Wire: Block-twisting Method

by Jamie Hall on July 22, 2010

In April, I did a post on wire-making tools, which briefly touched on methods of wire-drawing. There’s a proper discussion of some of the methods in Gold Bulletin: The Production of Gold Wire in Antiquity, by Andrew Oddy. This time, I’m looking at the Block-twisting method in detail.

In block-twisting, a square-sectioned length of wire is tightly twisted so that it’s four corners form a tight spiral which is then rolled between two blocks to smooth it and produce round-section wire.

– Kevin Leahy, Anglo-Saxon Crafts, Chapter 9, p149

As with the other methods I’ve outlined, using a draw-plate is preferrable in most cases, as block-twisting is a more complicated process, and much less exact. I’ve started with soft sterling silver wire, 0.95mm square. Ideally, this kind of work needs two people, each holding a rod to which the wire is attached – they turn the rods in their hands, in the opposite direction to each other; this causes an even twist along the length of the wire. If the wire is only twisted from one end, the twist tends to be tighter towards the worker. As I didn’t have someone to help me, I put the wire in a lathe chuck at one end and a pin vice at the other. I alternately hand-turned the lathe one rotation toward me, and the pin vice one rotation away from me – it wasn’t possible to do both at once, so the middle of the wire had a slightly looser spiral.

I eventually lost count of how many turns I’d done, but it takes quite a long time for the wire to lose it’s square cross-section, even when it looks quite round outside. As soon as the cross-section changed, the wire became brittle and granular, even with regular annealling – this appears to be a physical limit on this stage of the process, although it might be less of a problem with fine silver or gold. At any point up to this stage, the wire has a “barley sugar twist” which is decorative, and attractive in it’s own right.

Once the wire is annealled, the next process is rolling – the wire is rolled between two stone or metal blocks, causing the raised edges of the spiral to blend into the body of the wire. In my case, I used two steel bench blocks – I only processed a very short length of wire, but the result was spectacular. Almost immediately, the barley sugar twist became a round wire with a characteristic helical pattern still visible (0.95mm thick, the same as the original square cross-section). In Oddy’s article (see above), he suggests: “With very soft metals, such as pure gold, these grooves can be completely eliminated during the rolling process…”

Despite the lines, the wire is visually pleasing – the blocks I used were quite rough, and yet the wire had an smooth, even texture. Polished blocks would probably bright-finish the wire, and other textures might be possible using stone. Mechanically, the wire fared worse when bent, and they are probably best suited to straight lengths or very gentle curves. Bending a small circle caused the lines to open up, but the effect was acceptable. Right-angled bends, however, cause burring and deformation of the wire. This could probably be avoided by using high-carat wire, or annealling the wire before bending (which I didn’t think to do, although I usually would).

In future, I’d like to try a traditional starting material – I made the 0.95mm square wire in the rolling mill, but it should be cut from thick sheet, hammered square, or possibly even swaged into a square section. It would also be good to work in fine silver, instead of sterling – hopefully the metal would fuse, absorbing the lines better. I probably don’t need to twist the wire such much, and turning it evenly from both ends is important – if you look at the pictured sample, the spiral is much tighter at one end, where it was clamping. Cleaing up the blocks would make a noticable difference to the finish of the wire, which should be worked for longer. If I did all that, I might have a wire more suitable for bending.


Anonymous July 26, 2010 at 13:03

Jamie- again, you have a great blog entry with an enlightened moment of the past… Bravo and kudos to you… In the spirit of this experiment you may find solace in Herbert Maryon’s book Metalwork and Enamelling with its entries on wire twisting. The highlights are the use of the twisted wires in sword handles and decoratives.
-If you are going to attempt to fuse the spirals be wary of the grooves trapping impurities that will not allow the gaps to fill completely or trap gasses and may not roll out well. I have seen a version of a Mokume gane style twist done in your illustrated method- .25mm square 999 silver stock laid next to .25mm copper stock square… in a 1mm square piece checkerboard pattern (16 total pieces). Twisted, then fused, then rolled- then drawplated to wire size.
-I enjoy your writings- you should seriously consider a book from your thoughts and compilations as you make the processes of a bygone age available to those with no more tools- and do it rather “Primitively Eloquent”

