Squamata construction

[Crispvs]

Following requests from Moi (Vindex) and others I am starting a short thread on how I made my squamata.

First of all, the way I made it was in large part dictated by the hole pattern on the scales I had. This is why I opted to use linen strips to attach the scales (of which more later) rather than sew directly onto the backing fabric.

The first picture shows the type of scales I used, along with examples of the staples I used to join them side by side. All the scales in the cuirass are identical (aside from a few clipped edges here and there). The staples are each made from a length of copper wire slightly under an inch long.
I found that the cheapest way to buy copper wire was in electrical cabling which featured two side by side bundles each of seven strands of wire, separated by a single thicker wire, all encased in grey rubber insulation. One of my local hardware shops sells this cabling for £1.50 per meter, meaning I get fourteen meters of wire for staples for every £1.50.



As I initially intended to try to have the armour ready in time for Good Friday, when I started wiring the scales together I set myself a quota of eighty eight scales or more per day. This was based on the fact that the container I was using to carry scales and staples so I could work on it when I was out and about could only accommodate a row of no more than eleven scales. As intended, the number of finished rows grew daily. One day, when out, I accidentally wired ten rows of eleven scales the wrong way round. Rather than undo this work, I decided that as I intended to make a separate shoulder doubling I could simply use these scales on that, as it would not matter if these went the other way, as long as all of the other scales on the doubling did the same.



At this point I should say that as I was making the armour I paid considerable attention to information Peronis had given me on how he had made his first squamata.
In looking for a suitable fabric for the backing, I went looking for a heavy weight linen canvas. I did find a suitable linen canvas but then found and even heavier weight cotton canvas which was apparently the same grade as old fashioned sail cloth. I debated this for a little while before deciding to opt for the sail cloth, on the dual bases that it would bear more weight and would not be seen anyway.
Peronis had recommended using a T-shirt to make the initial pattern for the fabric backing and tailoring the chest slightly by cutting a small 'V' shape out of the upper chest starting at the neck and then sewing the edges together. Aware that Peronis and I are not exactly the same shape as each other I decided to follow his advice on the T-shirt but hold off from any tailoring until I had tried it on. I cut two sections of canvas of the same dimensions as the T-shirt, pinned then roughly together and then enlisted the help of one of my neighbours to do the fitting of the garment. I wore tunic and undertunic, as well as a subarmalis while we were doing this in order to get the fit right for wearing it in the field. This session allowed me to get the shape of the shoulder right, as well as the cut outs for the arms. I found that it was unnecessary for me to tailor the chest as Peronis had done. The body seemed right at the time but I later had to reduce it. More on that later. I sewed it up on one side but left the other side open to allow it to be put on. I also cut a small slit at the back of the neck to allow me to get my head through. Peronis had recommended double stitching all weight bearing seams and taking my lead from this I went further and triple stitched them.
Sorry there is no picture of the backing garment on its own.

When it came to sewing scales onto the backing garment I was faced with an issue imposed by the hole pattern on the scales I was using. As the scales featured, not a pair of holes for stitching directly to the backing as many scales do, but a single larger hole, I decided to follow the method used on the fragment of scale discovered at Carpow. Although the Carpow fragment features linen twisted into cord, when I tried this I found it made the next row of scales rise up too much (the original makers of the Carpow scale knew what they were doing, whereas I clearly did not when it came to twisting linen in this way). Accordingly, I cut narrow strips of linen instead, which I folded double for strength and laid along the rows of scales to cover the holes. I then sewed up through the holes, over the linen strip and back down again to secure the scales to the backing. For the sewing I used artificial sinew thread, again recommended by Peronis. If anyone reading this attempts the same method, be aware that linen frays and so if you want it to look tidy you will need to trim fraying threads.
Here is the squamata after I had sewn a few rows on. You may notice that it took me a little while to get the overlapping of the rows right, as this depended both on how tightly the scales were wired together and how tightly the linen strip was stretched and sewn down. Sorry you cannot see the linen strip in this photo. When I took it I had just laid another row on the squamata in readiness for sewing it down.



