Thread Strength Comparison Test

[spectre6000]

Premise:

Part of my attraction to leather and leatherwork is that it's a natural material. Sewing it together with plastic thread sorta defeats the purpose in my view. I did a ton of research a while back into various natural fiber thread options, and, as I'm sure many of you are at least peripherally aware, there's not a ton of really solid info out there and it's in a huge variety of non-correlated terms and units. So, I did a little leg work, made a bunch of phone calls, ran all over town, ordered a few things online, and came up with a wide variety of threads to test, then sat on it for a few months, then moved the pile to an unused room because I needed the space, then eventually forced myself to scratch it off the to-do list.

Methodology:

Pretty simple really... I tied a bucket to a pry bar without any sharp bends in the knot, suspended that on a fender stand, then slowly filled the bucket with nuts and bolts out of my spare/used hardware bucket one small handful at a time with a healthy pause in between. Once the thread broke, I weighed it on a calibrated scale (.25# granularity), and recorded the result. Repeat down the list.

Subjects:

I was primarily interested in silks, because I thought that was most likely to be the strongest natural fiber I could get my hands on. I couldn't find reference to silk being used in leatherwork beyond aesthetic embellishment, but also couldn't find any reason for that to be the case. It's a commodity fiber, and not a cheap one, so my suspicion was that it was just too expensive and variably priced to gain much traction in industry, and thereby hobby-land. Since cost is not in my top XXX consideration list, it didn't seem a deterrent. I located a silk importer (one of the very few in the country happens to be semi-local to me!), and got a bunch of samples. I also tried "beading silk"; a lower overall quality, but more uniform industrially produced product. Unfortunately, it's difficult to impossible to find in larger quantities, but I figured I could track back to the producer if it was worth it. I obtained a variety of linen threads of various numbers of cords, and both Barbour and generic. I also gathered the stand-bys, nylon and poly, in 69, 92,138, and 207. I ran across a small spool of Kevlar in my hunting, so scooped it up for giggles. 

Hypothesis:

My sincere hope was that one of the silks I could get my hands on in nominal quantities would be on par with the poly threads (the presumed non-Kevlar strongest). I really had no idea, just a hope. I expected the linen to be fairly mild. Nylon was expected to be a bit stronger than the polyester. I thought Kevlar would blow them all out of the water. Fortunately, the experiment was structured such that my pre-conceived biases had no effect... because I was pretty much wrong across the board!

Results:

Doing my best to recreate the spreadsheet here...

Thread-------------------------------------------Breaking Point (LBS)--------Notes

Kevlar.................................................................2.............................................Very fine, guessing ~38

#1 Silk................................................................4

Slik "Lixue"........................................................4

#2 Silk...............................................................6.75........................................Guessing ~92 thickness equivalent

69 Polyester......................................................8.5

69 Nylon...........................................................8.75

#3 Silk THEORETICAL.....................................9

Silk "Chilali".....................................................9...............................................~138?

Silk "Neva"......................................................10.5

92 Nylon..........................................................10.75

4-Cord Unbranded Linen................................11

92 Polyester.....................................................11.75

#4 Silk..............................................................12

3-Cord Barbour Linen THEORETICAL.............14.25

5-Cord Unbranded Linen.................................15.75

138 Polyester...................................................15.75

138 Nylon........................................................17.25

Waxed Nylon? 3-Cord (leftover Tandy?).........19.5

4-Cord Barbour Linen.....................................19.5

207 Nylon.......................................................20.25

207 Polyester.................................................21

6-Cord Unbranded Linen..............................24

5-Cord Barbour Linen...................................25

6-Cord Barbour Linen...................................32.25

Conclusions:

Like I said... It's a good thing I checked, because pretty much everything I expected to happen wasn't even close! I wasn't able to get my hands on #3 silk very easily for some reason, so I calculated a curve across the other numbered silks, and that's roughly where it should fall. I did the same calculation for the 3-Cord Barbour Linen (used to test something else a while back, and I only had about 6"), and checked my % delta against a published tensile strength chart, and was within 7%. With a sample size of one and so few data points to fit the curve to, I figure that's not too bad. I was super shocked by the Kevlar... I thought that would require a bigger bucket (and even switched to a bigger bucket in anticipation), but it didn't like it at all... I tried a second time, and had the exact same result. Both times broke at the knot, and it was the finest thread I tested, so that could have something to do with it. I had a ton of other exotic silks and silk-animal fiber blends (including camel and yak just because I happened upon them) blends, but they were all so weak that I didn't really feel it was worth boring anyone with extra lines on the spreadsheet. The unbranded linen was supposed to be at least as good as the Barbour, which has supposedly gone way downhill and is made in China or some such now, whereas the unbranded stuff is Ukrainian or something... Nope. You get what you pay for there. Sometimes the markup is for marketing, sometimes it's for quality. This is the latter case. 

Application:

If I need big strength in a small package, it looks like I'm using Nylon. If I need strength with UV resistance, it's polyester. I thought Nylon and polyester would be a little different in strength, but they're neck and neck the whole way. The silk was a total bust, so that rabbit hole will be filled in. For most things, I see people/companies using 92 or 138 threads, and the 3-Cord Barbours falls right in the middle there without being too visually obtrusive, so that looks like what I'll be using for the most part going forward! I shared this because, while most probably won't wade through all my floral prose, it might help some like minded person down the road not have to go through such a task. At least it's all now in the same place in the same units with the same methodology for relative comparison. Enjoy. 

Edited July 14, 2017 by spectre6000

[soccerdad]

This is really interesting, spectre6K. Thanks for all the work. I don't recall seeing this sort of thing all in one place for common leather threads. When I'm sewing leather bags and wallets, I think a lot about resistance to UV as well as to abrasion for items that get handled a lot, such as wallets. I've seen some info on these for nylon and polyester but not for natural fibers. If I remember correctly, nylon doesn't resist UV as well as polyester, a consideration for outdoor applications, but nylon resists abrasion better than polyester.

Thanks again!

[spectre6000]

I've also done some UV exposure testing on vegetable tanned leather, but not thread. The testing is all done (nearly 60 full days' worth), I just haven't done anything with the samples yet (they're all packed into the same spare room for forgetting purposes). I'm not sure exactly how you would do a realistic abrasion test for thread. It doesn't seem as simple as just taking sandpaper to it... There is an industrial standard, so someone has already figured it out, but I've not looked into it at all. Linen was the clear natural fiber winner for strength, which I know won't come as much of a surprise to some, but now there's actual context and some idea of relativity. This is pure conjecture, and as demonstrated above, is apparently not my strong suit with this sort of thing, but I think since linen thread is used in shoes, it's probably pretty decent on the abrasion side of things. I think I read that it has good uv, abrasion, and mildew (for another testable element) resistant properties, but that may or may not have involved waxing...

[oioi]

Great experiment. Thank you for sharing!