Looking at real surviving bronze objects from the medieval period leaves me amazed at how good they are. The thinness, the smoothness and so on. I have never been able to replicate it using clay based moulds, so far.
The question is, why?
Here’s a pair of spectacle buckles I have made by casting into clay based moulds:
They both have, to a greater or lesser extent, a pockmarked surface.
In other objects you can see what look like real bubbles, which may well be from oxygen dissolved into the hot metal. It’s my fault for not poling the metal, that is, putting a green stick into it to burn out the oxygen, but most o the casting I have done has been very close to the the right temperature.
So that’s definitely part of it. If the mould and metal are both very hot, then it will fill the mould completely, and if the metal hasn’t much oxygen in it, then there won’t be many bubbles.
This would explain some of the poor surface finish. And I have seen medieval bells with bubbles at the surface a millimetre across, often several grouped together, so they obviously had the same problem.
Now, here are some actual medieval belt buckles, found in the remains of a real 13th century or so workshop in Dublin:
Sorry about the picture quality, the photos were taken through a glass case. What you can see, hopefully, is that there are unfinished castings with some very rough surfaces. Which immediately makes it clear that a lot of fettling goes on with the unfinished castings. This was also suggested to me by a member of the public at Chatelherault a couple of years ago. I also wonder if the state of the sprue attached to them indicates the type of mould they were cast in. Pewter objects usually have a clear, sharply defined sprue with diamond cross section, because that is how the channels are cut in moulds. These by contrast are squarer in cross section, and so I wonder if they were made in clay based moulds. It is known that buckles and such were made in multi-layered clay based moulds such as one found in London, so they would have had the same difficulties I did with the quality of final product.
Another option is that they were cast into stone moulds. These would, looking at pewter castings, give very smooth, detailed surfaces. The only slight problem there is that there’s no real evidence for it. So few moulds survive, and I’ve not come across mention of stone moulds in wills or the like. Moreover, they wouldn’t last for so many casts, because 1100C Bronze or brass is hot enough to cause major thermal shock and a lot of expansion and contraction. Besides, most moulds I’ve seen have lead plugs, which might get melted out by the heat from the cooling bronze. I’m going to be doing some experiments this year to show it one way or another.
Yet another way to ensure a good quality, smooth surfaced casting is by making the mould up using very fine clay slip around the wax original, before adding the coarser clay. I haven’t succeeded in doing that method myself, but have spoken to people who have and they say it works well, even better if you add some fine carbon in so that there is a reducing atmosphere inside the mould, which means you get a nice shiny non-oxidised surface finish.
From Biringuccio, the early 16th century founder, there’s evidence for the use of casting using sand based recipes or fine powders, but of course such stuff wouldn’t leave obvious traces in the archaeological record, unlike the tonnes of clay based moulds which came from bell and cauldron moulds. Or perhaps some archaeologists have overlooked it? I have yet to try such substances, but they might also not have been much used in Britain compared to the continent. He does mention the use of multi-layer moulds like that found in London, and his instructions for making bells are almost certainly the same as how they were made in England, so it does seem worth trusting his descriptions and recipes.
The final option is that the way I am doing all parts of the work isn’t right. Hence the experiments with different mould materials that I shall carry out.
Before that though the obvious thing to do is do a quick and dirty test clean up of one of my own castings.
So here is a buckle, cast in a clay mould from pressing the original into the damp clay.
Before treatment with files:
What is immediately clear is that with a fairly small but finely cut file you can get a nice shiny smooth finish very quickly, with only a few strokes, and that will take out most of the unevenness of the as cast surface. Further work could also give each of the edges a sharp angle to it. So it is likely that some of the perfection of buckles is down simply to good fettling by whoever did it.
But as always, more experimentation and research required.