Warning: this was written for alpha only, and some bits don't translate well to The Net. I will try and get it rewritten, with maybe some graphics added in, someday.
I personally recommend that you get the drill, the vise, and the hacksaw. Minimize effort. If you want a good product, use good tools. Drills, hacksaws, and vises are all useful for a LOT more than just this -- they're like can-openers. A thousand household uses.
If no drill: Wind springs on 3/8 rod. This is done by drilling (with a borrowed drill, or drill press in a student shop somewhere) a 1/8" hole through the rod, about 1 foot from the end of the rod. This hole easily holds 14-gauge wire, when winding. To get the torque to wind the wire, put two bends in the short (1') section of the rod, to approximate a handle shape, and put the bearings on the rod to support it. Keep them in place by wrapping masking tape around the rod. Clamp the bearings in the vise and wind using the handle. (Ersatz bearings can be made by drilling a 3/8" hole in a piece of wood and clamping that in the vise or bolting it to something if you don't have a vise. This sucks but it works.)
If you have a drill: drill a 1/8" hole in the 3/8" rod, about 2" from one end. Chuck the rod in the drill and wind using the drill motor. (Don't overheat the drill. Be especially careful if it has less than 1/2 horsepower.)
If you have a drill and a vise (optimal): drill a 1/8" hole in the 3/8" rod about 6" from the end. Either clamp the drill in the vise (be careful) or put two bearings (3/8" x 1") on the rod immediately in front of the drill and clamp those in the vise. The latter method is much nicer to the bearing surfaces in the drill.
Warnings: ALWAYS wear leather gloves when winding and cutting wire. You could easily get a finger amputated by getting it caught in the wire while winding. Always wind the wire in the same direction. You will go absolutely crackers if you've got different windings mixed. Trust me on these.
Springs to rings:
If you have a vise: clamp spring horizontally in the vise, and saw down the side of the spring, perpendicular to the tangent. With a bit (well, a LOT) of practise, this produces beautiful, O-shaped rings with only a little gap to be closed up later. The cut should be as perpendicular to the wire as manangeable.
Saw with moderate pressure on the cutting stroke and almost no pressure on the return stroke. Wax the blade occasionally to prevent overheating. Use Lenox blades -- they last forever. I recommend using safe-jaws, pieces of bent sheet steel or aluminum, to cover the vise jaw faces, to guard against the inevitable slips. Tape a box to the vise below the cutting point to catch rings as they fall.
Without vise: clip rings off spring with clippers or bolt cutters. Wear a glove to prevent blisters on your hand.
Now, you have a box of rings. You need: about 10,000 rings for a typical shirt, which is roughly 60 feet of spring, or about 990 feet of wire. This weighs about 20 lbs. A coif is about 5000 rings, chausses (pants) are about 7500 rings.
There are places which sell rings in bulk, but it's not cheap. Chainmail.com seems to have better prices than most. Stainless steel wire looks nice, but if you ever really work your chainmail out, you'll find that stainless steel rings tend to shatter under weapon impact, where mild steel rings bend and absorb the blow. Bending and cutting the stainless steel is also a pain. Copper, aluminum, and bronze also work well, but are usually a bit soft, and tend to open spontaneously. They can be soldered, brazed, welded, or riveted, to ameliorate this problem, but those are subjects beyond the scope of this discussion.
Manufacture of mail from rings:
This is a very difficult thing to explain without the use of either live aids or pictures. (Some pictures added recently.) I am going to describe the manufacture of butted 4/1 mail, in which each ring connects with four others. This is practically the only pattern that was ever used in Europe. Asian mail had a LOT more variety than this, but that's a subject unto itself.
I prefer the latter. Everyone else in the universe prefers the former.
The primary thing to keep in mind at all times, when learning to make mail, is that each ring intersects four other rings, except for at the very edges. Chainmail has a definite warp and woof -- an x/y asymmetry in stretcheability. Look in a good dictionary and see if you can find a picture and xerox it -- that will help immensely.
There are two different types of rings in chainmail, ones that slant to the right, because they are held that way by their neighbors, who all slant to the left, and the left-slanters, who are held by the right- slanters. When you look at the edge of a piece of mail, it makes a zig- zag pattern of intersecting rings.
While you're pondering on that, grab some rings. Link them together to form a long chain, and close the links. If they're cut right, you should be able to slip them together without using the pliers, and then close them afterwards. Keep the Henry Ford principle in mind here: do one thing lots of times and then shift to doing another thing. It's more efficient. I make 30 five-foot long sections of linked chain in one session and then close all the rings in all the sections the next night. But for the nonce, make two chains that are each 21 rings long. (there's a reason for this unusual number, that I'll explain later.) The process of closing rings is tricky -- instead of crushing them together, you grab the ring with the pliers on one side of the gap, grab them on the other side with the other pair of pliers (or crescent wrench) and combine a bit of inward pressure with twisting. I usually twist the ring from its inital open state, PAST the closed state, into the open state on the other side, and then finesse it back closed, to get the nicest-looking closure. Getting this so that the gap is minimised and the ring is flat instead of potato-chip-shaped is a matter of practise.
