes a great building material

Pond cleaning in 2011: the pale grey, chalky pond clay makes a great building material. (The darker areas are pond mud on top of the clay).

Postscript on wattle and daub – 2013

We have a great resource in the pond, which we knew about from when we drained it in 2011, but we had not appreciated until this year its amazing building material potential: pale blue-grey boulder clay, which underlies much of Suffolk.

It seems to be made up largely of fine chalk interspersed with chalk pieces of varying sizes. When wet at the bottom of the pond it is extremely sticky but remains firm enough to walk on without sinking in. When dry, it is as hard as a lump of chalk. We were told by an engineer that in our part of Suffolk the boulder clay would typically be 25 metres thick. We are lucky, in that it comes almost to the surface in the garden, and its impermeability is the reason why we have an ancient pond. We had already found a lot of it used in the old wattle and daub panels.

We discovered its quality when making minor repairs  this summer to the other walls (not the one from which the cement render was stripped).  Some of the boulder clay was put aside when the pond was cleaned in 2011. We pounded it in a bucket for 20 minutes (for larger quantities it would be a builder’s bath) until it became plastic. Then it was mixed two to one with sand and lots of chopped straw added (in fact dried wild barley stalks, growing in the meadow).

This shrank much less when it dried than the yellow Essex clay daub we bought in 2009 and 2010 (for some major repairs to one of the walls and for filling gaps in the panels inside). Our pond clay is a much better repair material for damaged wall areas, including  holes and cracks, and it may also make a good clay-based render – to be tested. Memo to self: when the new foundations are piled, keep any boulder clay that comes up with the augur.

Instead of rendering or plastering, this time we just limewashed the clay. We had noticed areas of clay in the walls that showed no signs of old render; instead, there seemed to be many layers of old limewash flaking off. Perhaps we  will carry on repairing the other walls with just clay and limewash, and skip the plaster stage.

Techniques

Looking around a typical Suffolk village, houses repaired with lime are still quite rare. It is clear that many homeowners and builders have yet to be persuaded of the vital importance of breathability and the use of traditional materials on an ancient structure. There is clearly a huge backlog of work to be done and, judging by the stories we hear from time to time about the terrible problems found in buildings that were repaired thoughtlessly in the 20th century, the damage is mounting right across East Anglia’s huge stock of historic houses.

There is a problem, however, when you start digging into the detail. Even among the experts in these old techniques, there is plenty of disagreement, and sometimes it gets quite heated.  As experience grows again, after decades in which some skills were in danger of being lost, ideas are bound to evolve further.

So rather than try to pass on definitive advice, these posts have been setting out some of the techniques we used – or hired experts to use – and we also note where our own ideas and the advice from experts have changed over time, for example our decision to switch from the sand and lime render we used in the first year of the project to chalk and lime plaster for later phases. 

More generally, it is important to read and discuss, to take advice from more than one source and to think through the jobs yourself with an open mind, focusing on the particular characteristics of the building concerned. The key points are:

  • always protect old material, and never do anything to it that isn’t reversible
  • in the case of infills and coatings, never put in anything that isn’t breathable.
  • and of course, put some effort into learning about the techniques you decide to use (or hire contractors to use for you).

We win planning permission to extend

It has taken much of the year, but we have finally got permission to replace the old part of the house that collapsed several decades ago, which will add just over 50 per cent to the floor area of the building.

If you look carefully, you can see the outline of the vanished end of the building marked on this recent Ordnance Survey

The outline of the section of the building which fell down – the part closest to the pond –  is still shown on this recent Ordnance Survey map

One important factor was to demonstrate with photos and archives that there was a complete building on the site only a few decades ago, and certainly post-1948, which seems to be an important date.  I may have misunderstood it, but the gist seemed to be that the old structure legally still had some vague sort of existence, so that we were in a sense rebuilding, and giving back the farm complex (of which our house is a part) its previous layout.

