More Smelting Experiments at University College Dublin

More Smelting Experiments at University College Dublin

It was Tuesday the 20th of June and the weather was approaching that rare state of the skin-melting Irish heatwave. (Yes, these things do exist. The Sun rarely gazes upon this green island, but when it does, it does so with a vengeance most vicious.) In order to ease ourselves into the spirit of the summer solstice, and to show the treacherous orb in the sky that we do not go down so easily, we ran another bloomery smelt.

(I also happened to need a bit more raw material for my thesis research.)

This has been my second successful run since my first attempt to smelt Irish bog ore during the spring break in March. My method has changed somewhat since then, and I have in come a tiny step closer to understanding the work of Irish early medieval smelters. Therefore it is due time that I present a short report, while I work on my MSc. Thesis and more content for the ‘Guide to Iron Smelting’.

As I already mentioned in my report on the UCD Smelting Week, traces of ironworking are quite common in Ireland. Yet in most cases, the archaeology has been eroded all the way to the slag pit/bowl, which was dug into the ground. Therefore there is a bit of a mystery connected with the shape and height of the shaft, which would have been built above it. What we are a bit more sure about is that these are all slag-pit furnaces, since there are no finds of tapped slag in Ireland before the Normans came over.

The furnace which I am using at the moment is an adapted interpretation of the one excavate in at the site of Grange 2, Co. Meath. To this date it is best preserved find of an Iron Age or Early Medieval furnace in Ireland. It is unique in Ireland since it features 45 cm of the shaft preserved with a hole where the wall was broken in order to extract the bloom.

Grange 2 shaft furnace
Rare preserved remains of a furnace shaft at Grange 2 in Co. Meath. (Source: Kelly et al. (2011) Final Report: Archaeological Excavations at Grange 2, M3 Clonee North of Kells motorway scheme.)

Mimicking the dimensions of the Grange 2, my furnace was built of clay sand, with an 18 cm deep and 30 cm wide slag-pit, dug into the ground. Above this furnace bowl, I have erected a 50 cm tall shaft. This tapers from the 30 cm at the bottom, to an internal diameter of 16 cm at the top, just like the original. Since the archaeological find shows a broken edge at the top, adding 5 cm of height does not require great leaps of faith and imagination.

irish bloomery furnace
My adapted interpretation of the Grange 2 furnace.
bloomery furnace
The idea moulded into clay and sand.

I like to call this furnace type of bloomery furnace the ‘Bowl Furnace 2.0’, a.k.a a bowl furnace with a shaft. This a very similar setup to what Peter Crew used for his experiments with Iron Age bloomeries in Wales.

My main addition to the design is making space for a door. There is no proof that this was done in Early Medieval Ireland. Instead the wall of the furnace was most probably simply broken through. I added the door since it does not alter the mechanics of the furnace, yet having a separate door which can be broken out reduces the chances of fatally damaging the furnace shaft and thus making it easier to re-use. Being able to reuse the furnace by simply patching up a crack or two and inserting a new door makes it easier to run another smelting experiment, instead of spending another day building a new furnace, which is always helpful when you are somewhat short on time.

The tuyere gets built into this door, so that it lies c. 25 cm above the bottom of the pit. It is angled slightly downwards and extends about 3 cm into the furnace. This way the tuyere primarily provides a funnel to point the bellows at and helps direct the air blast.

iron smelting bloomery furnace in operation
The furnace in action. The flared tuyere channels the air from the bellows allows me to peek inside the furnace. Any drops of slag can be easily pushed away with a greenwood stick.
Drying the freshly inserted furnace door. (Photo: Aidan O’Sullivan)

Since I already had very fuss-free smelt using the same setup a month ago, I had my hopes up for this one as well. This almost had a feeling of routine to it. So different from some of my earlier experiments.

The day started with me coming in early to put a door on the furnace, insert a tuyere and light a fire in the pit. While was heating and the door was drying, I began to chop the charcoal. Soon I was joined by Sean, who took a break from his bell-brazing experiments, and even the Centre’s director got out of the office in favour of a day in the smoke and sunshine.

The first charge of ore went in around 11:35.

iron smelting pumping bellows
A furnace, a set of bellows, a bucket of ore and one Sean pumping away.

There we found ourselves, feeling as if the universe turned on the broiler function, pumping bellows, feeding a furnace and gazing into the blaze of a miniature sun seeping through the tuyere. Shades were worn. A cowboy hat was spotted. The only thing reminding us where we were was the melody of the Irish vernacular.

Sweet Jaysus! Now we are sucking diesel!

We used 15 kg of bog ore, which we gathered in a Midlands bog back in May. This was roasted and crushed a week earlier. The ore and charcoal were charged at a rough ratio of 1:1 by weight. A charge was therefore composed of 500 g of ore and 500 g of charcoal.

irish bog ore iron smelting
15 kg of Irish bog ore, roasted and crushed.

