A payback challenge got me into the business of making a lot of metal armor. My first decision was to make a knee length, half sleeve mail hauberk, which would have been considered an extraordinarily long mail coat for a norseman. Fortunately there is some justification from The Heimskringla in references to Harald Hardraada's mail shirt, "Emma," which was so long it was "nearly down to his knees."
The only mail shirt I know of from the Viking Age is from Germundbu, and, fortunately, there is some free online information about it.
The "ring diameter" of the Germundbu shirt varied, but 5/16ths inside diameter is a reasonable approximation. Data from five riveted rings showed a 7.8mm average ring diameter;; the "unknown rings" averaged 8.4mm. I assume the values given were inside diiameters; if so, 5/16ths inch (7.9mm) nominal inside diameter for mail rings seemed reasonable.
The wire used in the Germundbu shirt is not round in cross section. Riveted ring wire thickness varied from 1.1mm to 2mm. I used 1/16th inch (1.6mm) mild steel round wire.
I chose not to rivet the rings because I have better things to do with my time (metal arm and leg armor for example).
I tried to buy wire from a sadly misinformed local welding supply store (guys: Crown Alloys hasn't made that wire for more than two years... quit telling people you can get more of it overnight!) and ended up ordering thirty pounds online from the Ring Lord at a very reasonable price. With the benefit of hindsight I can't recommend the "expedited delivery" option, unless it comes with frequent flier miles. Eventually the wire got here, and the folks at the Ring Lord handled the problems with shipping promptly and profesionally. I'd buy from them again... and skip "expediting."
For a winding mandrel, I used a slotted 5/16ths rod. Compared to a mandrel with a drilled hole this saved some time because it was easier to remove finished coils from the mandrel:
At the end of a coil you cut off the wire supply, wrap the last few inches, then slide the coil off the end of the mandrel. Getting started and controlling coiling rate took a little practice, particularly when I got impatient and switched from my twenty year old 600RPM drill to my much newer 2400RPM drill! Here's an end view of a wrapped coil showing where the wire went through the mandrel slot:
I didn't double-wind the wire despite advice from knowledgeable friends. The theory is that the resulting gap would make it easier to knit, but in reality I open my links a lot more than 1/16th of an inch to knit them into place; I don't think a double-wind would help me much.
Double layered wrapping, where the wire wraps onto the mandrel and then by mistake wraps a second layer on top of the first, can be prevented by making a jig with a wire feeder hole less than one wire diameter from the outer diameter of the wrapped wire coil. Here's a view from the drill side of the mandrel:
Other than the 5./16ths rod, my wrapping jig consisted of a 2x4 in a bench vice with a drilled piece of angle iron screwed to it for a wire feeder.
A thirty pound, three foot spool of wire turned into ninety or so springs, each about eighteen inches long and 5/16ths of an inch on the inside:
For cutting links the fastest method I found (using tools from my basement) was aviation shears. Depending on which set of shears I used, I could cut 4 to 7 links in one stroke. Unfortunately the links furthest up the blade end up distorted in proportionally to the number of links cut.
Cutting four links at a time:
and here's the most distorted of the four:
I think the time save on cutting links is still an advantage even considering some of the links needed an extra squish during knitting.
I'm a big fan of the connect-the-chains method:
(1) build a lot of quintuplets (I start by simply closing a whole lot of rings)
(2) make the quintuplets into long chains (~65 quintuplets to go from the front of my knees, over my shoulder and down to the back of my knees). It helps at this point to have a lot of open links:
(3) then connect the chains into mail. The combination of link spiral direction and mail direction of lay makes a difference in how easy it is to thread a new link into the mail.
Is this "the fastest way?" Personally I'd bet "the fastest way" is to get somebody else to make the mail for you. Second fastest is to work on your mail a lot and get some help (thanks, Rollo, Gregoire, Odo, and Sunnifa!). Judge for yourself, here's what got done in one week:
Here's status at the close of week two:
Week three... I'm running out of patience, better get this thing done pretty soon...
Some information shamelessly plagiarized from Stephen Francis Wyley's web page, "The Gjermundbu Mail Shirt:"
Dimensions of rings measured (measurements are in millimetres).
Riveted rings: Ring diameters; 7.4, 7.7, 7.7, 7.7,8.3. Wire; 1.09*1.24, 1.35*1.45, 1.38* 1.38, 1.39*1.68, 1.44*1.50. (~3/64" to ~5/64)
Unknown rings: Ring diameters; 8.35 (sd=0.3, n=5). Wire; 1.2*1.4, 1.7*1.5, 1.7*1.5, 2.0*1.4, 2.0*1.52.