Crafting: Hide-faced Viking Shield Experiment



I’d been looking for a long time to find a source of very thin hide to try and face-and-back a shield. I had been told by my instructor at Asfolk Viking Martial Arts school that the evidence for a hide-faced shield was hit and miss; some were most likely raw wood, some were rimmed with hide and stitched, some may have been faced, etc. As with most things Viking, the relative paucity of physical artifacts means that every new find brings new and exciting information.

Nonetheless, if you’ve been following this blog at all, you’ve seen my learning the craft of making shields bit by bit, and that I also offer them for sale. One thing that always eluded me – mostly due to a lack of a good source for the hide – was the “parchment-thick” hide that my instructor says would have been used. I use goat hide claimed at 1 oz thickness (about 0.5mm thick) for the edges, but those hides are not large enough to cover a full-sized shield.

My existing “red” shield has been in use for quite a while now, and the edging, though one of my earlier trials, has held up well. I also made a pair of “three fox” shields, one as light as I could make it (less than 5 lbs!) of aspen, with a very light stainless steel boss (5 oz) so that the jarl of the Viking Encampment at the Minnesota Renaissance Festival could march with it, and a much more robust one of poplar, edged in deer hide, with a too-thick robust boss. Even so, that one came in at 7 lbs, which was several pounds lighter than her prior shield, which was too heavy for both performance and parade use.

Nonetheless, I wanted to try one of my own.

Hide in Hiding

In any case, I really wanted to try a hide facing. Not least because of how Thegn Thrand had demonstrated how a heavy cowhide facing, laminated with milk glue (casein glue) onto a 3-5mm pine core, could be ridiculously strong and tough.

I wanted to try and make one of similar dimensions for the core, but using the thinner hide attested by my instructor. Problem was, I couldn’t find the hide, and didn’t want to get a large deer, cow, or horse hide and stretch it.

Well, I finally found a source. Two, in fact, though I only found the second one after ordering a pair of hides of the first, which was 50% more expensive than it needed to be. I put a micrometer on the hides, and they were 0.4-0.5mm thick; reindeer hide available on drum sites can be 1.6mm thick, so its quite thin . . . and it was considered a “thick” goat hide. Others were available that should wind up being 0.2-0.25mm thick – about two human hairs or so.

In any case, I had no idea how this would go. I use Old Brown hide glue rather than casein glue – not because I think Thrand is mistaken, but rather simply because I have it on hand.

Carving the Blank

At least at this stage, I’m still using power tools to do things. More on that later. But I’ve got a circle cutter, power planer, and other tools to help me out.

I butted the planks together – they started as Menards 7/16″ Mastercraft poplar boards – to form the square blank and bound them under clamps with the hide glue. Titebond liquid hide glue, in this case; I have since switched to Old Brown. Crystalline hide glue, melted in a pot, is another way to go and likely what was used for hide glue historically. It has a remarkably fast setting time for tack, and I don’t have a workshop that will let me do what I need to do fast enough to make use of it. Alas.

So, I make the square, cut the circle, then shave down the glue drops with a curved scorp. Then sand off any steps in the butted boards, then plane it, sand it, taper it, and sand it some more. In the end, I went pretty thin with this one. 6-8mm at the core, and 3mm or less at the edge. The taper starts about 6-8″ from the edge of the shield.

Some pictures:

The sanding and tapering and sanding is a place where the power tools can cost you. The planer is . . . not gentle, and a belt sander not much less so. One of my seams split, but not badly. Fortunately, this is also where hide glue shines, as it’s pretty darn reworkable. I fixed the joint and then cleaned it up with water.

So that was that. The blank came in at 3.25 lbs, but I expected it to drop down to 3.0 to 3.15 once I cut the hole for the center. I’ve made the mistake of doing that too soon before, so I held off until I could secure and measure a proper boss.

Were I to be duplicating my red shield, I’d cut the boss hole now, then soak and glue on goat-hide edging, then stitch that in place. Then paint it, mount the handle, then the boss.

But that wasn’t the destiny of this one.

First Lamination

I determined I’d lay up the hide one side at a time, not least because I had no idea how the first one would go, and I didn’t want $130 worth of hide ruined in one go.

The hide itself was nifty. Nearly transparent, and it came rolled up. That gave me an idea.

I took a piece of 3″ inner diameter PVC, press-fit a cap on one end, and filled it with hot water. The rolled-up goathide went in that, and 10-20 minutes later it was plenty soft enough to work. I spread it out and painted glue on one side, then pressed the shield blank onto it.

Turning it over, I used a foam roller paintbrush to smooth out any bubbles, and that worked great. I then put it under some weights in my workout room downstairs to dry for a bit.

The water from the soaking dilutes and helps spread the glue a bit, which was good. It also soaks into the surface of the wood . . . with some surprising results.

I was expecting it to bend. I figured the hide would contract as it dried, doming the shield. Really, since the wood is so much stiffer along the grain, I expected it to “pringle.” I expected it to bend towards the hide.

Which it did, and did not, respectively.

It bent. Wow did it bend. But it bent away from the hide.

I tried to press it flat . . . and it cracked, much to my dismay. I had to think really hard about whether to soldier on and risk a lot more money on what really looked like a bust. I figured that a second lamination, if I could get it to work at all, would pull the pringle flat-ish.

