Working with Traditional Lime
Lime mortar and render
It would be nice to have a proper pan-style mixer but I can't afford the $1500 or more, and second-hand ones are non-existent around here. I also need a mixer that I can use adhoc, so hiring one is too inconvenient.
I bought the Ozito 125ltr cement mixer new for around $200 and just had to put it together.
It has turned out to do a great job. It will take 60ltrs of sand plus a 20ltr bag of lime putty in one load, which is perfect for my process. I can also change the angle to optimize the mix action.
There's only one manufacturer producing hydrated lime in the form of lime putty in Australia, on the other side of the continent in WA. While there are people using dry hydrated lime in their renders I am not prepared to risk it.
I bought through a reseller in Sydney and had a pallet load (40 bags) delivered to my local sand and gravel business in town by arrangement because they had a forklift. For a reasonable price they agreed to offload and store it until I could pick it up.
It's been maturing on site now for about three years.
2. Lime putty.
Determining the mix
Opinions on ratios of lime to sand vary widely depending on purpose but often also because of how the materials are measured and importantly the form the lime is in when measured. It's surprising that even the experts sometimes get the fundamental calculation wrong, which adds to the confusion of amateurs like myself.
A mix ratio shouldn't be given without clearly indicating what form the lime should have. For example, dry hydrated lime compacts to a much smaller volume when soaked to make lime putty, so a mix volume ratio using dry lime will contain much less reactive lime than the same volume of lime putty. Many historical mix ratios specified quicklime, which gives a different result again.
I'm influenced by the ratio specified by Mike Wye who uses 3:1 sand to lime putty.
I've developed a spreadsheet that will calculate volumes and weights of sand and lime putty given the coverage area and depth. It also accounts for sand voids in calculating volumes and weights of the separate material to achieve the overall render volume.
I did some very small scale experiments (literally about 300ml each) with the two grades of sand I could get locally: fine sharp whitish sand and coarser river sand (not sharp). I mixed both at 3:1 because both measured 1/3 void space, then repointed some patches of wall. Both bonded and hardened well and have not cracked etc.
I start with 3 x 20ltr pails of sand plus one pail of lime putty and get 3 x 20ltr pails of mixed lime render out that I can store and use as needed.
A pan-type mixer is normally preferred and sometimes one with rollers to force the mix together, However, good results can be obtained using a cement mixer if mixed for a longer time. I found the following process gave a consistent workable mix.
With mixer running:
- add one pail of sand (20ltrs) to the mixer
- add up to 10ltrs lime putty gradually using a scoop. Mix may initially "ball".
- once the mix has become reasonably consistent add another pail (20ltrs) of sand
- add remaining lime putty
- once the mix has become reasonably consistent again, add remaining sand (20ltrs)
- continue mixing and adjust bowl angle to maximum that will still contain the material in the mixer. Material should be tumbling without sticking excessively to the back of the bowl.
- There is usually enough water in the lime putty but some may need to be added for a looser mix. Stop the mixer, scoop some mortar up using a gauging trowel and knock twice on the side of the bowl. The mortar should stick to the trowel upside down.
- Restart the mixer and add fibres if making a haired render.
- Continue mixing for 10 or more minutes (the longer the better).
- Empty bowl into a barrow.
Shovel material from the barrow into pails and cover with a centimetre of water. I use water from washing the bowl to also wash the barrow and then to cover the render in the pails. The pails should have watertight lids.
The render can be stored indefinitely before use.
For small patching jobs decant the covering water in the pail then scoop render onto a hawk directly from the pail. Replace the decanted water when done.
For large jobs, decant and then empty pails into a barrow, or use from barrow if mixing fresh. Store unused render in pails as above.
How long between coats
I've been leaving a week between coats. Initial set takes 12 to 24 hours and I've been wetting the surface every 12 hours for the first two days then once a day for two days then letting it dry out completely before applying the next coat, which includes dampening to reduce suction as usual.
Keeping the render damp during the early curing stage minimises shrinkage due to the moisture level dropping. Saturating the pores in the render prevents carbonation so wetting needs to reduce and stop once the render has set enough to resist this shrinkage.
Here's a good site for troubleshooting.
So far, after six weeks, the patching I've done has had exellent results. I've seen only one shrinkage hairline crack appear. The mix was easy to cast onto the wall and then trowel. It adhered well to both basalt stone and timber with mesh and did not sag even in some areas where it was applied thicker than the recommended 12mm or so.
This is the best reference I've found so far for "real" information on how long it takes lime render or mortar to carbonate, specifically providing an insight into the shape of the carbonation progression "front" through the profile over time.
In image 3 I've summarised the results from table 4.5 on page 112 of the document for 4 month matured lime putty and silicate sand in the ratio of 1:3.
At 180 days the silicate sand mix has carbonated to 90% right through to 25mm (the depth of the test samples).
Page 115 of the document graphs the same carbonation depth data.
3. Carbonation progress over 180 days
It should be said that the curing of the samples was likely to have been done in an ideal environment with temperature and humidity set to deliver optimal carbonation rates, which is around 20°C and 60% respectively.
While the silicate sand render carbonates much more rapidly than the other aggregates tested, due to a higher porosity, it also has the least compressive strength. For render that is ok. For mortar, including re-pointing, use a different aggregate.
I mixed up the following recipe and painted it over a patch of the dark green paint in bedroom one, which immediately turned colour and softened, allowing me to scrape some paint off the wall. But not entirely in one application because it had not had time to penetrate. I need to thicken the paste somewhat if I want to strip paint in ernest, which is usually done by adding flour.
There are commercial milk paint stripping products that are essentially a mix of:
Sodium Carbonate = Washing Soda
Calcium Hydroxide = Hydrated Lime (slaked lime; builders lime; lime putty) (30-40%)
Some kind of thickening agent = possibly flour
I Googled "dissolving casein" and found this comment from Marcia Moss at
"Casein is soluble in alkaline pH as it forms sodium caseinate. Some people have used a solution of sodium or ammonium hydroxide. If you use a buffer at pH 8, it may help. But you may need to readjust your pH once you add the protein to your solution."
Further searching uncovered this document
"Casein has been used as a glue since the days of ancient Egypt, Greece, Rome, and China. It is mixed with sodium carbonate and calcium hydroxide, and then dissolved in water. The function of the sodium carbonate is to react with calcium hydroxide to produce sodium hydroxide. The sodium hydroxide reacts with casein, producing sodium caseinate, which is sufficiently water soluble to form a well dispersed sol, a solid dispersed in a liquid. When the sol is spread on the surface of objects to be glued, the sol wets the surfaces and adheres to them. When the water either evaporates, or soaks into the objects, the solid protein molecule remains, adhering to both objects."
Based on calcium hydroxide being 30-40% by weight and my calculation for a reaction that consumes all of the reactants, the mix would require approximately 1.5 parts sodium carbonate to 1 part calcium hydroxide by weight, or 45-60%. I have yet to add any flour to my test mix but the proportion in the commercial product based on this calculation could be from 0% to 25% is my guess. Because I'm using lime putty I can't directly measure my lime by weight as I have no idea what proportion is water.