Late medieval furnaces in Britain are very interesting. Not only if you like big fires, flames, liquid metal that could burn your foot off!!! but also because they were actually pretty advanced and grew larger as technical progress occured and there was a need to build bigger furnaces. The materials they were made of had to have several properties – be heat proof, so that the furnace didn’t catch fire or fall apart as the structure was baked like a piece of pottery; ideally locally available so that money wasn’t wasted on carting them around, and of course they had to be strong enough to take the load- structural and the weight of the hot metal.

The larger furnaces made by the 14th through to 16th centuries were easily 2 metres wide by 3 metres long by 2 metres high. They were substantial constructions, with walls a foot or two thick and ultimately used six or more cubic metres of stone, clay and other substances.
But they did wear out, and if you want to cast more or bigger items you’ll need to build a new furnace. So I was excited to find mention in the Lord High Treasurers Accounts of Scotland of building a new furnace in 1515. Someone called “Auld Juliane, Italiana”, was paid 12 pounds and 12 shillings for 6,000 tiles, specifically for making a new furnace for gunfounding.
The question is of course, what was wrong with the old furnace and how big would the new one be?

It seems likely to me that after Flodden and the accompanying political uncertainties (although the Scottish state as a whole coped quite well), that the old furnace which had been used to cast several large cannon had decayed. A couple of winters of frost damage to a rained on furnace would certainly cause damage. The other possibility is that it had been so well used that it was badly vitrified (that is, bits were turning into glass which is weaker and no use as a furnace material) and simply wasn’t up to the job of casting new cannon.

But how do the tiles fit into it?

Tiles are found in furnaces such as those excavated at York and elsewhere is as the base of the firebox. Tiles set on edge make a hard, long lasting surface for combustion to take place on. Since tiles have already been fired up to and beyond 1,000°C, they don’t react or change from the heat of the fire. Their presence would also make the raking of ashes from the firebox of the furnace much easier.
But examination of some of the pictures I have of medieval furnaces from excavations, e.g. the Bedern foundry in York, suggests only about 400-500 tiles are needed for that purpose given the space to fill. And the Bedern foundry made bells and cauldrons for decades; it is hardly an unrepresentative or unsuccessful site.

So 6,000 tiles, how exactly do you arrange them? They can’t all be for the base of the interior where the metal was melted, although this is another possible use for them. There aren’t enough to build the entire volume of the furnace. Which is where we come to the furnace found in Worcester. A conjectural drawing shows the tiles stacked up on the outside, acting as an outer layer, making the furnace weatherproof.
If you assume the tiles are the size of the suspected hearth tile found in Worcester, i.e. 26.5 by 15.7 by 3cm in size, then a furnace 3 by 4 metres by 2 metres in height would require about 3,400 of them to act as internal or external walling. Or if the tiles are 15mm thick like the Worcester roof tiles, then you need 7,000 of them. Or if they were 3cm thick but half the length you might need 7,000.
If the furnace was 5 by 4m as the main Worcester furnace was (And perhaps 2m tall), then you would need about 4,400 of them. There certainly were enough tiles to build a furnace with.

The end result is rather like this:

Which is my bronze casting furnace at Lanark Festival of History. The narrow tiles are similar in dimension to the ones at Worcester, only not as wide. They are there to give an effect, rather than be perfect replicas. Oddly enough Historic Scotland and other such entities would object if I asked to build a proper sized furnace in their properties, let alone the work required for such a project. Hence I have to compromise if I want to wow the audience with actual bronze casting.

It is clear to me that the furnace in Edinburgh castle would have been a reverberatory one with tiles on the outside, as at Worcester. A reveberatory furnace is a kind of furnace which is arranged so that the flames rise up from burning wood, hit the roof the furnace and beat down upon the metal from above, reverberating off the roof. Biringuccio’s drawings of them are like this:

The hole down at the bottom left is where the wood goes in, and you can see that the flames rise up and over the metal, before escaping out of a small chimney at the far end, which he doesn’t show properly.  The draft can be controlled by a hole in the base chimney with a cover on it.  Both Biringuccio and Benvenuto Cellini wrote instructions on how to build them, although Biringuccio used them in the overlap between 15th and 16th centuries, whereas Cellini is fully 15th century, dying in 1571.

Now, to get back to the Edinburgh castle setup, I am sure they had a reverberatory furnace already, because they cast a number of large cannon, using cartloads of wood to do so – they were buying 20 or 40 loads at a time. There is no other possible use except in a reverberatory furnace, and there isn’t water power in Edinburgh castle to drive bellows to pump air into a charcoal furnace. Possibly they could have built a wind furnace, but we don’t know what that would have been like and it wouldn’t have been as controllable as a proper reverberatory one.

Of course tiles would not have been enough for a full sized furnace. You would also need rocks and clay, but these are not mentioned in relation to the new furnace in the accounts. Perhaps they had some on site already, or were using the old furnace for parts. Either way it was a skilled business, carried out under the control of the master Melter, Robert Borthwick. By this time he seems to be acting in sole charge of the casting. Previously he had worked with the keeper of the castle and various foreign gunners, but now the accounts mention only him. Further research will be required to find out whether or not he was actually successful in casting more guns. I see no reason to doubt it he was or not, but the fragmentary nature of the accounts during this period do not help matters. There may be an account of the cannon owned by the king available somewhere else in the historical documents, but that will have to be ferreted out.


Dalwood, Hal and Edwards, Rachel Excavations at Deansway, Worcester 1988-89, CBA Research Report 139, 2004. 

Richards J.D. (ed.) The Bedern Foundry.  The archaeology of York.  Volume 10/3 Council for British Archaeology.

Smith C S and Gnudi M T (trans) 1990, The Pirotechnia of Vannoccio Biringuccio. The classic sixteenth-century treatise on metals and metallurgy (rev edn, 1959 reprinted 1990) New York: Dover Publications Inc.