Rudiments of building with stone

Over the two weeks, Ric taught us the basics of building with stone. This is enough to get us going independently of him, although we would expect to ask for advice as we go, and to learn rapidly from our mistakes!

Splitting Granite. Ric had brought up his set of ‘feathers and tare’, enough to fill 8 holes. We experimented with drilling into granite using our new rotary hammer drill and the 18mm bits, up to 60-70mm apart. We inserted the metal work, with the outer wedges facing the direction we wanted to split. We then got used to hitting the wedges with a small club hammer, listening to the pitch of the sound they produced. We hit the lower-pitched wedges until they reached the higher pitches. As the granite started to give the pins would drop in tone, so we moved between them, keeping them all up as high as possible. After a short period (we were not splitting huge blocks of stone, plus we were working on field-stone which had weathered from new) the granite would give a distinctive ‘pfftt’ noise and give way, pretty much where we wanted it to. We started easy, splitting a rectangular quoin stone across the middle, then worked up to a much more irregular boulder where we needed to spall-off a chunk to make it usable. So far, so good! Of course, in the best of all possible worlds, all our stones would be just the right size and shape, so we would not need to split them up!

As another exercise, we shaped up the sill of one of our new openings. The base for this inconveniently ran through several large boulders. We drilled in and upwards (to get the downward slope for the sill) in the big boulders and used the feathers and tare to split them to height. We used the angle grinder to trim smaller stones to height – cutting parallel lines 10mm apart to the required depth, then using the medium breaker to chip out the remaining stone. Once we had the area cut to size, we laid a mortar bed over it and left it to go off.

Preparing lime mortar: We are using a simple 3:1 mortar mix for the bulk of the building work, where we will not see it. This is two measures of building sand with a small amount of plasticiser, a measure of lime then another measure of building sand, followed by mixing for 20 minutes. This is a little light on the lime, hence the plasticiser, but is quite strong enough. Incidentally, leaving it for longer that 20 minutes is not good, it seems to thin out the mix and is less workable. I actually preferred mixing for 15 minutes not 20, but 20 minutes is the norm.

Laying rubble walling: We picked a door opening that was becoming a window and which needed a single course of rubble, built up to sill height. We laid a bed of mortar on the foundation blockwork and got it to slope up towards the back. We laid the sheet of DPC built in to the backing stonework over the mortar bed. Where it was rather short, we cut an extension sheet that we laid directly on the mortar bed, projecting out, and laid the original over it. This way, any water that gets into the wall will be pushed out the front of the wall. For the first two sections we did, Ric selected the large stones that made up the bulk of the course. We placed these on fresh mortar over the DPC and used small pieces of stone under the edges, to wedge the boulders in place. This done, we put mortar between the gaps and progressively filled in with smaller boulders and, for the narrower gaps, pinning stones. When we came across a strip of the expanded steel mesh embedded in the concrete blockwork, we folded the end back so it would not show and built it in to the stonework, to tie it all together. We part-filled the gap between the inner leaf of concrete blocks and the stonework with mortar, padded out with as much waste concrete as we could. We wanted to avoid hydraulic pressure from the mortar and packing from pushing the stonework out at the base of the section of wall, so we finished filling the gap the following day. We levelled out the top of the course of stones and, prepared the top for sills. The third section, we did on our own – and learnt a lot from it. We ended up being too cautious about using large stones, so ended up with a style of walling that did not really match what was either side – building up the course with a large number of smaller boulders.

The following day, Ric left us to our own devices to complete the walling above a pair of granite lintels he had reinstalled over a doorway and window. This needed one shallow course to tie in to a level in the existing walling, then a full-height one (35cm) above that to get up to the original top of the wall. Being above lintels, I ran a sheet of DPC from the stonework over the lip of the lintels, on a bed of mortar. I had packed a row of small pieces of concrete waste against the concrete blockwork, to raise the mortar bed upwards to the back, to encourage any water in the wall to run down and out. I spent quite a while assembling the 8 large stones we needed, hoisted them on to the scaffolding along with buckets of smaller stones, broken concrete and small wedges to prop stones up. We are using flexible gorilla buckets for the lime mortar, I discovered that I could loop the handles into our hoist and get full buckets up more quickly and easily than lugging up by hand – provided the handles do not give way. Once everything was in place, it was comparatively quick to place the lower, shallow course, back fill the gap and lay the upper course of big stones directly above. Technically we should have waited a day for the shallow course to go off, but we were short of time and accepted the risk that the lower layer might collapse under the one above – it did not, to my relief. We did include three boulders that were slightly too big, the bits that are above the level of the backing concrete leaf will need to be trimmed off before we can lay the concrete blockwork on top, that will raise the roof to the new level.

Preparing for pointing: We roughly filled the stonework with mortar as we went, knowing that next day, we would scrape it back to prepare for final pointing. After a day the mortar was hard enough to stand working, but soft enough to easily scrape back – a bit like granulated sugar that has got damp then dried out. I used a small pointing trowel to scrape mortar away from the edges of stones and to level it back a cm or two to a flat finish. We then used a churn brush to batter the mortar surface to a slightly rough finish. This will allow the final mortar fill to more easily bind to it. This initial picking transforms what initially looked to be messy stonework into a much neater and better defined finish. It is also at this point that you can really see how well the stonework matches what is around it. You have to live with it the way it turns out, but you can learn from it!

