A proposal on how to line up the overlap of riveted rings

The following photo essay came about as a result of discussions on the Yahoo Clubs riveted maille forum.  

Fitzlaughlyn noted:

"I've always been under the impression that the rings aren't round because the weight of the maille pulls the rings out of round."

I always attributed the out-of-roundness to wear, as well. I no longer do, though.

I have yet to see any maille example with perfectly round rings. It is possible, of course, that all the surviving rings have been deformed over time, and that none of the round ones are left. However, I have examined a lot of maille rings specifically that would be under little, if any load over time. For example, the rings at the bottom of a nasal on a coif, or the rings long the bottom of a shirt, or at the bottom of shirt sleeves. These rings bear no load other than supporting the weight of themselves, which is of course very light. And yet it is often the case that these rings exhibit the same physical characteristics as the rest of the rings in the garment. Except, of course, in cases where the rings in these areas are obviously meant to be different, either through intention or by later addition (i.e. lengthening of sleeves).

In summary, if the ring out-of-roundness was due to deformation under load, we would expect to find the rings around the top of a coif to be more out-of-round than those near the bottom. Likewise, the rings near the shoulders of a shirt should be more out-of-round than those near the bottom. Though I have no actual statistics for out-of-roundness compared to location in a garment, my general impression is that this is not the case.

Because there are no examples of maille with perfectly round rings, I now discount any theory which creates an overlapped ring that is perfectly round.  Thus I no longer put much stock in the theory of the funnel tool for forcing rings into an overlap condition. This process, too, would result in perfectly round rings, unless, of course, the bore were out-of-roune, and then we are back to the problem of exactly positioning the ring in the bore so that the overlap occurs at the desired place.

Of course, this means that cutting the rings from a coil with the overlap built-in is also incorrect.

On Left: Authentic rings.  Note ovoid shape.  On right: Reproduction rings cut from mandrel with overlap built-in.

I think it is more likely that each ring was simply overlapped by hand, the way that Erik is doing it. One can get quite proficient at it simply using their fingers and a small hammer.

The critical part of forming the ring overlap is that the two ring ends must lie very closely over one another. If the ring ends do not lie very close to directly over one another, when you attempt to crush them into one another during the flattening process the ring ends will either skip off of one another, or they will cross one another like the letter "X", only flattening where they cross. Neither condition is acceptable.

This is why I very much liked the method of cutting the rings from the coil with the overlap built in. By so doing, the entire ring maintains the same radius. Thus the ring ends match exactly the same radius, and thus lie directly over one another. The problem, as shown above, is that the rings end up exactly round. 

So regardless of how the overlap is formed, the critical thing is that the ring ends line up well, or you will likely experience one of the failure modes shown above.

One way to ensure a good line-up is to take your crudely overlapped ring (done by hand, with a hammer, whatever) and then stick a pair of small-jawed (needle nose) pliers up inside the ring. That is, one jaw of the pliers passes up inside the ring, and the other passes outside the ring, trapping the overlap in between. When the overlap is then squeezed between the jaws it thus flattens out that part of the ring and exactly lines up the two wire ends over one another. The ring becomes, in effect, D shaped, with the flat occurring where you squeezed the ring with the pliers.

I have tried this method, even with my perfectly-round pre-overlapped rings, and the result is much much more authentic looking than when I leave them round. I am thus not suggesting that it was, in fact, done this way, only that it does exactly line up the ring ends and produces something that looks better than what I'm currently producing.

I think it is plausible that the oval or D shapes seen are a result of some tool, similar to the pliers I use, being used to line up the ring ends.

A small, flat-jawed pair of pliers, used to "cross-squeeze" the overlap region.  This lines up the ring ends.

Round overlapped ring, cut from mandrel with overlap built-in

After "cross-squeezing" with pliers

After flattening

After punching

After riveting