An Introduction to Iron Smelting – Part IV: Furnace Design

An Introduction to Iron Smelting – Part IV: Furnace Design

You have a space for smelting selected, your equipment is gathered, you clay is mixed, you charcoal chopped and your ore crushed.

Now you are finally ready to build the furnace.

We are getting dangerously close to the actual smelting here. And that is what you came here for in the first place. Right?

But first you will need to choose what sort of bloomery furnace you want to build. 

There are many possibilities and examples available – both historic and modern. You are now wondering which to follow. Surely some must be better than than others for what you want to do. You do not want waste time re-inventing the wheel. You want to choose something that works well.


In this part of the series I will first go into the variations found in historic furnace. Then I will present a few designs, which have proven themselves practical and reliable. The detailed instructions on how to successfully build and dry a furnace will follow in a separate post.

Before you move on, I would like to remind you that this post is the 4th part of a series on getting started with iron smelting. Should you have just tuned in for the first time, or if you happen to need a refresher, then you are invited to take a look at the previous installments:

Part I: Theory
Part II: Tools of the Trade
Part III: Raw Materials

So make a nice cup of tea or coffee, and dig in.

One Concept Under Many Faces & Disguises

bloomery iron smelting furnace

There exists a multitude of variations on what a bloomery furnace is. Nevertheless there a few unifying characteristics to all of them.

The same thermo-chemical reactions happen in all of them, and to the same end. They are makeshift features meant to provide the conditions where iron ore can be reduced to metallic iron and form a bloom.


They all have in common that they have some kind of shaft where the reduction happens, and an air intake at the bottom of this shaft. Bellow the air intake there also has to be some space for the slag to drip down into and accumulate.

Beyond that, there begins the plethora of variations on the theme.

Some furnaces shafts can be tall or low, slender or wide. A furnace can have a slag pit dug into the ground, or it can be completely built above ground. It can be free-standing or built into a bank. So on, and so on…

For a list of all the variations that were historically used in Europe, I again invite you to take a look at Radomir Pleiner’s freely available magnum opus ‘Early European Smelters’.

Your choice of furnace will ultimately depended on several factors. The first question will be whether you are doing a specific reconstruction of the proces for the purposes experimental archaeology, or if you are just interested in producing some bloomery iron/steel for use your craft projects. Should the former be the case, then your choice will be determined by the time and place, which you are investigating. On the other hand, should you mostly be interested in the end result, then your choice will be determined by the ore used, preferred smelting technology (slag tapping or no slag tapping), and the range of easily available materials.

With that said, I will now break down the furnace into its constituent parts and present their most common variations.

Basal Pit

How deep should you go?

The basal pit is what we most often encounter as an archaeological find of a bloomery furnace, since unlike the above-ground shaft, the impersonal forces of entropy are a bit more kind towards features dug into the ground.

The purpose of the pit of the basal pit is to collect slag during the smelt, so that it does not block the air intake.

There are 3 possible variations on this theme:

  • Slag Pit (deep)

  • Slag Bowl (shallow)

  • No slag pit

The type of basal pit can tell us something about the smelting method used.


A deep slag pit (think at least as deep as it is wide – i.e. 30+ cm) indicates no slag tapping and a considerable amount of ore being smelted per run. This is your classic ‘Slag-Pit Furnace’.

All of the slag produced hopefully ends up in the form of a drippy block in the pit. This type of smelting is typical of early time periods (Iron Age) and usually involves quite large furnaces. It generally begins to be replaced by (often smaller) slag-tapping furnaces during the course of the 1st millennium AD.

This type of smelting presents a challenge, since the slag pit has to be pre-filled with organic matter (greenwood sticks or straw) so that a fire can be started above the pit. The idea is that during the smelt, the slag drips down and burns away the filler rods.

I will not go into much detail on this type of smelting, since I still have limited experience with using deep slag pits. Should you still be interested, then I invite you to take a look at the following smelting report, published by the Wareham Forge and the Dark Ages Re-Creation Company (DARC).

iron smelting bloomery furnace slag block denamrk
A massive slag block from Jutland (DK). Weighing 200 kg and dated to 200-500 CE, it is a typical representative of large furnaces with deep slag pits. Currently kept on display at the National Museum of Denmark in Copenhagen.
iron smelting bloomery furnace germany
Well preserved slag pit furnaces from Northern Germany, dated to the early 1st millennium CE. Note the thin tall shafts, multitle air intakes and cluster of slag blocks. Since the slag block was often left in the ground after the smelt, a new furnace had to be built next to it, thus creating a battery of furnace remains. Kept on display at the Archaeological Museum of Hamburg. (Photo: AMH)

A shallow slag bowl (usually 10-15 cm deep) indicates either slag tapping, or a slag-pit furnace used for smaller smelts.