The final pointing can wait, we would want to pick out the adjacent old mortar in the original walling and repoint it all together, to get a uniform finish.

Our lime mortar mixes

On our ‘working with lime mortar’ course, Hans Norling convincingly addressed a simmering dispute amongst those who use lime mortar – ratios of sand to lime. There is a common rule of thumb that it should be around 3 sand to 1 lime, by volume. It turns out this is most likely from the days when quicklime was used. When quicklime is slaked, it expands by 1.8 times. Making an equivalent mixture directly from slaked lime (using NHL or lime putty) would need nearer 2 units of sand to 1 unit of lime to get an equivalent mix.

We have made the choice to do most of our building with lime, including all the concrete blockwork more than 1m above ground. Below 1m, it will be tanked on the inside face and have the liquid DPC (synthaprufe) on the other i.e. there would be nowhere for lime mortar to breath. For those who work with cement mortar mixes of 6:1 or 4:1 and who pay £5 per bag rather than our £12 for lime, this might seem extravagant. We are paying a premium, but lime is a lot less dense than cement, so goes quite a bit further. We really do want to make the structure as breathable as possible. It is also great to work with, it stays workable for a lot longer than cement mortar. If we have left over mortar at the end of the day, we put it back in the mixer next day, add new sand and lime and carry on as normal. Cleaning up is quicker and easier. We do have hessian to keep the sun & wind off and a canvas to work under when it is wet, but most of the time, it is all very well behaved.

The downside to lime is that, over time, it starts to bind to metal objects such as trowels and the cement mixer, forming a hard crust of what I expect is calcium carbonate. In the cement mixer the mortar sticks to it, preventing it mixing properly and making it a bit of a swine to empty. At intervals I have spent a happy hour with a chisel tip in our light SDS drill, chipping it all out.

Our normal mix for non-visible mortar has indeed been 3:1 by volume, using building sand:NHL 3.5 and whilst it does go quite plastic after 20 minutes mixing, it is more workable with plasticiser added. It also tends not to stick very well as it starts to dry out. For laying blocks, we are making a slightly wetter mix which works well enough. When building rubble wall, we need the stiffer mix and stickiness is not such an issue. I quickly found the value in splatting even non-adhesive mortar into narrow gaps to fill them. If you miss, it is messy and a waste of mortar. If you get it right, it does fill all nooks and crannies. Strength-wise, 3:1 is quite adequate for us.

I experimented with increasing the lime content and decreasing the plasticiser. As I got towards 2:1 and no plasticiser, the result was indeed smoother, sticker and easier to work with for longer. It is also significantly more costly! So we will go with 3:1 for the bulk of the building, but I will use 2:1 for final pointing, where adhesion is very important. We will also be using our sharp sand from Bridgend quarry in place of building sand. We have two big bags, this should do for most of our final pointing. We will experiment with the pointing mortar at a future date, we can adjust the sharp sand by adding building sand or a coarser grit if we need to.

Training to use lime mortar

We were tipped-off by another steading-converter (the far side of Ellon from us) that a 2-day ‘working with lime mortar’ course was scheduled for the week we were arriving, to be held in Banff. This was organised by the Scottish Traditional Skills Training Centre, who do a variety of courses such as ‘managing woodland’ or ‘traditional window design’. We enquired and were instructed by Ric to attend. So we spent two days at Banff castle, about a 50 minute drive over to the Moray coast. This was run by master-mason Hans Norling, a Swede who has lived in Scotland for 16 years, giving him a most unusual accent. He co-founded his current company Masonry & Lime ltd and employs 16 staff. He was very laid back and really knew his stuff – we covered the theory & practice of picking out mortar, mixing lime mortars, pointing stonework, applying harling and using limewash. This may sound pretty dull, but when faced with loads of walling to build and several hundred square metres of picking & pointing, it has been a lifeline.

We got to meet 9 other like-minded souls including a couple over Fyvie way and, a bit spookily, the owner of a farmhouse just south of us, at Auchnabo. He is between us and our Architect, I guess his building is one of the points of light we see at night.

Most of us had some previous experience working with old stone buildings and there were sighs of relief at various points as we realised we had not irretrievably wrecked whatever we had been working on. Which is not to say that we will do a better job now we know how! My moment was when we got his blessing for using Singleton & Birch NHL 3.5 lime – good stuff, he said. They were using St Astier on the course, which gives a rather whiter finish. He was also increasingly adding quicklime to his mixes, to make them sticker and more workable and to give a softer finish. His view is that old lime mortars would have included locally-produced quicklimes.

We asked for advice on removing our 50+ square metres of concrete render along the entire north wall, he could not give us much comfort – he did unbend enough to say to use an SDS drill with chisel tip. Using mechanical tools on old stone is a bit heretical in the building preservation industry. I followed his tips and tried an area where I allowed the point to stay in any one place for only a second or two at a time. This mostly worked, the mortar was much more likely to end up crumbling and falling away, than where I left the tip in one place until it broke through the mortar. It potentially reduces the damage done to the stonework. It did not work so well where there was a thin layer of mortar strongly bonded to individual boulders. [A further tip on removing hard cement pointing from stonework, from an English Heritage publication: Run a diamond blade down the middle line of the pointing, to help break it out without damaging the stonework. I will look at getting a blade for our small angle grinder.]