Now, the first critical step: lay one chain down so that it is flat. When I say flat, I mean that every other link slants in the same direction; the links alternate tilting directions. The links at each end of the chain should be tilting in the same direction, because the chain is an odd number of links long. If you look at this chain end-on, it should make an X shape. It takes a bit of time getting used to laying chain out this way; rotate the end as you lay it out, if necessary, to get it to lay flat.
Here is a picture of two lines of mail lying flat in big picture form.
Second critical step: Lay down the other chain so that it's identical to the first one. The terminal rings should tilt in the same direction as those of the first chain. Carefully scoot the two chains together so that they overlay each other slightly, making sure that none of the rings flip as you move them.
Here is a picture of two lines of mail all snuggled together in big picture form.
Third critical step, and by far the ugliest step of them all: Link the two chains together. Open a free link to about three wire diameters, (that's about 1/4"), to form a helical segment, and, using a corkscrew movement, wind this open link through four links of the chains. I'll try to describe this more. You want to connect four links, two from each chain. These are NOT adjacent links. You are connecting alternating rings -- rings that all slope in the same direction. So, if we number the rings in our chains from 1 - 21, from left to right, and we have chains U and L, like so:
1 2 3 4 5 6 7 8 9...21
U: O C O C O C O C O (I am notating the rings so that O rings
L: O C O C O C O C O slant one direction, C rings another.)
I am assuming that you have the chains set out so that the ring U1 (an "O") has its lower edge on top of the upper edge of L1. If this is not true, ie if L1 is above U1, rotate the whole works (the table or book or whatever) 180 degrees so that U1 is over L1. (This inverts the order of the numbering, of course, since you're now assigning the moniker L1 to what was U21.)
What we are attempting to do is connect, in order, rings U1, L1, L3, and finally, U3. U2 and L2 should NOT be intersected or included.
Here is a picture of a single ring just put in place in big picture format.
Then, after closing this (and totally messing up the entire pattern, no doubt) we will reflatten the chains and connect U3, L3, L5, and U5, in that order, with the next ring, and then U5, L5, L7, U7, etc. This order of ring-linking is quite important. Then just continue doing this, adding one chain at a time, until you either finish or go mad.
It helps, when you've become proficient at this, to lay all the helical connecting links in place, and carefully lift the entire row by the helical link on one end, under tension, and then close them all, one by one. Again, the Henry Ford Principle: screw for a while, then close them all, in batches.
If you cannot get this to work no matter how hard you try, I offer an alternate technique that doesn't work for me but does for other people: make your initial lengths of chain, alternating one ring and then two rings and then one, etc. I represent this as o=o=o=o=o=o=o where the o is one ring and the = two rings. Then try the same process as listed earlier, laying them flat-wise so that they alternate slant directions. (ie all single rings tilt up towards the top and all double rings tilt up towards the bottom.) Then try to interconnect just the double rings with another identical chain, in the same manner listed, ie U1, L1, L3, U3. (this is assuming your chain ends in a double-ring. Don't bother trying to interconnect single rings. They're already connected to four rings so they're no longer of interest. Only try to play with rings that have space available.)
Confused? Like I said, this is not an easy matter to teach in a hands-on manner, much less a purely alpha graphics medium. The process involves stacking one chain on the end of the growing block and adding it on each time. As you'll soon notice, this involves alternating the pattern of adding corkscrew rings, as you'll add (if you're doing the single chain addition technique using chains 21 long) ten rings to one run, then 11 to the next, then 10 again. With a bit of practise, you can load all 10 (or 11) corkscrew rings in at once, and then CAREFULLY pick the line up and close them all in one pass, which speeds things up TREMENDOUSLY. But you invariably end up dropping the line on the floor, or having a bunch of rings fall out of their places so that they're only encircling three rings instead of four, when you start this, so it gets annoying.
Why 21 rings long? I hear the question resounding from the furtherest corners of the Earth. Well, you remember my claim that there are two types of rings, leftists and rightists. When you're making mail (we're voyaging into motivational techniques here) you are going to get VERY depressed, at about the time you've got two square feet of mail and are adding these diminuitive quantities of rings on. The project ahead seems overwhelming; there is no sense of progress. So what I recommend is that you sit down with groups of 210 rings each. Well, don't bother counting, just grab a big thirty-foot-long length of chain, and from it, cut seven 21-link-long chains. Adding corkscrew rings, turn this into a nifty little block of mail, measuring 10 rings along the edge and 21 along the length, which is 210 rings. The astute among you might have noticed that the number of corkscrew rings is 1/3 the number of rings contained in the chains of 21 -- which is unsurprising. Anyway, when you've made about forty of these little chunks, you can sit down and assemble them into a shirt, and it's nice to just watch this pile of chunks grow. Stack these squares on the floor. Pile them in interesting formations. It gives one a sense of accomplishment lacking from adding one ring onto this huge piece of mail, then adding another one, etc. It is also a LOT easier to carry these little pieces around.