More shortly.

Crispvs

[MDF]

Wow,
very impressive and patient work hock:
I have always wanted to make one but I always put off ...
As you have build the scales? How weight is it?

[markusaurelius]

Thanks for starting the write up. I am going to make my own shortly as well, and this will be most helpful. Are you finding that as you make the shirt that it shrinks? I have heard that as you apply the scale, it can cause bunching underneath that will make the shirt tighter, making it smaller than when originally fitted. Not sure if I will need to fit it a bit larger to accommodate this.....

[Crispvs]

Stitching on scales continued steadily but I found it took a good deal longer to sew a row on than it took to make in the first place. I found that it took me just on two hours to make enough staples to wire together the daily quota and make four rows of twenty two scales (or eight rows of eleven), whereas it took me around two and a half hours to sew a full row of scales onto the backing garment. This was partly because of the difficulty of pushing the needly fully through the sail cloth, and I often had to resort to pulling it though using pliers.
I found that going fully around the body normally required three and a half of my rows of twenty two scales, meaning my daily quota of scales could account for a little over two actual rows on the squamata. This continued until I reached the arm holes, at which point the rows became much shorted and the time taken to sew on each row was reduced accordingly. As I intended to add shoulder doubling I decided not to cover the upper area of the back with scales, as this area would be covered anyway, as well as saving on weight in the finished armour.

As the number of rows attached to the backing garment increased, naturally the weight increased as well. Although I had triple stitched the shoulder seams, the material either side of the seams started to pull apart. To combat this, I cut four patches from the same material and sewed these over both shoulders, both inside and out. I folded the edges over as normal to avoid fraying but took advantage of this by making sure that both the inner and outer patches on each side matched up with each other, ensuring that as I sewed them on, the needle passed through five layers of sail cloth, producing a very strong pair of reinforcements for the shoulders. Unfortunately I did this late one night and in my tiredness I did not sew them on identically, but this has not diminished their efficacy.



Having reached the arm holes I decided it was about time I started work on the doubling, as how I sewed on the chest and collar scales would be influenced by how the doubling lay over them. I enlisted the help of my neighbour again to sort out the size of the doubling and using a rough mock-up we managed to sort out the right widths and lengths for it, which allowed me to vary the width of the doubling so that it was the right width where it joined the back but widened as it approached the shoulder, before narrowing again to lay over the chest. As I was measuring the material for the doubling itself, I was very careful to make sure it was absolutely symmetrical and according with the measurements we had taken earlier. As I sewed scales on though, I found that small variations in the tension of the rows pulled the shape out of symmetry somewhat, although in the end not enough to make a serious difference.
I had wrestled for some time with the problem of how to join the scales on the front and back of the doubling over the shoulder. In the end I decided to lay the rows straight on the chest flaps but introduce a curve which would allow the scales to curve around the neck and over the shoulder in a manner similar to the method which would be used for a round collared squamata. Therefore I placed short sections of scale at both front and back to create expansions to allow the correct curvatures to be achieved (the lower edge of the back, being straight, also required expansions to induce the correct curvature). I found too that in order to get the scale coverage I had to clip some of the scales at the edges to shape. In this photo you can also see some of the linen strips.



