Something different – iron smelting

A re-enactment group I know are good friends with a venue in the Lakes District, near where iron used to be mined. They are interested in carrying out a medieval iron smelt at some point in time.

This blog post therefore is a wee summary of what I know about it at the moment, based on one particular interesting paper I found recently, along with some pictures.

The paper is this one:

There are many topics that the paper brings up, all of which are important if you wish to have as easy a re-creation of a medieval bloomery as possible. The simple fact is that our ancestors spent generations refining their metallurgical techniques, and ignoring what the archaeology and practical experiments tells us would condemn us to spend many years in re-creating the same things as have been done before. I see no need to do that, because, as with Tudor bronze casting, it will be quite difficult enough as it is even if we do everything right, so it is important to learn from previous attempts.

In the abstract is a list of topics that are examined that seem to me to be very important and I will first discuss them below based on my knowledge without having properly read the paper:

the nature of the furnaces, the bellows and blowing rates; the ore, charcoal and clay

types, quality and treatment; and the operating conditions, the products and the losses of material through the refining process.

To start with, the nature of the furnaces. I have been unable to find out much about lakes district furnaces in the late medieval period, but it seems obvious to me that since blast furnaces had not reached that part of Europe, they were the typical chimney shaped bloomery furnace. This will have to be fully confirmed by reading more widely in the archaeological literature.

They are called bloomery furnaces because they produce a ‘bloom’ of iron, like you can see being hit with hammers in the darkness in this photo:

Striking iron bloom

The furnaces would have looked something like this, with the second photo showing slag flowing out of the furnace:Iron smelt bloomery at work

Iron smelting tapping slag


As you can see above, the air comes in the side, a bit up from the bottom, and then when the job is done the slag is tapped out the bottom and the bloom of iron also taken out and bashed with hammers to squeeze out slag and consolidate the actual iron. This is hard work, better done with a water powered trip hammer, which they did have at that time and there must be evidence for them in the area. The charcoal and ore go in the top at regular intervals.

The next topic listed is bellows and blowing rates. How much air goes through the furnace affects how much of the iron ore is reduced, i.e. has the oxygen removed to make the pure iron, and also the temperature because as you know, blowing on a fire increases the amount of oxygen going into it and therefore the burning rate. But were their bellows water or hand powered? Most modern re-creations seem to use electric pumps, because that saves a lot of work and means the airflow can be measured better.

Next is ore, charcoal and clay types, quality and treatment. An issue with making a reproduction furnace might be that there aren’t many good quality ores left available for us to smelt since they will have been mined out in the early modern period. That aside, the ore was surely crushed into small pieces, but how small? Dust would be too small and probably not agglomerate properly in the reducing zone of the furnace. At the other extreme, large lumps would be too big to react properly, so the question is how small to make it all, pea sized? Walnut sized? Different ores will present slightly different problems with the firing due to their different chemistry, not to mention that some require to be roasted before smelting to turn the sulphide into the oxide.

I recall also that the bloomery firings I helped with years ago, the charcoal was sieved, so as to remove the smallest stuff. Charcoal reduces in diameter as it burns up, exposing a greater surface area to the air and therefore requiring more air to be introduced, but also increasing temperature if you do add more air. And if the charcoal is too small it forms a wall blocking the passage of air.

The clay is important insofar as the furnace structure shouldn’t fall in on itself and not react much with the charcoal and iron inside.

Now, after looking through the paper, I can elaborate on a number of points.

Firstly, the furnace structure can vary quite a lot. It seems at least one medieval furnace had a long entry way through which the slag was tapped. It also had a lot of stone forming the structure of it, covered with clay.

This clay used to line the furnace is important because it can react with the slag or the iron. It seems it is also possible to build a furnace with wattle and daub as a main superstructure, giving it a structure that the clay can be built up around quickly and easily. This can then be burnt away once it is dry. The walls should not be too thin in order to retain heat well, or rather, if you make the wall too thin, although it makes it easier to build, too much heat will be lost and you won’t reach the right temperature.

Then there are the bellows. The experimenters went through various sizes and numbers of bellows. It proved difficult to maintain consistent blowing rates during a smelt, and then during the next one the blowing rate would change, due in part to changes in personnel. Interestingly, electric fans seem to have given too fierce a blowing rate. The experimenters eventually built some non-historical piston based bellows, which permitted good control of the variables.

It is important also to consider the nozzle and it’s location with the air hole, not fully in it, so that the inside of the furnace can be seen and any blockages removed by a rod.

One of the most interesting things is the charcoal. Once you have used various sorts of charcoal you will know the difference between them, and the experimenters found a big difference between beech and oak, birch and alder, in flame colour, burning rates and bed stability. But oak wasn’t as good for the smelting, despite having been used a lot in history. Alder was best. Oddly enough I think alder is a common tree in such wet areas as the lakes district.

The 25-40mm fraction of charcoal was used for smelting. It was added in batches with the ore, but half of it needed to be used for initial filling and pre-heating of the furnace.

It is clear too that if we want to make an experimental bloomery we shall have to check the type of ore and go for as good a quality one as possible. It should be possible to use bog iron. The ore quality is important, but so is consistency of chemistry, if necessary it should be ground up together and sifted together so several smelts can be carried out with as few variations as possible.

Experimental proposals

Regarding doing experimental smelts at the proposed location, we shall require a good 6 or 8 people in order to ensure nobody gets too tired. Ideally it would also be done as part of a bigger re-enactment event, ensuring a steady stream of visitors. We would require clay, ore, charcoal, sand, some stones for building the furnace up and of course bellows and some sort of stand for them.

None of this is impossible, but I think the hardest points will be building a good furnace, getting decent bellows and good ore. Normal barbecue charcoal won’t be right for it, rather it might be better to make our own hardwood charcoal the year before and keep it dry somewhere over the winter. As for space, not a lot would be required, only say 5 by 7 metres to ensure a good space for working in. There will be some sparks too, out of the top of the furnace, but aside from the heat the actual hazard is low. The liquid slag will be safe enough at the bottom of the furnace, although some might spurt out when we beat the bloom with hammers.

As a project it would likely take three or four weekends to come to fruition, between building and drying the furnace and doing some test runs in order to make sure the furnace and equipment work properly. The actual smelting run can take 8 or 12 hours, depending on all the factors listed above. As an educational tool it is quite a striking one, and can draw in aspects of history, archaeology, chemistry and engineering.

It’s certainly something I would like to do at some point.

2 thoughts on “Something different – iron smelting

  1. I did a iron smelt in Ecclesall woods with uni of Sheffield using this type of furnace. Its probaby worth talking you to them as they will have looked into a lot of the points you raised, even though they were recreating an earlier period. (They’ve done this process quite a lot as experimental archeology.) They built the furnace, that we used, 3 days before so the clay could start to dry out. The first few hours after it was lit was further drying and warming it up before getting it hot. Took us 8 hours on the bellows with a team of ~10 people before we opened it up. It was an incomplete melt and it’d stuck to the furnace. A lot of damage was done to get it out before patting it into an ingot shape. Apparently most smiths won’t use / touch this type of smelted iron because the ‘grain’ is random/mixed up like burr/knotted wood, not smooth like modern commercial metals, and therefore it’s difficult to work. They’ve found 2 smiths in the uk who are willing to give it a go.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s