We judged the temperature by observing the colour of the blaze, as seen through the tuyere. We kept it at a bright yellow to white heat, which led feeding in a new charge every 7 minutes.

I was not disappointed, for the smelt ran smoothly, with no issues beyond having to occasionally push a droplet of slag away from the tuyere opening. After less than 4 hours, the furnace devoured the last of the ore. Now it was merely a matter of adding a few extra charges of charcoal to help the last of the ore join the bloom, and then it was time to burn down what was left in the shaft.

Around 15:45 I started tearing out the door. The belly of the furnace was filled with a bloom, which needed some convincing before it could be torn out. The fact that I did not have time to finish a large enough pair of bloom tongs before might have had something to do with that.

The bloom came out with a large portion of the slaggy furnace bottom attached to it. This we managed to mostly knock off during the initial compacting with a sledge hammer. The resulting bloom weighed 3.8 kg, although I think that a considerable portion of that will melt out as slag when I reheat it.

bloomery iron smelting
Prodding the bloom with a wooden pole, trying to compact it a bit before I even open the furnace.
bloomery iron smelting furnace opening
Time to break open the gates of hell.

bloomery iron smelting furnace opening

bloomery iron smelting
Get out you hot unconsolidated mass of iron and slag!
iron bloom smelting
A bloom emerges from the pit.
iron production iron bloom
Trying to knock of the slag bottom and compact the bloom while it is still hot.

bloomery iron smelting

Hit it like you mean it! Try to miss the research centre’s deputy director. (Photo : Aidan O’Sullivan)

iron smelting bloom consolidation

iron smelting bloom consolidation

bog iron bloom
The Result: 3.8 kg of iron bloom.

After the Smelt


Two days later I went in to document the aftermath.

The furnace suffered a few crack. A large one down the back formed during the smelt, while some of the rest were the result of me trying to yank out the large bloom. Nevertheless, after a bit of patching with wet clay, the furnace shall be good to go for at least another run or two.

There was very little slag left in the furnace, this mostly took the form of shiny slag prills and droplets mixed in with the charcoal left in the bowl. A lot more slag was found around the wooden stump, where we did the compacting, which confirmed my suspicion that I managed to pull out the furnace bottom with the bloom.

There was also a nice pile of gromps or bloom fluff left in the furnace. Maybe if I added a few more charges of charcoal, some of these would have had time to fuse with the main mass of the bloom.

I am now considering running a small furnace or hearth, where I re-smelt the bloom fluff from the last two smelts. Just to see what sort of result this gives me.

bloomery iron smelting aftermath
The Aftermath. Bits of charcoal, furnace door, tuyere, slag and gromps scattered in and in front of the furnace.

blomery iron smelting aftermath

bloomery furnace damage
This will need some patching before the next smelt.
gromps bloom fluff
A pile of gromps, lost little bits of iron fluff, which never had the chance to become a bloom.
There was relatively little slag left in the furnace.
bloomery smelting slag
The rest of the slag ended up as debris, scattered around the stump where we did the initial compacting.

Ideas for further Experiments


Despite the fact that I have now enough usable bloomery material to do my research, there are some puzzles which I would like to answer, should I find the time.

Although it seems that I am producing usable blooms, which seem to resemble what we find in the archaeological record, I have yet to produce a proper furnace bottom, which fills up the bowl of the furnace and stays there during bloom extraction.

I order to do that I will try to further increase the amount of air pumped into the furnace, raising the rate of burn to the point where a charge goes in every 5 minutes (10 minutes for a kilo of charcoal burnt). I am hoping that this increase in temperature will ensure better separation of slag and iron and a more liquid slag which better fills out the bottom before it solidifies.

I am also thinking that having 25 cm of pit bellow the tuyere might be a bit excessive in this scenario. Therefore I might bring the tuyere down to ground level at the next smelt, which means c. 18 cm between the bottom and the tuyere.

Any other thoughts or suggestions? Do write in the comments bellow.

Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInEmail this to someonePrint this page

4 thoughts on “More Smelting Experiments at University College Dublin

  1. Is there anything between the burn area and the slag pit below? What is to prevent the bloom from falling down into the slag and resulting in what you’ve found: a mixed lump of bloom and slag?

    1. There is only charcoal in there. Although this is a relatively shallow slag pit, you often get an unburned section right at the bottom. Heat always travels forwards and upwards from the tuyere. Therefore slag starts solidifying after it drops bellow the level of the tuyere. I hope that lowering the height of the tuyere should raise the temperature in the basal pit.

    1. Interesting. I definitely plan on running a smelt where I try to run it as fast as possible and see what happens. Then I can compare the results with the previous smelts and the rare find of a bloom from early medieval Ireland.

Leave a Reply

Your email address will not be published. Required fields are marked *