Around this time it became necessary to try on the partially completed squamata in order to make sure the fitting was still correct. I did this but found that I was unable to do my military belt up around it, as its flexibility was not great enough to allow it to squash in sufficiently. I found I needed to reduce the waist by four inches (I have yet to develop a beer gut or succumb to middle age spread). This meant removing material (and scales) not only from the open ends but also from the closed side. This was a pain in the neck as it meant removing scales not only at the waist but also above and below this point in order for the armour to sit correctly without looking odd. I measured an inch in from each edge and an inch out each side of the side seam and then drew diagonals out from these points to the seam again, the upper diagonal extending almost to the arm hole. I removed all of the scales which fell within the three areas to be removed and sewed a new side seam along the new lines, as well as cutting along the new lines at the open side to remove the now surplus material there. I then joined up the scales at the closed side. In most places I needed to add another scale to do this, although in some cases the gaps was too narrow, meaning I had to cut some scales to be narrower and punch new holes (I always punch holes in brass rather than drilling them, as it is quicker and more accurate than drilling, as well as being the method the Romans would have used [once the hole has been punched, normally with a normal centre punch, I file the torn edges of the hole and flatten the metal again with pliers]), before joining up the rows again.
Having done all this extra work (around nine hours across two evenings) I needed to perform the acid test and tried in on again with the belt. This photo was taken in the mirror for reference once I had found that the fit was now correct and also shows the partially finished doubling in about the right position.



Next installment coming up shortly.

Crispvs

[Crispvs]

Marco,

I haven't actually weighed the squamata yet (come to think of it, in fourteen years of wearing segmentata I've never got around to weighing that either), but when I pick it up with one hand the weight feels about the same as that of a modern British solder's body armour when picked up the same way, if that is of any help to you.

Markus

I did anticipate bunching which is partly why I initially made the backing garment as wide as I did. However, the contraction in the material was far less than I had expected, hence the need to cut pieces out. This is largely influenced by how much the rows are stretched as they are sewn down. I found it was useful to have a degree of play in the rows, which seemed to largely overcome the lateral contraction. There was some vertical contraction, but most of this disappeared once there was sufficient weight in the armour to drag it down again. Of course, I knew that I could compensate for any contraction to a large extent anyway due to being able to determine the width of the leather edging on each edge of the open side to ensure a correct fit, which would be one of the last tasks to do on the armour, of which more later.
I did find that having to induce quite a sharp curve around the neck and the inner edge of the shoulder doubling as well as the contractions in row numbers (of which more later) meant that there was some contraction both in the neck hole and the inner edge of the doubling, making them both quite close fitting, but I did not find the same problem where rows were laid straight.

Crispvs

[Vindex]

Wow!

Thanks Crispus. Really pleased I asked you now - shows how much work you've put into this.

Even more impressive than the original photo!

[munazio planco]

avete omnes

I am looking at how is able to do this kind of work so patient and precisious whit very big respect ...after ten scales I could finish in the psichiatrist hospital whit a patera in the had and a flower in the hand... :grin: :grin:

[markusaurelius]

Thanks for the details Crispvs. I may make mine open on both sides of the body, like a full muscula would. It would make it far easier to add or remove scales, then having to remove a whole row and try and take one out of the middle.

[Crispvs]

Thanks everybody.

Right, next installment.

I decided that as the doubling was not just a separate piece but a part which would influence how I ended up finishing much of the body, I decided I should finish this before doing anything else on the body.
I had already punched holes in many of the scales in order to attach the leather edging and as I completed the attachment of the scales, I completed the circuit of holes around the scales along the edge. I then measured around the entire edge and cut a one and a quarter inch wide strip from a piece of leather of a suitable weight which was a couple of inches longer than the measurement I had taken. I then sewed this strip onto the edge of the doubling, folding it around the edge and pushing holes through it with my awl which also passed through the holes I had already punched through the scales, normally making eight at a time, before sewing tightly though these holes with the sinew thread (chosen as it was less likely to be frayed by the brass edges). Once I had sewed all the way around I continued around for a second time to ensure that the edging was very secure.
Taking a lead from the method of attachment used on the scale shirts found in France a few years ago, I also sewed four short leather straps to the inside of the lower edge of the doubling to attach it to the back of the body and punched a single hole in each (actually I did this before I attached any of the scales, as I would not have been able to once the scales were on, but it seemed to make more sense to mention it now.