This began as a project to convert into a home a Grade II listed farm service building that had the same legal status as a barn and which did not even have planning permission when we bought it. Now  we’re into a far more ambitious affair, essentially a self build of a home with three bedrooms, two reception and a kitchen diner, spread over a very long period.

It has been more than six years since we bought the property, nearly five years since we started work and more than four years since we moved in. We still aren’t going to rush, and do not expect to finish the new-build section for another two years. Our project is in fact characterised by its extreme slowness!

But speed doesn’t matter – indeed it is a great advantage in some ways to move at a snail’s pace: if we had built the original small extension for which we had permission at the same time as we repaired the old building then we would have closed off the new options that have emerged. Proceeding slowly also gave use time to restore the pond and landscape the garden, which in turn led to the idea of the new extension, whose centrepiece will be the ‘pond room’ which overlooks and almost overhangs our 250 square metre pond.

The contrast with 2007-8 could not be greater. Then, we had difficulty persuading the council to accept an enlarged porch, which was just big enough to squeeze in a shower room. Now we are adding two substantial rooms and a lobby.

We used the planning consultants and architects (Hollins of Framlingham) who won us permission in 2008 to convert the old building into a dwelling, and they dealt with the new design and application very efficiently and effectively. They organised a pre-meeting with a Mid-Suffolk planner and the conservation officer, who fed back some useful and interesting suggestions, which we accepted, before the final drawings were submitted.

The result is a design that will be modern in its detailing – including powder coated aluminium windows – to avoid the look of a pastiche. But it will have enough similarities to the listed building to harmonise with it. In particular, the roof pitch will be the same and we will use lime render and limewash for the walls.

We also decided to build the extension (or replacement, if you like) as a completely separate structure joined to the old building by a new lightweight single storey lobby, to allow the old and new to move independently. One will have deep piled foundations and the other is still on brick footings lying on clay. We arrived at this aspect of the design ourselves by playing with Google Sketchup. We had successfully used a separate building with a light connecting corridor for an extension to a previous house, which had been built in 1830 of bricks resting on clay.

In fact, with the new project, if we join the new to the old at first floor level, through the old doorway that still exists in outline, we would not add much to the usable area of the house. One of the old bedrooms would become a corridor from old to new. We would rather put up with the  small inconvenience of a second staircase.

Because the site is tightly defined by the edge of the pond and the floorplan of the previous structure (the footings are still there, with a couple of the sole plates), compromises had to be made on the layout; in particular we will need to use a spiral rather than a standard staircase, and the stairs and the shower room both take space out of the pond room.

But… there are still a lot of repairs to do to the old house, so we’ll get to grips with those before we start work. That will give us some time to think about the extra money we have discovered we may have to spend on the foundations: the engineering consultant said they must be piled, because they are close to our pond.

We also need to take time to think about how we manage the project: do we get a single contractor in, do we manage subcontractors ourselves, how much DIY do we want to undertake? There is quite a lot of useful information in self-build books and magazines, particularly Homebuilding and Renovation.

We have two outline estimates from single contractors, both way above our initial cost idea, taking it to over £2,000 a square metre, far more than we want to pay. The current thought is that we break the project down into four: finishing the old house; building the piled foundations (where a main contractor would be hiring the same specialists to do it); building a watertight shell; and finally all the finishing work including plumbing and electrics and rendering the outside in lime, all of which could be done by sole trader local specialists who we have found reliable during the repair stages. They are too small to be VAT registered, so at least their labour charges would not be taxed. We will do as much DIY as we can cope with, and have the skills for.

To be continued!

2012 repairs – pond, well, patio, drainage

New brick patio, with the well protected by a steel over with limestone slabs on top.

New brick patio, with the well and electric pump protected by a steel cover with limestone slabs on top.