We can see which of the two options was used in a specific case based on the slags found with the furnace, since tap slag has a specific, flow-like, appearance. These furnaces often tend to be narrower than the previous type.

Slag-pit furnaces with shallow bowls are, for example, common in Iron Age UK and Ireland. In Ireland they are still used until the end of the early medieval period (400-1100 AD). It is also what I used during my university research on early medieval ironworking.

iron age UK bloomery iron smelting furnace archaeological remains
The remains of two iron age furnaces from Wales. These have shallow slag pits, roughly 15 cm deep. Also note the clearly visible door arch and thick walls on the the right furnace. (Source: Crew 2013)

The absence of a basal pit is a good indicator of slag-tapping. It means that that all of the furnace is built above ground. At the same time, it also decreases the chances of any part of the furnace being preserved in its original shape or position.

It should be kept in mind that building the furnace above ground makes effective slag-tapping possible, but does not make it compulsory. The air intake still needs to be placed above the bottom of the furnace, and by placing it high enough on the shaft, the effect of a slag pit is mimicked.

bloomery iron smelting early medieval shaft furnace
By gazing through the gaping mouth of the furnace on the right, you can clearly see that it is built completely above ground.


Constuction Method

A Question of Raw Materials.

Furnaces can be constructed in two main styles:

  • All Clay

  • Combined Method (Stone & Clay)

According to the first method, you construct the entirety of the furnace out of your metallurgical muck – the clay mixture which you learned how to prepare in the post on raw materials.

The second method builds a base out of stone and a shaft above it out of a clay mixture. The base then requires an opening into which a clay door is inserted, so that you have somewhere to place the air intake and extract the bloom.

Something like granite is good to use for the furnace base, due to its high heat resistance. Soapstone is even better, but few people have copious amounts of the good stuff laying around. You should stay away from using limestone, since that way you run into the danger of either turning the bloomery furnace into a lime kiln, or adding possibly unwanted fluxing agents.

The choice of construction method was mostly determined by the local geology.

If you have plenty of good, heat resistant clay available, then you will likely simply build a furnace out of a good clay mixture.

If there is a lot of heat resistant rock available, and the clay is a bit less reliable, then you will want to build the base of the furnace out stone, since there is where you find the most intense heat. The stone will be merely mortared together with some sandy clay. The shaft can still be made out of the clay mixture, since the temperature drops quite quickly once you move up and away from the air intake. The limiting factor and most likely thing to start melting will then be the furnace door.

Subsequently, in some truly outlandish landscapes, the ancient smelters had to come up with some peculiar solutions. For example, in Viking Age and Medieval Iceland, bloomery furnaces were supposedly built of turf, lined with basalt stones and lastly smeared with silty clay.

Why, you might ask?

Due to the youthful, vulcanic nature of Iceland’s geography there is practically no clay on that island, since that is formed in the long term weathering of rocks. The selection of stone is limited as well. But there is a lot of turf going around. In any case, if somebody actually tried to to smelt in this manner, then let me know. I am interested in how that went.

A variation on the second method is also used by some modern smelters, who in search of a quick furnace, build the base or most of the furnace out of firebrick.

salme viking market bloomery irn smelting furnace
A furnace with an embanked stone base and short, tapering, clay shaft. Built for smelting limonites and bog ores. Smelting course at the Salme Viking Market 2016. (Photo: Ragner Saage)


How many shapes can a chimney take?

You will find that there is more than one way of building a shaft. Therefore furnace shafts come as:

  • Free-Standing or Embanked
  • Straight or Tapered
  • Tall or Short

The first and most striking will be the choice between a ‘free-standing’ and an ‘embanked’ furnace.

A free-standing furnace is self explanatory. The furnace stands alone like a chimney. This is the type of furnace most commonly used by bloomery smelters today. When the shaft is meant to be tall (more than 1 meter), then the construction process requires considerable skill and patience. You might even need some kind of internal support, so that the whole thing does not go pear-shaped and lop-sided while the clay is still wet.

Embanked furnaces are built into hillsides and natural terraces. This means that the shaft is supported from the back and sides, while only the front is open. Once you cut the space for the furnace the into the bank, then building the rest is a fairly quick and easy matter.