The bend took a bit to figure out, but I am hypothesizing that there were significant stresses in the face of the wood, and hide or no, when I took wet hide and put it on the face of the shield, it relieved the stresses on that side, allowing the boards to bend the other way. Either that or the weight itself, pressed into the dry face of the shield, forced it to relax. Not sure, but not what I was expecting.

Nonetheless, I kept on with it, and the next day, put glue on the board this time (mostly to protect my floor), soaked the hide, and put it on, and under pressure. A bit more cracking, but not too bad, and I resolved to leave it under weight for at least 24 hours, and not fiddle with it.

The results were surprisingly pleasing!

Most of the bend had warped out, and when I again pressed it under nearly 200 lbs of weight overnight, it was flat enough (though it did a little S-curve at one end) that I was able to use the circle-cutter to open up the hole for the handle. I’d gotten a 9-oz boss from my instructor, who make them himself, made of mild steel to historical dimensions, taken from actual museum pieces measured with calipers. In Scandinavia and Germany. So pretty sweet.

Edge and (oops) No-Stitch

I have a rig and process that I use for the edging. I soak the hide strips in water, then coat one side with hide glue, which dilutes and spreads quite evenly. I use a swarm of binder clips and popsicle sticks and a jig to keep the edging on and keep it as flat as I can. My later shields have gotten quite nice and smooth. This one was complicated by the hide facing being a bit uneven, and I had to re-laminate it at the same time as I glued the edging. In the end, it worked out . . .


The edging . . . huh. I got so excited after I finished edging it that I forgot to stitch it with linen thread. Ah, well, I can do that later, I guess.

The Finished Product

In the end, I grabbed a standard oak 1×2, and shaped it with my shiny new bandsaw, which worked much better than my jigsaw. I then used a hand-rasp, again recently acquired, to shape and round the handle. Finally, I sanded it smooth and rounded the back of the “D” at the handle. I made the handle a bit narrower than my red shield, perhaps 1.0-1.125″ wide at the waist, gently increasing to the full 1.5″ width of the 1×2, then tapering back down to 1″ wide at the edge.

I think it’s the nicest handle I’ve used.

That left me ready to mount the boss and handle. I was pretty excited about the new boss. It had a very wide flange, and I was able to drill the holes exactly the size of my copper nail shafts, which made for the smoothest rivet-and-peen process yet.I had to work a bit to get the shield facing to be as flat as I could, but I’m still pleased with the results. My “red” working shield that I train with is on the right . . . the one with edging and stitching but no facing. And a gigantic boss that’s a legacy of prior mistakes.

Ultimately these two shields came out to be exactly  the same weight. They’re both 6.0 lbs and you actually need to get to two decimal places to see a difference. That’s 96 ounces.

Where’s the weight?

  • Core: 50 ounces
  • Edging: 5.5 ounces
  • Boss: 9 ounces
  • Handle: 13 ounces
  • Facing and glue and hardware: 18.5 ounces (1.16 lbs)

The two full hides themselves were about 1.2 lbs total, and about an inch or so got trimmed, plus the hide that was laminated to the core. That means that each facing was, including glue, about 9 ounces . . . a number that I’d expect to get cut in half if I use the “thin” hide I found on sale later.

I took the shield with me to a birthday party at the Asfolk Longhouse and used it to teach and train with. It’s much more . . . sprightly? Nimble? Responsive? than my existing red training shield. It moves in the hand a lot better than my other one. Some of that is the smaller opening for the boss, which helps the one-handed manipulation of the shield, as well as the superior taper of the handle.

But it also feels to me that the stiffer laminate is just more responsive.

And if Thrand’s videos are any indication, the new shield “should” be tough as hell. As the saying goes, our ancestors weren’t stupid. If they did all this work, they did it for a reason.

A Successful Experiment

So, that was that, and it worked out well.

The flaws in the experiment was that the goat-hide was too thick, and too expensive.

The next experiment, I’ll build a jig and laminate both sides at once. I’ve already got it designed in my head, and I feel really good about that. I’ll also order the thinner hide from another source. That should save me at least $50 in raw materials.

Also, I have acquired a large quantity of quarter- and rift-sawn poplar from some 40-50 year old trees my father in law had to cut down, so my next shield will use some great wood, with vertical grain. It’s drying right now, and soon I’ll resaw it to 1/3 of an inch thick, and take care as I make it into a blank to mind the flatness. That should give me a lot less material to remove, and I may try my hand at using a spoke-shave or hand-plane to smooth and taper it, rather than my power tools. That should make the next shield come in close to 5.25 to 5.5 lbs, and be even more robust thanks to the vertical-grain, rift-sawn wood.

I will definitely offer these shields on a commission basis. I think, though, that relative to a wood shield alone, the facing will probably add quite a chunk to the cost of the shield, but I won’t quote a price just yet, because I need to work out and reprice the entire line I offer. Regular wood, stitched, edged-and-stitched, or faced.

3 thoughts on “Crafting: Hide-faced Viking Shield Experiment

    1. Effectively, that’s what I eventually wound up doing. I used a lower and upper board of 1/4″ melamine, because it’s uniform and flexible. I then line it with waxed paper. I laminate the hide both sides at once, covering it with another waxed paper/melamine board, then pile on bags of water softener salt. That allows the whole shebang to be mildly compliant to the tapered edges of the shield. Works very well!

Leave a Reply

Your email address will not be published. Required fields are marked *