  • The remaining pond mud, piled in the garden – perhaps 80 tons – was removed and used for landscaping adjustments, and grass was laid.
  • The well was properly capped with a steel cover, screwed down, on which limestone slabs were relaid.
  • A steel structure was built by the same people to make loft access safer.
  • A brick and stone patio was constructed outside the kitchen with a channel for the power supply and hose of an electric pump for topping up the pond. The flow rates were tested. At the worst of the 2011- 2012 drought, when the well was at the lowest level we have recorded, it produced more than 2 cubic metres of water a day which would be adequate to top up the pond in a summer drought. At other times its productivity was much higher.
  • The shallow trench round part of the house, which had been dug roughly to prevent damp earth lying against the brick footings above the inside floor level, was redone more carefully. A low brick retaining wall was built, the bottom of the trench the was lined with geotextile, and several inches of gravel placed on top. It stopped the damp coming through the lower bricks and discolouring the limewash on the inside of the brick footings.

 

New stairs and a restored pond – 2011

  • Felix Oliver, our next door neighbour, replaced the old stairs with a new oak staircase to a similar pattern but much better finished. The design was negotiated with, and approved by, building control. The result has been widely admired. It fits perfectly at the top to a very uneven sloping floor. Felix is a professional wooden boatbuilder as well as a specialist in oak-framed buildings – see this link to  Suffolk Timber Frame Buildings – and his boatbuilding skills show. He also installed 3 new oak studs to replace the rather agricultural – and recent – softwood posts that had been there before.

    New oak stairs

    New oak stairs

  • Insulation was put in the roof above the north bedroom, bathroom and landing, but only on the flat surfaces – the side sloping surfaces were too difficult to access. Part of the the loft area was then floored so it could be used for storage. We discovered that modern polystyrene insulation had been laid under the tiles of the roof when it was redone about 30 years ago and it seems to be intact. There is still a substantial airgap round the rafters, which is reassuring.
  • The south and east walls were limewashed, pending full re-rendering. All the external window surfaces were oiled with linseed. We couldn’t decide what colour to paint them and linseed oil seems a very good wood preservative and water repellent, so we may just keep coating them for a year or two.

    Limewashing

    Limewashing

  • The pond was cleared, removing hundreds of tons of mud, using it to landscape the garden and shrink the overall size of the pond back to where it probably was before the banks were broken down – see the blogroll link to the pond blog for all the details. This took from July to nearly the end of the year so very little was done on the house during this time.

Limewash

Limewash

The wall repaired in 2009 was limewashed the following spring with a colour made up to our specification by Ted Ingleby, the well known traditional paint manufacturer of Glemsford, Suffolk, which he called Rodgers Flint. We asked him  to match the very attractive colour of the render itself, which came from the sand we used. (We took it to him painted onto a flint). The recipe is on his file and we have continued to order it.

There turned out to be a bit of controversy over grades of limewash – one for the those into the detail of repairing old buildings. Ted is very keen on an ancient Suffolk recipe using tallow, which he markets for outdoor use. It can cover a wall effectively in three coats rather than five and is very waterproof, and has considerably better coverage and durability than some other versions of limewash. But various books we consulted advised against tallow as not breathable enough, and suggested linseed oil was better, or even plain limewash in sheltered positions. Three of our walls are very sheltered.

The urgent repairs finished. The darker area is cement render that has yet to be removed.

We decided to add new coats the following year, 2011, using  Ted Ingleby’s interior limewash, with linseed oil, on the outside. This has kept its full colour and has not deteriorated after two years. We also use his special pozzilime mix for non porous surfaces, which has tiny glass spheres in the mix, to paint the old cement render. This old render is to be removed, but not yet.

Inside the house, all the panels and the exposed brick footings on the ground floor were painted with an off-white Ingleby limewash, containing linseed oil. (The colour was called ‘half tint’).

Wattle and daub – inside repairs

There was some damage to the inside panels of wattle and daub during the replacement of the sole plates. The building contractors re-rendered the bottom 30 – 50cm cm or so of the outside of the wall to weatherproof it over the narrow gap that appeared, using lime mortar. But on the inside there were a number of breaks and gaps in the clay where the old panel infill fell out when the building was lifted. Some thermalite blocks, which had been used in previous repairs, also fell out.

