scutum covering again

[Peroni]

Wow! Hadn't thought of the huge headache of drilling all those holes (probably in the neighborhood of 200 for a scutum) with a hand-twist drill!

Probably with a bow attached, it takes a hell of a lot of work out of the job :wink:

[M. Demetrius]

I think I'll do a stress test on gluing wet and dry rawhide onto plywood. I'll get back in a few days with some kind of result, or no result, whatever.

I generally use the "glue moist with water glue technique" with cloth, and it does fine. I wonder. If the glue actually penetrated a little into the rawhide, then it all dried as one mass, possibly the wood would fail before the glue bond failed. If so, that's about all you can ask of glue, right?

[Marcus Mummius]

I'm very curious for the results! Thanks!

Vale,

[Matt Lukes]

I think I'll do a stress test on gluing wet and dry rawhide onto plywood. I'll get back in a few days with some kind of result, or no result, whatever.

I'm doing one right now with hide glue- dry rawhide glued to wood, wet rawhide glued to wood, tanned leather soaked in hide glue glued to wood. This morning the dry rawhide and leather are nice and dry and I'll subject them to some bending tests in a bit. The wet rawhide is still rubbery, so it'll probably be a couple of days before I know if it has adhered as well as the dry stuff...

[M. Demetrius]

Well, you're ahead of me, since I haven't cut the rawhide bits yet. Let me know. I'll follow up if it makes sense to do so, with a stress test...I have th idea to put a ring in the rawhide as it's glued onto the wood, let it dry, then hang a weight from the ring and see what happens.

I await your results first, before proceeding.

[Matt Lukes]

Okay, my basic tests are done. Bear in mind these are qualititative, not quantitative. No measurements were made and the different materials simply compared to one another.

Materials and Methods:

I used average-strength hide glue (315g), mixed 1:2 glue:water, heated to about 70C. The rawhide was unbleached and about 1mm thick (dry). I don't know what type of hide it was, except that it wasn't pig. The leather was chrome or alum-tanned pigskin lining about 1.5oz, about 0.6mm thick when saturated and dry. The wood substrate was birch (scutum leftover) 3mm thick and 3.5cm wide.

The leather was soaked in the hot glue for 10 minutes and agitated to ensure it became saturated. One section of rawhide was soaked in warm water for 1 hour and then blotted lightly. The dry rawhide and birch substrate were sanded lightly with 200 grit sandpaper to roughen them slightly. Glue was applied to both the wood and various material patches to ensure a continuous bond. A piece of plastic was then applied on top and then a second piece of birch plus some clamps to ensure all was held tight until the glue gelled. After 15 minutes, the clamps, wood and plastic were removed to allow all to dry. One corner of the wet rawhide patch was accidently lifted but the remainder was considered sufficient to permit testing. After 24 hours the wet rawhide patch was found to have been lost during some accidental movement of the wood. A previous piece applied to 1/8" plywood 24 hours previously (total drying time 48 hours), however, was intact and substituted for this qualititaive test.

The birch strip was tested by bending it along its length with the experimenter's thumbs (that is, me ) immediately behind the centerpoint of the patch of each material. The qualitative, relative force (as it appeared) necessary to crack the wood and tear or separate the material was then judged as more or less than the others and they were ranked strongest to weakest. A further test of plain wood was then conducted as a control.

Results:

Strength

1) All materials appeared to increase the force necessary to break the wood considerably. Neither the dry rawhide nor the soaked leather separated from the substrate before tearing, and both tore only as the wood broke- apparently partly at least because of the splintered wood edge. Unfortunately, the wet rawhide substitute piece might not have been entirely dry as ultimately wasn't terribly hard to tear with my bare fingers.

2) The dry rawhide was stronger than the glue-soaked leather, but not as much as might have been expected given that the former was about twice the thickness.

Glue Performance

1) hide glue will bond wet rawhide, but it's quite easy to accidently lift sections and thus spoil the bond. The glue gels when it cools down, not when it dries, but only solidifies when it dries- so the glue remians a gel as long as the rawhide is wet. Since the wet rawhide is so flexible, it's VERY easy to separate it from the glue for a day or more I should think. After 48 hours, as already stated, it wasn't 100% dry either.

