DISCLAIMER: I am not an engineer. I am not trained in materials testing. I am not Chris Harmston. Many people consider me a dundering idiot. No one takes advice from me. This is clearly not statistically sound testing. If you get hurt or die because you read some gibberish on the web, it's your own damn fault.


April 5-8, 2006. Rain and rest days in Vegas. What better to do than surf rockclimbing.com and pick fights? Anyways, a thread came up that wondered whether you could backup rap anchors by cutting sewn spectra (or other tech webbing) and tying it. (No doubt you've heard that they are slippery and do not hold knots like nylon. If you haven't heard, you've got some reading to do..)

Anyways, I thought that a tech sling ('tech' defined here as 'made out of something other than nylon') would probably hold fine with a double fisherman knot, and would almost definitely hold fine with a triple. And because I use tech slings and would like to be able to back up rap anchors (and use the rap anchors currently tied with tied tech webbing) without fearing imminent death, I decided to pull on some knots.

Apologies to those who like good looking webpages, correct spelling and punctation, etc. For that, try elsewhere on this site. There are a couple images added to this page, but I have before/after photos of all test samples in the directory in which this file is in. Feel free to check them out if you're really, really bored, or if you can learn things about the testing by looking at them. I also took video of each test, but they all look the same and you can't really see anything except accessory cord stretching a lot, so I won't be posting it and eating up my bandwidth.

I do not have a load tester of any sort whatsoever, so I used some accessory cord. I figured I would put a piece of this cord inline with the tech webbing knot, and seeing what broke would give me a ballpark estimate of what the strengths were.

Anyways, the bits and pieces I used were as follows:

1. New 4mm nylon accessory cord. 'Monster' brand. Blue with purple and orange tracers. Rated to 3.4 kN.
2. New 5mm nylon accessory cord. Bluewater brand. Green with orange and black tracers. Rated to 5.7 kN.
3. Barely-used 6mm nylon accessory cord from my bits-and-pieces bin. I might have prussiked on it, once. Red with yellow and blue pattern. The average of the 6mm nylon ratings I've found is 8.7 kN. However, I will assume the lowest found rating of 7.2 kN.
4. Used Bluewater tech runner (cleaned off a bail anchor somewhere). In good shape. Not fuzzy, but obviously used. Green/white pattern. Something like 5/8" width. Rated to at least 22kN.
5. New Wild Country double-length tech runner. Something like 10mm width. White, with black stripes on side. Rated to at least 22 kN.
6. 7mm nylon in good condition, from a bail anchor. No obvious sun fading. Slight fuzzing. Purple with light blue and dark blue tracers. Manufacturer unknown. Lowest 7mm nylon rating found: 9.3 kN.
7. 7mm nylon in excellent condition, from a bail anchor. No sun fading, almost non-existent signs of use. Purple with black and grey tracers. Manufacturer unknown. Lowest 7mm nylon rating found: 9.3 kN.

For all tests, I tied the accessory cord as a single strand with overhand loops in each end. Overhand knots are generally estimated to reduce the strength of cord by up to 40% (from fish's website) so I will estimate the breaking strengths of my accessory cord 'fuses' as 60% of their estimated strength. I then tied the tech sling material around the fuse (also a single strand with loops on the ends). I then pulled this series between my car and a streetlamp anchor. With the 6mm and larger cords, I had to gun it to break something. All knots were dressed and set before testing. I put no biners inline because I didnt want biners coming at my car at high speed when cords broke.

Anyways, to start off I wanted to see how good tech slings would hold up with the overhand loop. Conventional wisdom says that a double fisherman or a triple fish would be less likely to slip than the water knot, and the overhand loop is the same construction as the water knot, so I wanted to start low and work up. The overhand loop on a single strand is also a common way that rap stations are created out of cannibalized slings to maximize the usable length.