Fitting hazel ledgers and sticks to an empty panel

The empty panels were filled from the inside with clay daub. After letting the work dry and crack for six weeks we put on a clay plaster made from daub mixed 2:1 with lime mortar and finally a thin coat of haired lime plaster bought ready-mixed from Anglia Lime. This was also used to plaster the thermalite blocks which had been used to fill in the old kitchen doorway, and was very effective in strengthening a section of old crumbling plaster in the kitchen.

Not strictly wattles – which are woven – but a tied hazel lattice, as taught on the Essex clay daub course.

There were two completely empty panels and a number with large holes. In all these cases, hazel ledgers were fitted to notches made on the studs and hazel sticks were tied to the ledgers with hemp string. Clay daub was pushed into and around the hazel. If these had been completely open panels, then it would have needed two people to do the job,one inside and one outside; because the outside render was intact one person could manage the infilling.

We used clay even where the broken panels proved to have been filled with thermalite blocks, on the suggestion of the wattle and daub instructor, who said this might give an early indication of whether moisture was building up behind the panel – the clay would absorb water and quickly discolour if it got damp again.

Pressing clay daub into the wall

Repairing a damaged panel

The clay inserts under the modern blocks were mixed 2:1 with lime mortar for added compression strength because they were supporting a substantial weight of material. Once dry, a clay and lime mortar mix appears to make a very strong block. In fact, at least one nearby farm outbuilding is built of clay lump – dried blocks of clay daub – without any supporting timbers. The clay blocks were very visible because of the deterioration of the wall, though it has recently been repaired and lime rendered.

Making clay infill, supported by hazel, under sections of damaged Thermalite blocks.

More repairs – 2010

  • Chris and I filled the very large number of holes left in the kitchen and living room by the replacement of the sole plates, including some complete damaged panels. Wherever possible, we used hazel lattices and clay daub for the repairs, with a skim of lime render inside, finished with limewash. We salvaged old clay and plaster. See ‘wattle and daub’ and ‘wattle and daub inside’ posts.
  • The interior ground floor woodwork was painted and the brickwork limewashed.
  • An electric storage heater Aga was installed in the kitchen.
  • We deliberately spurned a fitted kitchem. The idea is that separate pieces of kitchen furniture unattached to the walls allow the walls to breathe and stay dry – important given our repair philosophy of using the natural characteristics of the materials to keep the building dry through encouraging breathability.
  • We paid two local landscapers to lay hardcore and a stepping stone path from the entrance to the front door, and to make a small brick patio outside the front door, linked to the doorstep by limestone slabs on a brick plinth. They also made a low brick wall to hold back the slightly higher level area outside the kitchen window, which we gravelled. They erected a garden shed we had bought.
  • A fencing contractor installed a full size five bar farm gate. We later replanted the damaged hedge either side of the gate with hawthorn and field maple.

First stage repairs – more photos

PHOTO GALLERY

Before work started - plastic to keep the woodpeckers off. 2008.

Before work started – plastic to keep the woodpeckers off. 2008.

Preparing to insert the flitch beam.

Preparing to insert the flitch beam.

Inserting the new sole plate in the kitchen, early 2009.

Inserting the new sole plate in the kitchen, early 2009.

Connecting the power, early 2009

Connecting the power, early 2009

Waste pipe from the first floor - early 2009.

Waste pipe from the first floor – early 2009.

Installing the bath, spring 2009.

Installing the bath, spring 2009.

The temporary window - to be a door later.

The temporary window – to be a door later.

Excavating for the power cable, early 2009.

Excavating for the power cable, early 2009.

Sole plates and new footings done, getting ready to lime render the gap, using a permeable membrane.

Sole plates and new footings done, getting ready to lime render the gap, using a permeable membrane.