This type of (slag-tapping) furnace is very common in Central Europe 700-1200 AD, especially Eastern Central Europe (e.g. Hungary, Moravia).

bloomery iron smelting furnace early medieval moravia
A chain of embanked Moravian furnaces in various stages of decay and repair. (Photo: K. Lokajickova)

Beyond that the shaft can be either tapered’ or straight’. In other words, it can be conical like a volcano, which means that it is wider at the base and it narrows to the top. Or it can have parallel sides like a chimney, which gives it a more or less constant diameter.

The difference between the two is that the first option (tapered) makes it easier and quicker to build taller, since you need less and clay as you progress. The use of a lower shaft with straight sides on the other hand, allows you to remove the bloom from the top, should you want to play with that.

Lastly, the total height of the shaft may also vary based on the type of ore used, desired outcome and preferences of the smelter.

The taller the shaft, the more time the ore shall spend travelling the way from the top of the shaft to the tuyere. More time spent in the shaft means more time given for the ore to reduce to metallic iron, or even pick up carbon. On the other hand, a larger shaft means that you might also need to pump in more air to keep things moving. There is also a slightly higher chance that something may get stuck.

Easily reducible ores, such as bog ores and limonites, can perform well in lower shafts. I have found 40 cm of shaft above the air intake to be enough for such starting stock.

Something denser and richer, such as a magnetite or haematite will require a slightly taller stack, should you want to unlock the ore’s full riches.

Should you want to focus on making high carbon blooms (bloomery steel), then you will also want to make your stack slightly taller. This will ensure that your ore is well reduced and ready to start picking up carbon by the time it reaches the hot-spot.

As a rule of thumb, when selecting the shaft height, you may go with a 40-50 cm of shaft above the air intake as the starting point and then see if further adjustments are required.

bloomery iron smelting furnace early medieval ireland
An early medieval Irish furnace, used during my experimental research at University College Dublin. It features a low, tapered free-standing shaft, set above a shallow slag bowl.


Are we sculpting Fortress or an Eggshell?

Furnace walls come in two varieties: ‘thick’ or ‘thin’.

On one end of this spectrum you shall find furnaces with walls no thicker than 5 cm – often even thinner towards the top. Essentially eggshells with violet flames gushing out at the top.

On the other end there are stout, immovable, objects of clay, with walls 20, even 40 cm thick.

We know that both were used in the past, with quite some shades in between as well.

The extremes illustrate the two way ways of coping with an inherent challenge of all bloomery furnaces. Namely that they are clay and sand structures, which have to contain an intense heat. They have to run for hours on end at temperatures, which lie very close to what even the best metallurgical muck can take.

A furnace melting is a sign of a smelt failing. Therefore this needs to be prevented.

Furnaces with thin walls manage to deal with the heat by dissipating it quickly.

Furnaces with thick walls accomplish the same thing by providing a thick insulating layer and some material which may melt before the damage is severe.

In practice, all furnaces will tend to melt and vitrify somewhat during the first part of the smelt, when they coming up to proper working temperature. But they should also soon reach an equilibrium where the melting will stop spreading. Once the furnace is broken in this way, the damage should not spread much in the following smelts. Therefore trying to replace what has molten away during a smelt may often result in only more melting.

bloomery iron smelting furnace medieval france
The reconstruction of a thick-walled dreadnought of a medieval French furnace. This thing was a monster – consuming ore and charcoal by the crate. Fête du Fer 2015, France. (Photo: B. Beranek)


Built with care or made by force?

At the end of the smelt, you have to somehow get the bloom out of the furnace. This is most often done by breaking open the front.

This again presents two possible solutions.

Option A: You tear out a hole in the furnace wall by using some kind pointed iron rod.

This gets the job done, but there is a high chance that you have cracked quite a bit of the furnace wall in the process. Should you want to run another smelt in the same furnace, then you might need to do considerable amount of patching up now. There is even a chance that you have (accidentally) completely torn the shaft asunder in the heated battle with the bloom.

Option B: You have a removable (sacrificial) clay door built into the front of the furnace.

The furnace is built with an arch, into which a door is inserted after drying. The two are never truly fused. Therefore once the time comes to open the furnace, you just break out the door, with (hopefully) minimal damage to the furnace.

As always, both solutions seem to have been used by the ancients. And as you can probably notice, I generally recommend the second method, since rebuilding the furnace after every other smelt is a pain. After all, you probably find more interest in smelting than construction clay shafts.

