As you can see on the 'Courses' overview, one of the specialist programs I offer is 'Bloomery Iron Smelting'. As this is a very 'on the fringe' interest, I normally only offer this program typically every second year. Further, the program is complicated by the raw materials cost (over $400 worth consumed), so there has to be at least TWO paid registrations for me to proceed. This year, I was contacted by Keith, who was determined enough that he would fly in from Edmonton (!!) to participate. With that kind of enthusiasm, I committed to run the course. Ongoing conversations with Bill Short of Hurstwic lead to him registering two, eventual three additional people. (The Hurstwic team drove up from Southburough MAS.) Because all of these participants were keenly interested in specifically Norse / Viking Age methods, the normal use of the simpler 'Econo Norse' design was shifted to a historically based 'Norse Short Shaft' styled furnace.

This smelting effort marks the 77th time I have personally ran a bloomery furnace, so the layout and process is well established, only the personnel involved presented any 'random element'.

 - Clay Cobb Construction : the usual mix of equal amounts (by volume) of powdered clay, rough sand, and shredded horse manure.
- Overall size : 30 cm ID / 60 cm + tall
- Tuyere : cylindrical ceramic 'kiln support' - with ID at 20 mm and overall 20 cm long.
- Tuyere Set : roughly 15 + cm above base, at 22 degrees down, 5 cm proud of interior wall.
- Furnace constructed on a base plinth of standard bricks, the central space filled with charcoal fines.

The first course day consisted of mixing clay cobb and building the furnace itself.

Matt prepares clay cobb 'balls' while Barbara builds.	Detail of the build, rope (attempting) to contain the lower bulge, with sand / ash packing inside.	After the evening long drying fire - the body sculpted.

As you can see, the shaping of the furnace in damp clay did get a bit out of control.
In this case, no interior form was used to ensure a consistent cylindrical shape. Hand building such a large form is certainly more difficult than it may first appear - especially for those not experienced with working with wet clay. Eventually, the bottom spread was at least partially controlled by banding rope around the exterior. Packing material of 50/50 sand and wood ashes both helps to stabilize the structure, but also helps to pull some of the moisture out of the damp clay mix.
The end result was more a conical than cylindrical shape. There can be certain advantages to this during the firing sequence, but at the cost of making a top extraction difficult (if even possible).

While Matt and Barbara undertook the bulk of the heavy work of mixing clay and building, Keith and Bill were also busy preparing the other raw materials. There were 10 large bags of charcoal (80 kg total) to break and grade to the required size (between 25 to 5 mm). Also the prepared analog to break into suitable sized pieces (ideally 15 - 3 mm). All dirty work!

For this smelt, the well proven DD1A analog was used, with roughly 30 kg prepared:
- 22.5 kg red iron oxide powder (Fe2O3)
- 2.5 kg whole wheat flour (10%)
- 2.65 kg collected forge scale as enrichment (roughly 10 % Fe3O4)

Smelt Day - starting the pre-heat with wood splints, with the air system in place.	'Speaks with Fire' - about 2/3 through the smelting sequence.	Keith adding ore.
Nearing the end of the sequence ('Actually kind of boring - if nothing goes wrong.')	Starting the extraction - bracing the front of the furnace as the arch is pried open. (image by Neil Peterson)	Matt reaches to get the bloom mass, Barbara ready with hammer in hand ('Over to the stump' from me) (image by Neil Peterson)
The mass pulled free, ready to move! (image by Neil Peterson)	First compaction stroke by Barbara, while Matt holds. Note amount of slag still clinging to the bloom. (image by Neil Peterson)	Two strikers at work, still gentle impacts as loose material is knocked clear of the bloom. (image by Neil Peterson)

In keeping with the 'Nissen Method', the first charges were not of ore, but iron rich tap slag recovered from earlier smelts. A total of 5 kg of this material was charged, as a series of 1 kg additions. The huge advantage to this method is the fast establishment of a working slag bowl system inside the furnace. This done before the more valuable ore is added, typically resulting in a significant increase in possible yield.

For this smelt, the totals were :
- ore = 29.3 kg
- charcoal = 65 kg
- time = 6 1/2 hours, plus burn down and extraction (another 1 1/2 hours on pre-heat)
The full smelt data is available HERE

image by William Short

In the chaos of final exaction and packing up in early dark, I never got any suitable scaled images of the final bloom.
Final Weight = 11 kg
Yield = 36 %

Needless to say - this is excellent.
Part of the reason is most certainly the use of the 'Nissen Method', which in the past has been found to add anything from + 5 to + 10% to the final yield. Even still, with a 30 kg ore addition, our past results are more typically 20 - 25 %.

Later, Bill had the resulting metal analyzed, and prepared this brief overview of those findings - HERE
(as PDF - used with permission)