The fireplace after the floor was excavated, early 2009.

The fireplace after the floor was excavated, early 2009.

New sole plates in the corner of the living room.

New sole plates in the corner of the living room.

Installing a ring main before the floor was laid.

Installing a ring main before the floor was laid.

The chandelier installed in the main bedroom, 2009

The chandelier installed in the main bedroom, 2009

Loon and cabin for the builders 2009

Loo and cabin for the builders 2009

Materials for the limecrete floor

Materials for the limecrete floor, early 2009

Protective covering on the back wall after the footings done - to keep the woodpeckers from the clay. Early 2009.

Protective covering on the back wall after the footings done – to keep the woodpeckers from the clay. Early 2009.

The brewing hearth and the door to the demolished part of the building.

The brewing hearth and the door to the demolished part of the building.

Excavation nearly done.
Excavation nearly done

Further work – 2009

 Further work 2009

Summer, autumn, winter

Celebration - moving in!

Celebration – moving in! May 29, 2009.

  • The east wall was repaired with clay, but in patches where there were holes, and on the north wall there were also areas that needed repair.

    New Morso wood burning stove

    New Morso wood burning stove

  • We planted a beech hedge on the north boundary of the kitchen garden.
  • A Morso wood burning stove was installed by us in the sitting room. It has a rated output of 10Kw maximum, though we were toild to take all these measurements with a pinch of salt. It does however create a lot of heat that spreads throughout the building after a few days, as the massive chimney brickwork warms.

Wattle and daub

We learnt the basics of wattle and daub on an Essex County Council course at Onchors Farm near Braintree. The course tutors were Lydia Bucknell and Peter Roe of Traditionally Plastered. This was a prelude to a long term programme of repairs to the daub, inside and outside.

The first thing to think about is a supply of clay daub. On the course, we learnt to make it ourselves. The basic ingredient in that part of Essex was a pale yellow clay, containing small chalk pebbles, from a pit on the farm, which was quite  similar in texture to the material in the walls of our cottage.

This was was the kind of repair we had to tackle – deep damage to the clay panels.

At Onchors farm, the clay was mixed with sand, chopped straw and cow manure and then trodden (literally) until the texture changed to plastic and doughy, not unlike plasticine. It took 3 people 30-40 minutes to tread one large wheelbarrow load into the right condition. Traditional builders used to short circuit this laborious process by leaving cows tethered in the clay pits to tread the mixture. Modern machinery can be used to make daub in quantity, but do not use cement mixers, because the material is too glutinous. Onchors farm has a machine very like a large bread doughmaker, with rotating arms. Given our experience of DIY daub making, we decided to buy ready made daub from the farm in tubs.

There are many recipes and much variation in advice on this subject. We used a straight daub mix of clay, sand, manure and chopped straw for filling deep holes or complete panel replacement. For shallow repairs and a base coat for render on top of the daub, we used a mix of 2 buckets of daub to one of lime mortar. The mortar was mixed from 3 buckets of sharp sand to one of mature lime putty.

The lime putty can be made up from slaked lime, but this is a messy process and the putty should also be stored for some months and preferably a year or more to mature it, which improves the consistency. Matured lime putty can be bought nowadays from ordinary builders merchants as well as specialist suppliers of building materials for old houses. Our supplies came from Travis Perkins.

The tools required are basic: for the daub, a plastering trowel, preferably one with a rounded end, a bucket and mixing spade, rubber gloves, and a hose with a fine nozzle or – preferably – a hand-held spray gun for dampening the old materials before applying new; for the wattles, a half inch or three quarter inch chisel and a mallet or hammer. We were taught how to split the hazel sticks with a frame and a special tool, but when investigating our house we found the builders had used unsplit sticks, so we did the same.

Old clothes or overalls are of course vital, because daub gets everywhere. Where the daub is being applied mixed with lime mortar as a first render coat, then a plasterers hawk is very useful for holding lumps of material while holding the trowel in the other hand.