Jamie Hall July 26, 2010 at 14:42

One of the issues with the above method is the use of sterling silver – it was oxidised when I rolled it between the blocks; as you suggest, the compressed spirals contain impurities due to the oxidation. The difficulty with oxidation, as I’ve discussed elsewhere, is the lack of strong acids in the medieval period – if I had pickled it after I’d annealled it, I would have needed at least a day of pickling in vinegar or similar every time.

My next experiment is hammering out square wire, which I’m doing with fine silver. If I can make square wire well enough, I’ll try to repeat the block-twisting experiment with that, and see if it solves the problem.

There are some of Herbert Maryon’s books on Amazon secondhand, so I’ll try to get some. Thanks for the tip-off. And the compliments ;)

tsanko October 18, 2010 at 15:14

Wonderful ..thanks a lot for posting a good informitive blog

larraine April 24, 2013 at 21:07


great to find such an interesting site —- I am studying silversmithing and struggling to make an even barley sugar twist
in silver strip (0.9mm thick x 15mm w x 400 long) with twists about every 2 / 3 cm. I hav’nt yet mastered a tight enough
twist (by hand and vice) without it tightening in one spot. Any good ideas?
regards L.

Jamie Hall April 25, 2013 at 00:05


It’s hard to say without knowing all the details, but there are a few things you could consider. First, is the strip too thin? What are the measurements? If the width:thickness or length:thickness ratios are too different, that could be a problem. For width:thickness, 1:1 gives you the easiest twist, and as you start getting beyond 3:1, twisting is going to become problematic – I’ve done more advanced versions of this experiment (or more primitive, I should say), and if you have a high width:thickness, it can actually start to curve along its width.

Second, how soft is the metal? As a rule, you work with metal that is well annealed, right? But that isn’t always what you want. Sometimes a little hardness is a good thing – when you apply force to annealed metal, the metal nearest the force will absorb most of the energy by deforming – but if it has a little hardness, the force will be transmitted along the length of the strip.

Third, how are you applying the force? Ideally, you want an assistant; you grip the strip with tongs or a swiss clamp at either end, and then each turn them to your left (or each to your right, depending on the way you want the twist to go). That will ensure a reasonably even spread of twist; if you do it all from one end, the twist will bunch at that end. I daresay that you’ll never get a completely even spread, but you really want to avoid it bunching at one end (unless it’s an intentional decoration).

Any help?

Jamie Hall April 25, 2013 at 00:18

D’oh! Forget that you’d supplied a measurement. 15mm x 9mm gives you a width:thickness ratio of nearly 1:16. That’s way too high. It’s not impossible to twist a strip with such a high ratio, but you might need to use a different method. I haven’t done it myself, but maybe you could try setting your rolling mill so that the strip just fits through without being rolled thinner. Then have one person turn the handle of the mill while you apply the twisting force close to where the strip exits the mill. In that scenario, you’ll want it well annealed, because you want all the force to be soaked up locally.

Yet another idea would be to jacket the strip; again, I’ve not tried this myself, I’m just thinking out loud for you, but if you had 3 strips together (using copper for the outer ones, if cost is an issue), you would get a better width:thickness ratio.

larraine April 29, 2013 at 23:04

thanks Jamie
the strip measurement is 15mm wide x “point” 9mm. Some interesting ideas – rolling mill
sounds worth go. I’m practising it all in copper at the moment. The lack of annealing makes sense.
I quite like the look of the curve created (you do mean in section do you?)
I’ll chat about your advice/ ideas at college – great help thanks. I’ll let you know how it goes!

Comments on this entry are closed.

{ 2 trackbacks }

Previous post:

Next post:

WordPress Admin