NOTE: I am here ignoring the hypothetical possibility of toppling or even lifting the whole shaft at the end of the smelt. I definitely would not recommend such acrobatics to beginners.

bloomery iron smelting embanked furnace door tuyere
The aftermath of a smelt. The remains of a furnace door with an embedded tuyere. (Photo: K. Lokajickova)

Air Intake

The tuyere dilemma. 

Every furnace needs some kind of air intake, so that a fire may burn at the bottom of the shaft. It is through this entry point that the air is forced into the furnace, thus raising the blaze to smelting temperatures.

A furnace may have one, two or several air intakes.

A single air intake is typical of most smaller (narrower) furnaces, commonly used by smelters today. This will generally cover smaller prehistoric furnaces, early medieval, Viking and medieval furnaces. When the furnace has a door, the air intake will be placed on the door.

When two air intakes are used, they will be placed on opposing sides of the shaft. This feature is seen on larger and wider furnaces, such as some of the prehistoric slag-pit furnaces (some of these even have 4 air intakes).

When several air intakes are used, they will be placed radially and evenly spaced around the bottom of the shaft. Again, this is a feature mostly seen on wider furnaces.

The use of more than one air intake seems to be an attempt at providing a more even heat and increasing the size of the hot-spot. Nevertheless, things are not so simple. The volume and pressure of air forced through each intake still has to be high enough, so that they add up n a single hot-spot and subsequently one large bloom. Otherwise, the use of several air intakes will just lead to several small blooms forming.

I have personal experience only with using single air intake furnace, therefore this guide will focus on that. Should you want to pursue other paths, then you might need to adapt my information, and possibly figure out a trick or two on your own.

This air intake itself will again manifest in two principal forms:

Option A: a simple blowhole – a narrow opening in the furnace wall or door.

Option B: a tuyere – a clay tube inserted into the furnace wall or door.

In each case, the diameter of the opening will measure something in the range 2-4, possibly 5 cm. I generally stay on the lower end of this scale.

Anything too large, and you might run into trouble with air and embers being blown back out of the furnace. Make the opening too small, on the other hand, and you might run difficulties when trying to point the bellows that way. Or when poking into the furnace to move away that pesky bit of slag, which just blocked airway.

Unlike what most people believe, there is no real need to use a tuyere.

They have a certain, annoying, tendency to crack when you start poking into them during the smelt.

Also any clay pipe that you insert deeper than 3-5 cm into a furnace, will almost surely melt after a short while. Therefore any idea that a tuyere helps move the hot-spot closer to the centre of the furnace is wishfull thinking at best.

A tuyere is also not necessary to set the angle of the air blast, since the same can be achieved by adjusting the angle of the bellows.

Nevertheless, tuyeres are found in archaeological contexts related to ironworking – sometimes even embedded in furnace walls. And should you look at some photos of me smelting, then more often that not, you will see me using a tuyere. How come?

One reason is certainly habit. This is how started smelting and this is what I got used to. The other is the way I actually use the tuyere.

First of all, I insert the tuyere only an inch deep into the furnace. (See above for reasons)

Secondly, I use it as a funnel which makes it easier to point the bellows in the right direction. Therefore my tuyeres strive towards a reverse trumpet shape. Since I do not insert the bellows into the tuyere (there is no need or advantage to doing so), the tuyere also functions somewhat as a valve. As the bellows are opened to fill up with air, the ambient air gets sucked into the tuyere, further decreasing the chance of flames being sucked out of the furnace and into the bellows.

In a nutshell: Use tuyere just barely reaching into the furnace, or none at all. It matters not. Either way works.

bloomery iron smelting forging archaeology tuyeres
A selection of clay tuyeres from archaeological sites across (early) medieval Europa. (Source: Pleiner 2000)

Air Intake Position

Regardless whether you opt for a tuyere or a blowhole in you furnace design, you will need to dedicate some thought to the position of the air intake.

You will need to have some space in between the tuyere and the bottom of the furnace, so that the slag has somewhere to go before it starts clogging up the air intake. 10-15 cm of clearance tends to be enough for most applications. With slag-pit furnaces, gap is already provided by the pit. With other types of furnaces, you will regulate this by changing where on the shaft install the air intake.

You will find that often you cannot increase this depth far beyond 20 cm, since at this point you run into trouble with a cold spot forming on the bottom of the furnace. You will recognize this by a layer of unburned charcoal at the bottom, with the slag solidified above it.