Back to the 21 question: well, if we made each of these chunks 20 rings long, so that they would each contain 200 links, we would end up with two varieties of blocks, randomly determined by how we add corkscrews. One would have rightist links along both ends, and the other leftists. By ends, I mean the termini of the chain segments you've linked. The rightist blocks and the leftists are distinct, and a pain to connect to each other. If you don't believe me on this, play around with making some patches that are only, say, six or eight chainlinks long, and see what I mean. An odd length guarantees that, from some combination of rotations around either the x or z axis, the chunk will match edges with any other chunk. (For any chemists out there, I *THINK* that an even-width, odd-length chunk of mail has a C1v point group, while an even-width, even-length chunk has a C2 group, implying that the latter is chiral (which is the entire point (pardon the pun) of this discussion, while the former, odd-length, is achiral.))
This is a very subjective and experimental area of mail. I personally have a 40" chest, so use ten panels of the above-mentioned patches for the body tube, which I make three panels long, or 30 panels in toto. The circumfrence is 105 rings. (10 x 10 + 5 rings used in stitching them together, because only every other ring counts, since half of them aren't added onto the top row but onto the second. Count it out, it'll make more sense.) Now, to make the straps that will convert this tube into a vest, I use these same patches of 210. I propose, for the purposes of this discussion, to use four whole patches, and use two patches made from bisecting edgewise one normal patch into two pieces, each of which are 10 rings on edge, 5 rings along the side. I connect all this together to form two strips, each 27 rings long and 10 rings wide. (why 27? Because we added two rows worth of helices, when we connected the three patches.)
Now, with a bit of math, (ie 4 strap attachments x 10 rings width/strap = 40, 105 - 40 = 65, and since I'd like the space between the attachments to be equal, coz it seems to fit well, 65 / 4 = 16 (drop the fraction) so I'll attach one strap at one end, completely, and count 16 rings, and attach another strap, etc. This will take some adjustment and fiddling; error towards extra space in the armpits and front spaces.) With this, I will have a vest. You'll get the hang of it. Then, make some stuff with diagonal edges to serve as the front neckline, and fill the back neckline hole, between the straps, up very near to the top. This involves just puttering about, and seeing what you can fit that still looks nice, and (IMPORTANT) fits over your head.
Sleeves are most easily made by continuing the material from the tops of the vest straps right down to about the elbow, and continuing them around to meet and close on the other side. However, this will leave a very nasty bit around the armpit, which doesn't connect at all well. Indeed, armpits have always been the downfall of mail, and doing a sophisticated armpit is quite difficult. I say, leave them open for your first suit. Just throw it together and enjoy it as it is.
The original ones just ran them together, perpendicularly, extending the material from the chest up past the armpit and then jamming them together just under the biceps. I personally put the ugly seam line in the armpit, against the pectoral muscle, because I think that helps the flexibility beneath the armpit, which is critical for comfortable fit and maneuverability. It isn't easy to make these work out such that you can put your elbows against your sides.
There are advanced topics, that involve gusseting and fitting in greater detail. These involve alternating series of rings that only intersect three others, and rings that intersect five others, which, in groups, act to expand or contract chainmail. The process of making these is that of making diagonal pieces and attaching them along the sides to normal pieces of mail, and is insanely complicated and mind- warping, so I am going to leave that for later. Ditto on riveting processes, and armpit work. I do include a helm pattern for chainmail layout, which is unusual in that it relies only on nice, diagonal patches rather than the sort of chaotic gusseting seen in more typical single-ring, diverging patterns. This serves as an introduction to gussets, which work excellently for following contours.
On the care and feeding of mail:
If you use galvanized steel, as in electric fence wire, it should not ever have a problem with rusting. If it does tend to rust, for lack of a coating, there's not a lot that is highly effective, that I know of. Coating it with heated wax works ok, although it tends to be messy on the padded shirt beneath it, and dipping it in lacquer works moderately well, but leaves a mottled appearance to individual rings, as a result of lacquer build-up at the contact points between rings.
References that have helped: Charles ffoulkes' "The Armourer and His Craft"; "Arms and Armour of the Medieval Knight" by David Edge and John Paddock, and a series of articles by Martin Burgess in the Antiquaries' Journal, circa 1957, about mail construction and design.
This page created on 6/4/94, Last Modified 1/24/00
Back to the Index