Conclusions

Well, technically, I can't really say anything for sure since the wet rawhide test wasn't properly completed. I see some difficulties with working with rawhide has a number of problems. Having worked with hide glue a fair bit now, I can say that it wouldn't be easy to use it to secure a pice of dry rawhide to a scutum face. For large areas, it's usually necessary, because the glue gels fairly quickly, to work in sections- difficult to do with a semi-rigid material as you cannot fold it easily without accidently pulling up already glued areas. That would suggest that wet rawhide would be the thing to use, however it has its own problems. As already stated, it's very easy to separate the hide from the glue even after the latter has gelled. Also, rawhide contracts when it dries and I do wonder if that might be part of the reason it wasn't as strong as the dry rawhide- something I hope to find with the repeat test.

Applying glue-soaked tanned leather would be easy enough if very messy. Thoroughly saturating such a large piece of leather could be problematic- a large container and a lot of glue would be necessary, and a specialized setup to heat it. Luckily, hide glue can always be turned to gel again with water and heat, so it's not hard to clean up.

Here are pictures- the top is the dry rawhide, the bottom is glue-soaked tanned leather:

[Matt Lukes]

One further thing, my test was conducted to break the potential outer layer materials from the inside- to test the material's tearing capacity. On a real scutum the impact would come from the exterior and the wood wouldn't flex the same way or NEARLY as much. It seems to me the major danger to the covering from an impact on the scutum would be separation of the material from the wood, so I'm going to repeat these tests and try bending the wood the other way to see how/ if it separates.

[Marcus Mummius]

HI Matt,

Thanks for that test! Could you also try a piece with regular white glue instead of hide glue?

Kind regards,

[Matt Lukes]

I'm already on it Jef I'll hopefully have white glue results tomorrow

[M. Demetrius]

Give it an extra day, so you can tell if the wet rawhide has had a chance to dry all the way, plz. Just a thought, OTOH, do what you think is best, 'cause you're in charge. U da MAN!

[Matt Lukes]

Well the reason I wrote out everything I did was so that you guys would be able to see any flaws there might be or make recommendations for further tests or changes to those already done

Indeed I'll let the wet rawhide test piece dry longer this time- probably 3 days- as it's necessary to know both how strong it ends up being and if it'll even work with hide or white PVA glue.

[Neuraleanus]

Apparently that is how the Dura Europo scutum was made. I know from experience that glued linen over birch plywood makes something akin to fiberglass. I made my last batch of scuta that way and they came out extremely strong. I used ordinary Elmers Carpenter's glue.

[Matt Lukes]

That'll indeed be the last phase to this little basic research project Lee- to see how readily (or not) these outer layer materials separate from the linen layer.

I didn't include that yet because thus far the tests were only to examine how the outer materials stack up to one another, so as long as the substrate is consistent, it didn't really matter if it's just plain wood or linen-covered wood. In fact, since it adds strength to the system, the linen would actually make it more difficult since I'm only gauging how hard it is to break a strip of birch with my bare hands.

I doubt very much that if material A separates from wood more readily than material B, that it would be any different for linen, however it would be necessary to actually try it to really validate the idea.

Now have you actually used rawhide for entire scuta Lee? If so, you can certainly help with the practicality of it- moreso than my little tests and theories can do.

[Tarbicus]

Stitching around the rim, nails or rivets for fixing the umbo, and (if it's actually the case) the nailing of metal decorations must all have contributed as well to the rawhide being held in place?

[Matt Lukes]

Maybe a little- but I honestly doubt it was a lot. If it really were that helpful, or if more help were in fact necessary, I'd expect there'd be purpose mechanical fasteners at regular intervals over the face of the scutum to maximize the enhancement. Mechanical fasteners can be important for thicker materials that are glued together, because these can be forced to move relative to one another and damage the glue bond, but I wouldn't expect a thin layer of rawhide, that has some inherent flexibility to it, to have need or even benefit much from them.

It's occurred to me too that given the scutum's already very strong construction, 3 plies of strips, their grains running at right angles to one another, the outer layers with their grains running with the main curve, plus a sheathing layer of glue soaked linen, the added strength of a front outer layer may not be as vital as some might think. Indeed for a simple single-layer butted plank shield, it's essential, but for a scutum? To be honest, I can really see the 3-ply plus glue-soaked linen being quite sufficient- the outer leather or rawhide layer being just a little enhancement as opposed to a necessity.