Lastly, the air intake should be slightly angled downwards, so as to help blow air towards the bottom of the furnace. A downwards angle of 20-30º shall suffice. This will help keep as much of furnace base as possible at a high enough heat, in turn letting the slag flow down, forming a bath, which will help with bloom formation.

A Final Piece of Advice on Furnace Design

When designing and building the furnace, it is best not to have the door, or the shaft just above it leaning into the furnace too quickly.

Having a portion of the clay furnace leaning over the air intake puts that part of the furnace dangerously close to the hottest part of the fire. This in turn increases the chances of the clay melting and giving you trouble as it starts dripping and obstructing the air intake.


For the best results, try to keep the lowest part of the front relatively vertical.

A Table Set With Variables

All of the variables of the previous section can be summarised in the form of the following table of common bloomery furnace variations.

bloomery iron smelting furnace variations typology
Call it a useful infographic or Harald’s Bloomery Furnace Typology. In either case it can be used to describe the vast majority of historically occurring furnace designs. At the same time, it allows you to come up with a plethora of variations on the spot.

I will now share with you some possible designs, which have proven to work well.

The Result: A Bloomery Furnace 4 Ways

Among the myriad of design possibilities I have chosen 4 types, which I have all personal experience with using. They have all proven themselves relatively quick and easy to build and reliable for smelting quantities of ore in the 15-25 kg range.

Because of these traits and my preferences, you will notice that the furnaces share many similarities in their dimensions. You could even argue that they are roughly the same type of furnace, constructed in different ways. A furnace slightly tapered shaft, which is 20-30 cm wide at the bottom and tapers in diameter and thickness toward the top. They may or may not have a slag bowl. They are all also designed with a door arch in mind.

They may all easily be modified to suit your personal preferences and situation. Therefore you can make some alterations to the diameter of the furnace, add height to the shaft, change the wall thickness or even add/substract a shallow slag pit.

Shallow Slag-Pit Furnace (The Master Thesis Furnace)

early medieval ireland bloomery iron production furnace design


I used this furnace design during my postgraduate year in Dublin, to produce iron for my research on The Ironworker’s Craft in Early Medieval Ireland.

It features a shallow slag pit, and a low free-standing shaft with tapering thin walls. The shaft could have been built taller, but I have found no need to do so when smelting bog ore.

The design is inspired by the remains of a 6th century furnace found at Grange 2, Co. Meath, Ireland. It is a small and efficient clay furnace – quick to build and easy to use. It is designed so that smelting sessions consuming 15-25 kg of bog ore can be run without slag-tapping.

Free-Standing Slag-Tapping Furnace

medieval viking bloomery iron smleting furance design

This is a narrow and tall furnace with thin walls. Built completely above ground, it is the archetypal slag-tapping furnace preferred by most smelters today. It is appropriate for most Viking Age and (early) medieval smelting operations.

The tall shaft allows richer ores to be smelted as well. So if you plan on smelting magnetite or haematite, you will probably want to build a furnace along these lines. Compared to the previous design, this tall shaft also requires a bit more skill and patience in its construction.

Stone Base Furnace

stone base viking medieval bloomery iron smelting furnace design

You should make use of this design if your local clay less than ideal. In other words: slightly prone to melting. By using heat resistant stone such as granite, you can build a very sturdy base in short time. If you want to further ensure success, you may form the door from store-bought or specially sourced fire-resistant clay.

Embanked Furnaceembanked early medieval moravian hungarian bloomery smelting furnace

I got acquainted with this furnace design when I visited the iron smelting meeting at Adamov, Czech Republic. The iron smelters working there almost exclusively used furnaces like this, since they were based on local finds from early medieval Great Moravia.

One thing that can be said about an embanked furnace is that it allows you to be very sloppy with the construction and still get away with it.

Later I found out that since this furnace type is very practical to built, it keeps popping up across Europe at different time periods. Nevertheless, they are most often associated Eastern Central Europe in the Early and High Medieval period.

The Takeaway

There are variations and types of furnaces out there and they can all be made to work. Some are just easier to work with than others.

If you are dabbling in bloomery smelting because of an archaeological research project, then you will definitely be trying to replicate a specific find or a group of finds.

But if you are just interested in producing some bloomery iron and getting some experience with smelting, then you will do well to choose one of the presented furnace types, and sticking to sensible dimensions.

Have I missed your favourite furnace design? Do you know of something even more peculiar then an Icelandic Turf Shaft? Got any questions? Do not hesitate post in the comments bellow!

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

Leave a Reply

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