[Matthew Amt]

Bingo. Thank you, Randall!

Matthew

[William P]

personally i cant see how this process of cutting washers from an iron tube would be cheaper than punching solid links from iron sheeting..

i mean it seems like to make links out of piping, your taking this extra step and employing a crapton of new processes and steps and tools and skills needed to turn the iron into a tube, and to then make consistant links out of that tube..
especially if its meant for making the hundreds of thousands of maile links needed for the tens of thousands of hamata worn by the troops.

[Ralph Grinly]

I dare say that TODAY we could make solid rings from lathe cutting a tube. But I'm not so sure it would be possible to produce them IN BULK in Roman times. All the discusion here sofar seems to centre around whether or not the romans could even make iron (or steel) tubing of the required size. I'd focus more on the OTHER side of the claim..did the Romans even HAVE lathes that were capable of working down to millimetre tolerances, esp with cutting tools ? Don't forget..every hamata needs many thousand of links, 30 ? -40? thousand maybe ..or more. Assuming half of these are solids..thats 15-20 THOUSAND very similar rings PER Hamata..multiply this by the many thousand hamata's needed for a legion and you are looking at a pretty massive industrial capacity in lathes, all working to pretty similar tolerances. I just don't think the roman's, for all their organisational abilities were up to the task Working up links form punched plates, or drawn wire would be much more cost effective, I believe.

[Jean Thibodeau]

[Aleksei Sosnovski]

[Matthew Amt]

[Dan Howard]

Not according to the summary that Christian posted. I've refrained from speculating further in this thread untl someone else reads the article and confirms its findings.

[Greg Thomas Obach]

hello, my 2 cents

punching holes from sheet- you can simply cut it up and add it to your next bloom furnace charge
saw cuttings from tube- save it and add to next furnace run

you don't waste anything with either way

the material they used was wrought iron ( as was mentioned ) it has many sort of fibers in it like a piece of green wood
- it likes to split along the fibers... this can be very frustrating
- you'd want the fibers to flow around the ring for strength

unless you really fold and refine the wrought many times ... then it will act more like the iron we know today ... as the coarse pockets of slag will be finely distributed ... then it will act less like green wood

the rolled tube and forgewelded would be the easiest and i believe would be the strongest ... the ends of the tube have to be scarfed ( tapered ) so that it will forgeweld evenly and look seamless

i've worked with a decent amount of wrought iron and Phos containing wrought iron... it is a cool material but does require attention to the character of the metal when constructing shapes ... otherwise it may split when used

eg... chain links were made of round stock with scarfed and welded ends.... and not of plate with the centers punched out ...

Greg

[Johan Gemvik]

I believe future science in this field will show many variants of solid ring making was in practice thoughut the history of maille.
As modern tests show it's possible to get these results in a number of ways, from punched plate, forged from wire, and why not cut tubing as well. Each have their strengths and weaknesses and requre different skillsets and tools.

[Johan Gemvik]

Dan, I've studied the Gjermundbu pretty thoroughly, but we're talking Roman maille here where you're the prime expert.

On the Gjermundbu maille* the solid rings are drop forged to be rounded off. Eriks closeup shows sharp corners, but I suppose this is a replica he made. Are the original roman mailles that way too, sharp cornered (and worn down by wear) or drop forged to be rounded?

I found this image from the Canterbury find (http://users.stlcc.edu/mfuller/Canterbury/Arbeia.html) but I can't really tell from the photo if the edges are just worn down or made that way.



Is there any research done on this?


If a ring is drop forged post cutting a lot of the look to it will be from the final forging.


Pic of the Gjermundbu maille





* For those who may not befamiliar with the Gjermundbu maille, it's from viking age but made with very small but thick links similar to the roman style maille.

[Dan Howard]

Vegard was convinced that Scandinavian solid links were punched from plate. Until Greiner's article I thought that all the evidence for Roman mail was for punched links too.

[Sean Manning]

I have now read Greiner's article, although my German is limited so I am not sure if I understood every subtlty. Here is a summary. I may post two or three key paragraphs of the German later.

The first page summarizes the advantages of Roman mail, suggests that its widespread use implied a system of mass production, and discusses changes in the form of mail over time. He notes an example found at Rainau-Buch with a cavalry helmet in a context dated 229-254 CE. He calls the larger riveted rings with a rounded cross-section hamae and the smaller solid rings with a rectangular cross-section circui citing a poem by Sidonius Appolinarius. In general he is not cautious about extrapolating from the Rainau-Buch lorica hamata to all mail in the Roman imperial period, and this weakens his argument.

The second section describes the production of hamae, following the usual “make wire, wrap it around a rod, slice the wire into rings, rivet the rings closed” theory.