And so to work: the wall on the west side of the kitchen was very badly damaged. A lean-to had protected it but had disappeared. There were large holes in the clay dug by woodpeckers  seeking the insects that lived in the fabric. It was far from watertight, and beginning to dissolve. Early on, we had covered it in a  plastic sheet to slow the deterioration.

The plastic sheet was rolled back to start repairs

From June to September 2009, the exposed clay daub, which was most of it in this area, was repaired. New hazel supports were made where the holes were particularly large. The method taught at Onchors farm did not, strictly speaking, use wattles, which are woven lattices, but instead vertical sticks tied to horizontal ‘ledgers.’

The ledgers were wedged horizontally into notches on the oak studs and the vertical sticks were tied to the ledgers with simple knots. There were already notches on the oak studs, but in one of two places where the ledgers had disappeared we had to make new notches, because the old ones weren’t usable.

We found a child’s stitched leather shoe sticking out of the crumbling daub above the west window to the kitchen, looking rather like a moccasin. We were superstitious enough to leave it where it was, because we suspected it was part of some mediaeval building ritual.  It was eventually covered with new clay daub.

Part of a child-size leather shoe can just be seen sticking out of the clay

The original hazel sticks supporting the daub panels were tied together with brambles rather than twine. Some of the sticks that were visible were surprisingly big. They were not split, unlike the technique  of using large split sticks that we were shown on the Essex course.  We did not split our hazel, and instead used rather smaller diameter complete sticks from the coppiced hazel at the end of the meadow.

There was some very firmly fixed cement render left in places on the wall, and this was left for later removal while the urgent repairs were done.

The deep holes, some of which went right through the wall, were filled mainly with clay daub, after soaking  the wall repeatedly with a fine hand spray gun at intervals for 24 hours beforehand. This was necessary to have any hope of getting new material to stick to the ancient daub. Aftyer this experience, I started soaking the wall a couple of days beforehand in other parts of the building.

After the major clay repairs, the whole surface, including minor holes, was thickly plastered – as advised on the course – with a clay-based render made up from two parts  daub to one of a lime and sand render (by volume).  The lime made the clay stickier and also harder, though it was difficult to work into the surface enough to stick properly, even after a prolonged soaking. The surface was then striated with a diamond pattern using a sharp stick to make a key for a final coat of a conventional lime and sand render

The loose material was brushed off

The clay and daub mix was left to dry for as long as possible – six weeks to three months, depending when it was done . The first week or so it was left covered with the plastic sheet and repeatedly sprayed to reduce the cracking.  (In future, we will use hessian instead of plastic, because it will keep the atmosphere damper). Though we started in early summer, the work went through July and August, which is not ideal, because the heat caused the materials to dry too fast.

The final coat was three sand to one lime putty in a thickness of one to two centimetres. The daub and lime mix of the first coat seemed to have done all the cracking for us, and most of the lime render showed very few fissures. However, it was done in four separate panels, using different thicknesses of daub and render to test adhesion and cracking. One area  of render did crack quite badly. This was where the first coat was thin and the top coat thick  – so best to have a thick first coat and thin top layer. It is also important to keep render damp for several days, to slow the drying and minimise cracks.

Vital to keep dampening the old and new clay

Preparing one of the big holes for filling with clay

Putting on a coat of daub mixed with lime

Archaeologists don’t get along with builders

The floor was also of special interest to the archaeologists who investigated the building, which was a condition of planning permission. Once the concrete had been removed, they meticulously worked down through the old floor layers using trowels and logging every piece of broken pottery and other material they found. They were on site for nearly a week, so there was constant tension with the builders, whose timetable was at risk because of need to investigate each layer under the floor. The archaeologists tried hard to be helpful, and did shift quite a lot of spoil for the builders. The main find was the base of the old brewing hearth (see ‘A house unlived in for centuries’). Bricks were taken away for dating. We are still waiting for the results.