Caherconnell bloom forging 1

Preliminary Report

Yesterday I started compacting a bloom fragment recovered at the Caherconnell Furnace Festival in Ireland late August this year.

The bloom piece was found laying next the the furnace constructed by the 'Irish Builders' team (father and son) who were new to iron smelting. They undertook the build on Friday, initially working under my direction, until tension arose over their accepting instruction. This had resulted in a furnace considerably distorted from the standard layout used by others at the event. (1) They would return on Saturday and undertake several smelt attempts, working basically without supervision. Because of this, and their total lack of any record keeping, the exact method to produce the bloom that they then abandoned, is unknown.

The ore used by the majority of the working teams was Irish Derryarkin bog ore. I would call this an older deposit, found as long layers under accumulated peat (not the kind of fresh deposits as found at the stream edge at L'Anse aux Meadows). The Derryarkin is an extremely rich ore :

Ore             Fe2O3     MnO     TiO2     SiO2     Al2O3     MgO     CaO     P2O3
Derryarkin    94.5901   0.5823  < LOD    0.1656   < LOD     2.2201  2.0916  0.048 

You can see that the iron component as almost pure Fe2O3 (elemental iron @ 70 % = 66 %)

The standard method used at Caherconnell in 2023 was 10 kg of ore per smelt. Individual blooms were extracted and only roughly compacted, mostly striking off any clinging slag and collapsing to a pancake. I did not observe any of the teams working in the edges of their blooms. Generally teams were getting weights at that point of 1.5 to 2.5 kg, respectable return yields for such small ore amount smelts.

The bloom piece recovered was s clearly crumbly in texture and contained visible slag. It is unknown if this was the complete return from one smelt attempt, or piece broken from a larger mass.

 

It was small, only 570 gms total. The volume was measured via water displacement to 100 cc. This gives a density of 5.7 gm / cc, which is in the range of other blooms I have measured. (1)

As a further check on potential 'quality', the piece was cut in two, down the notch visible above.

After cutting, the total weight was 555 gms, one at 243 gm, one at 312 gm. These pieces are actually small enough to make manipulating them a bit difficult in the forge (in terms of being able to both hold them with tongs and still be able to hammer them).

The compaction process was undertook using a bituminous coal, deep fire pot, forge here at Wareham, employing a cavern styled fire . (In comparison, compaction heat at Caherconnell was using fairly shallow, coke, forges.)

The first step was to flatten each of the pieces, under the hydraulic press. These were heated to a normal welding heat, then quickly transferred over about 25 feet to the press table (so given the small size, likely means some dropping from ideal temperature).

The 312 gm piece flattened

After flattening, the two pieces were both roughly the same size, but the initial 312 gm piece was reduced to 8 mm thick, the 243 gm to 6 mm thick. For additional welding heats / hand hammering was taken on the flat faces (two either end) on both pieces. Then a set of four welding cycles to work the ragged edges in. During this work on the larger piece, a fragment of 42 gm cracked off.

The plates were shaped slightly to roughly match each other. The result was two small plates, 184 gm (from 243) and 196 (from 312). The total return at this stage was 76% (from bloom). At this point they were water quenched and taken to the grinder for spark testing. The larger plate was found to spark as relatively carbon free. The smaller plate had a large difference, one end at carbon free, the other showing similar to a mid carbon (1045) content.

After weighing, these plates were wrapped with a single loop of fencing wire to hold them together, returned to the forge and welded together via hand hammering

The majority of the wire would break off during the first welding course, and so is not considered to have added significant mass. Again a total of four welding heats were directed on the flat, four more on the edges. During this set, the bar would break into two pieces.

An attempt was made to compress / weld in the two ragged edges. The two pieces were forged to roughly the same size, with different thickness. Again these were looped together with wire, and hand forge welded together using a similar sequence of four / four. Over this process, the block would again fracture, several times.

gromp fragments recovered to right

The overall result at this point is one small bar, showing some edge cracking, and three roughly similar square pieces. The total weight of all is 275 gm, with a further 27 gm of small fragments recovered that broke off various edges during forging. This places the effective loss (into bar) at this stage at only 50% from bloom. 

The metal is continuing to be very crumbly in texture, prone to large fractures during forging. It needs to be pointed out that the metal was brought up to welding heat before each hammer cycle, and never worked below an orange heat. The reason for this cracking is not clear. The overall carbon content is low, which normally makes for easy welding up and drawing out. I had thought phosphorous content from the ore might be a cause, but on a return to the ore analysis, this element is in low concentration. (3)

This bloom is proving quite frustrating to work with, primarily because of the way it continues to break into small fragments. At this point (basically three compaction phases) it still remains a long way from a refined working bar, yet the losses are swiftly increasing. This compared to about two dozen other blooms I have rendered down into bars, where the usual return has averaged about 65%.

The pieces remaining are too small now to easily manipulate in the forge. The best next step, given the small pieces, would be to 'cheat' and MIG weld the fragments on to a long bar on one edge, then weld that tall stack together.

1) The layout for furnaces was set for a circular 25 cm ID, stack height in the range of 30 - 35 cm. The furnace built was an oval cross section at 30 x 35 ID. This changed the internal area from the intended 156 to 265 cm2. A significant change (and potential problem) put against the fixed volume electric blowers available.

2) A report covering 30 blooms created at Wareham over the last 15 + years shows the average density at 6 gm / cc : 'How Dense are you?'

3) It is important here to remember that the majority of blooms created at Wareham have used our red oxide analog, which contains no elemental phosphorus.