Sewing Eyelets on Nylon and Polyester (PET) 6mm Cords for Break Tests

Pics by Harlan W S
Aug 11, 2023

First, heed this warning: If you try to sew rope you will eventually DIE! Leave rope-sewing to elite outdoor-gear purveyors, who have good lawyers!

This note is really the background for an on-going study on the effect of knots on force dissipation in ropes. In those tests, the cord is pulled to breaking, which invariably occurs in one of the knots. It is obvious that to find the breaking strength and force dissipation behavior of the in-line knots, the end attachments of the cord (from which it is pulled) must be stronger than any inline knots.

So in the tests REPORTED BELOW, we verify that we have a way to pull end-to-end on unknotted rope, and recover the reported rope strength

So how do you make attachments at the ends of a rope section, so that the rope stength is preserved? Most knots will weaken the rope by 35-50%. There are a few knots (such as the modified trilene) that are almost full rope strength, for nylon and polyester (but not Dyneema); they take up a lot of rope and must be cut free after the test. There are capstain clamps and big bollards that involve wrapping the rope (many times) around a cylinder that is at least 20x the diameter of the rope, and they get near rope full strength -- but use up a lot of rope.

Or, if you really really really know what you are doing, you can sew eyelets at the ends. (WARNING WARNING ROPE-SEWING IS THE REALM OF THE DEVIL! I take no responsibility if you choose to sew rope!)

I've covered how I sew rope elsewhere. My "equipment" is shown in the last frame of this page. Aside from common sense about sewing, you want:
(number of stitches through both strands)*(strength of thread) > 2*(reported strength of rope). For these tests, each termination had at least 42 stitches.

For this hand-sewing, I use 100 lb test Spectra (UHMWPE) braided fishing line. One interesting property of braided Spectra: it is easy to form a cut end into a flat blade with the flats of pliers, and then very easy to push the flattened end through the eye of "reasonably-sized" needle. Normally I would suggest polyester or nylon thread to sew polyester of nylon; but these tests below seem to show the Spectra is fine, and it is damnably hard to find poly or nylon thread that is more than 25 lbs breaking strength. The worry is that the Spectra will cut right through the weaker nylon or poly yarns; and that probably happens, but only near the breaking strength of the rope, so who cares?

10-06-2023 NOTE: "Capstan Clamp" test
I made a capstan clamp from two 1.5" (38mm) OD U-shaped iron pipes, and wrapped bluewater cord around each pipe 4x (and tied off with half-hitches), then pulled to breaking at 1922 lbs. In principle, this system should approach the native strength of the 6mm nylon rope. This test gives me confidence the sewing works well, as the cord with sewn loops actually broke at 1.011x the reported MBS, and the capstan clamped-rope broke at
0.983x the MBS. I have no good way to judge the likely uncertainty, so 1.011, 1.000, and 0.983 may be effectively the same number.

Click on any image to see larger version

a20230806_093312
My linescale
loadcell sits in
a recess in a
massive block of
EVA foam,
inside a case.
b0_PET_vs_nylon_sewn
Results of two pulls
behind my jeep. Both
cords broke in the
stitching, but ABOVE
the rated strength of
the pure cord.
bluewater_nylon_sewn_break
Broken Bluewater. It
looks like it broke below
the stitching, but in fact
the core broke at the
stitching, and the sheath
broke below.
Maxim_PET_sewn_break
Broken Maxim.
nvector_sewn
One problem: near
the bottom of the
stitching, the
live strand (left)
is pulled at a slight
angle (red line) to
force line (black
line) which reduces
the strength.
sewing_kit
My gear for sewing.
The shears are
micro-serrated, and
will cut UHMWPE
and aramids like Kevlar.
The brown object is a
palm thimble.