The third section presents the existing theories of solid ring construction: either they were punched from sheet, or they were welded from wire. S. Drescher showed that mail found at Sörup had its solid rings constructed by welding. Its welded rings had a round cross-section and a delicate seam. The Rainau-Buch circuli have neither feature. Circuli with a rectangular cross-section have traditionally been assumed to be stamped from iron sheet, with D. Sim describing the hypothetical process in detail. Stamping iron faces special technical challenges because iron is a hard material. He argues that if the rings were made by punching first a narrow hole to stamp out the hole then a broad one to stamp out the ring, the holes would not always be centred and there would be burrs of stretched metal facing up on the inner rim of the ring and facing down on the outer ring (so that their cross section resembled an integral sign). He can neither find these fragments on the Rainau-Buch armour nor find marks of filing to remove them. He also objects that neither punches nor dies nor stamped sheets of metal have been found, although I would expect that the scrap sheets were sent back to the smelter.

The fourth section proposes his own theory for the manufacture of solid rings. He looked for a technology which was common and highly developed in the Roman empire and thought of lathes. He cites examples of lathed stone, wood, bone, bronze, glass, and iron in the Roman world. He suggests the following process to make rings identical to the Rainau-Buch circuli: Take an iron bar, lathe it into a cylinder 8 mm in diameter, drill out the bar from end to end with a 6mm bit, turn the tube against a cutting blade to make slices 0.6mm thick. This leaves a trace of metal on the inner edge of the rings, but none on the outer, and this resembles the Rainau-Buch rings. He also notes that the Brokaer, Vimose, Thorsberg, Hedegard, and Kastenkov mailles have rounded outer edges (which could be produced by a cutting blade with the right shape) and that when one ring of the Brokaer mail had its cross-section taken it had a burr on the inner edge.

The fifth and final section describes his experiments making rings with a 50 year old lathe. He produced 35 circuli in about half an hour, and believed that he could have doubled this rate with practice. Therefore, the 15,000 solid rings of a small lorica could be made in 11 worker-days at 14 hours a day and 100 rings per hour. He suggests that another 9 days would have been needed for the riveted rings, allowing this hypothetical factory to make a lorica in three worker-weeks.

My opinion of this theory is of little weight, since I haven't read the necessary scholarship and am not a smith. The advantages are that it would have easily produced uniform rings which resemble the Rainau-Buch circuli. The objections seem to be that it would waste 65% of the iron bar, produce its waste in the awkward form of dust, and use up a lot of expensive steel drill bits. The pre-industrial ironworkers I am familiar with were very reluctant to waste good iron, because it was expensive relative to their own time. He doesn't seem to propose any change in construction methods over time, and he does suggest that the tubes were bored from solid stock rather than being forge-welded around a mandrel then smoothed with a lathe and centre-bore. I don't think that the later method would be incompatible with his theory, and it would reduce the waste in his process.

[Randall Moffett]

Sean,

Thank you! I think you answered all my primary questions and your conclusions seem fairly solid. I guess someday when I get German under my belt I might take some time to look it through but I had wondered about what his basis was.

I have punched disks from plates and never had any of the issues he is talking about.

Having used a metal lathe before I would never have made millions of rings in this way if I could punch them. Maybe the Romans even had a stamp that did multiple rings for all I know.

RPM

[Jean Thibodeau]

[Mark Shier]

Why would the Romans build complicated machinery when they had cheap or free labour? Slaves if the rings are being made by a contractor, soldiers if it's being done in-house.

[Jean Thibodeau]

[Randall Moffett]

Mark,

Good point. Several technologies never advance much because of this same factor.

RPM

[Sean Manning]

[Craig Shackleton]

One point that kind of baffles me; the conclusion is that it would take longer to construct the solid rings than the riveted rings. I realize this might not count the riveting time, but it still doesn't seem efficient. Especially because it means that there are two processes now, and the extra process is the inefficient one. Stamping seems like it could be much faster, and by using a rolling press punch, which is IMO no more technologically difficult than a lathe, it could become tremendously faster. I can easily speculate a rolling press that would punch the hole then the ring sequentially as the iron feeds through the roller, keeping the hole centered nicely.

I believe the lathe method is technologically feasible, but right now it seems an implausible choice.

[Sean Manning]

Yes, he doesn't seem to justify that, but its a short article. When he says 30 seconds per riveted ring (geschlossenen Ringe) he may be just talking about the time to insert and rivet the rings but I'm not sure why the time to make the solid rings counts but the time to make the riveted rings does not. The verb he uses is is auffädeln "to string."

I can't find much about the author online, but he seems to be an archaeologist who has worked on Rainau-Buch. So this might be a case where someone entering a new field has original ideas but brushes over complexities that a specialist would notice. I would be interested to know what Erik Schmidt thinks of this theory.