I would have made this post longer and covered more but as my computer has been freezing so often that it has taken over an hour to do this much I will finish this either later this evening or tomorrow. Sorry about that.

Crispvs

[Titus Marius Secundus]

Did you mention how you made the scales? I read most, but honestly not all :-(

Sam

[markusaurelius]

And I was wondering what the dimensions of the scales are, and how many you needed to make the whole armor.

[Crispvs]

Right, here we are again.

To attach the four straps on the doubling to the body, I made four buckles. I cut four pieces of 3mm brass rod (sorry but for some reason I did this bit in French measure - I'm not sure why - perhaps I was tired or had imbibed rather too much), each four centimetres long. Holding each piece in turn with pliers by one end I heated the other end with a propane torch and then hammered it flat against the hogging last I use as an anvil. Having done this I punched a hole through the flattened end (this was much harder than it had been with the scales, as the hammering had work hardened the brass). Having made the hole with a punch, I then widened it to the required diameter with a needle file. I did this for each end of all four pieces of brass rod, leaving each somewhat lengthened with an eye at each end. I then heated each in turn with the torch and bent the ends round until the eyes faced each other. Owing to my lack of skill in metalwork, each ended up being a slightly different shape and width. Next, I cut four pieces from a slightly narrower piece of brass rod, each corresponding in length to a particular curved piece, although each was deliberately slightly longer than the width of its partner element. I heated the ends of the four new rods to anneal them, before inserting them into the eyes and peening the ends over with the hammer. Pins were easily made from further pieces of the narrower rod, one end of each being annealed and then bent around the new bars with pliers and then worked back and forth a few times to ensure free movement. The buckles were completed with plates cur from brass sheet. I punched a row of three holes in each of these and filed these into slots. I then folded these over the bars, the pins passing through the slots, before flattening them and punching two pairs of holes through both layers of each. Sorry there are no photographs of any of this.

I attached the these buckles to the back of the body in positions corresponding with the positions of the four straps, simply by sewing them on and then attached three short rows of scales between the buckles. So as not to impede the movement of the buckle pins I decided not to attach these short rows with linen strips and sewed them on in the same ways as the buckles, so as to hide any exposed backing material from view.

Having seen to all this I attached the remaining rows of scales to the chest, extending up to the collar, which I surrounded with a single row of scales ending either side of the split at the back of the neck. I also sewed on two short rows of scales either side of the split, which were overlapped by the row around the neck. Most of rest of the area to be covered by the doubling I also left free of scales.
Once the scales had been attached around the collar, I stitched on edging in exactly the same way as I had done with the edging on the doubling, edging the split at the back as well. In the process of doing this I also inserted two short pieces of leather thonging, which passed out of the edging either side of the top of the split, through holes punched into the edging for that purpose. These were to do up the split once the armour was on and were prevented from being pulled out by knotting the ends prior to insertion in the edging.

Once I had done all this it seemed high time to see to the side closure. Again I donned my tunics and subarmalis and enlisted my neighbour's help in measuring the gap between the edges of the open side. This allowed me to both cover the edges of the scales and the knotted ends of the linen strips, as well as conveniently making up for any contractions in the width of the armour. I cut two pieces of leather about twice as wide as I had done for the other pieces of edging. Following a suggestion from Peronis, prior to cutting these pieces of leather I drew a nominal line down the centre of the back of each and left an extension of an extra inch and a half on one side of this line. The shorter side of the line was equal in length to the length of the vertical edge of the open side. I then proceeded to sew the edging onto both sides. In most cases I was able to sew through holes already present in the scales at the edges but in some places I did have to punch further holes. In the process of doing this I inserted a longish piece of thonging in the top edge of one side and a somewhat shorter piece in the bottom edge of the same side. As I reached the bottom of each side, the extensions I mentioned above were folded up and sewn in to produce a neat finish to the bottom. Once I had finished sewing the edging on, I sewed around its edges with linen thread to make the edging stiffer. To finish the closure I punched seven evenly spaced holes through the edging piece with the pieces of thonging at top and bottom and nine holes through the other piece of edging. Initially I had punched fewer holes than this but found the closure was better with nine holes. The long lace from the top is threaded through the holes and is then simply tied to the shorter piece of thonging at the bottom.

Once I had the closure sorted out I was able to find the true centre of the chest. Having done this and with close reference to how the double lay against the chest when attached, as well as to the positions of underlying linen strips, I marked the position for the attachment of the breast hooks and punched and filed through this. Once I had filed this hole (which went through two scales and the backing material) to the right diameter I attached the breast hooks, pushing the already annealed stem of the central button through the hole and peening it down over a washer on the inside.



Once the breast hooks had been attached, I performed the same operation to attach the buttons on the doubling for the hooks to grip.

Next, I extended two rows of scales over the edges of the shoulders, running them down the back edges of the arm holes, and curving around to meet the rows of scales between the buckles. I then punched holes through most of these scales to facilitate stitching and then, for the final act, sewed the edging onto the arm holes in exactly the same way as I had done for the shoulder doubling.



Here is a photo of the back of the body, showing the buckles and other features described above.



Finally, here again is Caballo's picture of the finished squamata being worn by yours truly.



Sorry about the lack of good pictures for some parts of the process. Unlike when I made my sword scabbard, when I made the squamata I did not really think of doing a 'how-to' and did not think to photograph every stage of the process. Apart from the last three photos (two of which were taken today), the photos I did take were really only taken for my own comparison purposes and so that I could track my progress.
I hope that it has been of some interest though and that it has not ended up being too long winded. :roll:

Crispvs

[Crispvs]

Markus,

The scales are each three quarters of an inch wide and one and a quarter inches long at their longest extent. I honestly don't know how many scales I used. I lost count around the time I started sewing them onto the backing and have not taken the time to count them since, but I think it was probably between four and a half and five thousand.
Regarding your suggestion to leave your's open at both sides, you could do that, but I think it would end up taking nearly as much time to do and would certainly take a lot longer to do up when it came to putting it on. The two to three minutes that it seems to take to lace mine up would be doubled.
Incidentally, I didn't need to remove any rows of scales to modify the girth measurement. I simply opened up the staples and removed the scales from the affected areas only, joining them up again once the adjustment had been made, as described above.

Sam,

I didn't make the scales myself. I was lucky enough to have been given a box full of them some years ago by a friend who had found them surplus to his requirements.
It is possible to buy scales from a number of places. Deepeeka used to sell individual scales but I am not sure if they still do. If you have a look at Matthew Amt's Legio XX site he has a page listing suppliers which may be worth a look as well as a page which shows a way you can make scales yourself. I recall there being a very good thread here on RAT a few years ago about good ways of making large numbers of scales. If you search back about five years you may be able to find it, although many of the pictures may have disappeared by now.

Crispvs

[MDF]

It 'an awesome job ... some years back I made a chain mail, and I promised myself not to repeat such work more with hundreds of pieces!

Many compliments!

[Theodosius the Great]

Crispvs,

Excellent!

You did a neater job than I did.

Thanks for posting your work. I like how you tapered the waist! I'll try to make that modification to my squamata.

I hope you don't keep the shirt lying flat on the ground. Whenever I do that the staples on the side plates become warped due to the weight and eventually come loose. I hang my shirt on coat hanger using several wooden hangers.

Have you tried cleaning your shirt yet? I've been trying several methods but no matter which one I use the process is very time consuming. So far I like using a dremel with a steel brush.

I lost count around the time I started sewing them onto the backing and have not taken the time to count them since, but I think it was probably between four and a half and five thousand.

I used about 1,400 plates but I didn't make a doubler and my plates are slightly larger.

The most tedious part, I agree, is the stitching. At least when you're making the rows or the staples you can watch a movie at the same time.

~Theo