Testing Round 1

I tied a 4mm cord onto the new WC tech sling. The tech sling had a overhand knot for one loop, and a figure 8 for the other loop. The 4mm cord broke. I then tried the same test with 5mm. The cord broke. I then tried with my 6mm cord, which also broke. There was no perceptible tail slippage in the tech sling knots, and the breakage of the 5mm and 6mm cords would suggest that the overhand in the WC tech sling was good to around 720 and 900 lbs, respectively. All failures were at the accessory cord where it entered the knot.

Just in case it was just this particular WC tech sling, I also tried the 6mm test with the same setup, but with the fatter older green bluewater tech sling. Unexcitingly, the cord broke. No tail slippage through the knots.

I had originally planned to find out what it took to make a 'slippery' tech sling overhand knot fail, and then try double and triple fish knots. However, I am convinced that they will be even stronger, and have no desire to spend any more time than necessary getting stared at by unimpressed las vegas bystanders.



I would feel perfectly comfortable rapping off new 5 or 6mm accessory cord tied with an overhand loop. Obviously, other people are too because it is a common cord to find at rap stations. (In fact, some lunatics even use them for cordalettes..). The fact that the cord was consistently breaking before the tech sling overhand knot even started to slip makes me confident about rapping on tied tech slings that are similar to the ones I tested.
    Further, the rating estimates also makes me feel comfortable. The tested knots were not slipping at loads of an estimated 900 pounds (~ 4kN), and I doubt too many folks can generate that much force while rapping. Yet further- most rap stations have two or more equalized pieces which would greatly diminish the forces on any one strand.
    It gets better: if the knot were tied in a continuous loop (as it often is in rap stations) the forces on the knot would be much less than i tested with a single strand. End conclusion: I trust double and triple fisherman knots in tech sling to rap on. I would even rap on the overhand loop if I didn't have enough slingage to use a fisherman knot.

Testing Round 2

This time, I tried bigger nylon. For the first test, I set up an accessory cord 'fuse' (as described in the previous round) of 7mm nylon (cord #6 in the table, good condition). I put this inline with the WC tech runner tied on one side with an overhand loop, and on the other with a figure 8 loop. When pulled, the failure was at the accessory cord where it entered the knot. Same as in the prior testing round. No perceptible tail slippage.

Second test was with another fuse of the same 7mm nylon cord, but with two pieces of tech webbing (putting two pieces inline allows me to test four separate knots at the same time, looking for tail slippage, etc). One piece was the 10mm WC sling (tied with overhand loops on both sides) and the other was the BW (tied with figure 8 loop on one side, and overhand on the other). When pulled to failure, the nylon cord broke at the knot. There appeared to be some tail slippage in this test, as one of the overhand loops in the WC sling was noticeably larger than it had been when it was set up. However, I can not be sure, as it appears (from the pictures I took before and after) that I may have forgotten to set this knot before the test and the perceived loop-lengthening may have been the knot setting itself under load.

For all subsequent tests, I marked the tails with a permanent marker to be able to quantify tail slippage more accurately.

Third test was with the excellent condition 7mm nylon (cord #7 in table). I set up the fuse inline as for the last test, with two pieces of tech webbing. One piece was the 10mm WC sling (tied with overhand loops on both sides) and the other was the BW sling (tied with figure 8 loop on one side, and overhand on the other). When pulled to failure the bluewater spectra broke, where the webbing entered the figure-8 loop.
    This is surprising for many reasons. First: because I would have expected the webbing to by much stronger than the 7mm nylon. Second: if the webbing had failed, I expected it to fail by slipping, not by breaking at the knot. Third: if I am achieving webbing-breaking forces, I would expect the overhand knot to fail, not the figure-8. It is possible, because this is used webbing, that there was something in this webbing's history that weakened it significantly. The fact that this setup, identical in every way to the last test, broke at the webbing instead of the cord suggests that the newer condition cord is significantly stronger. There was no perceptible tail slippage.

Fourth test was with the same newer 7mm nylon. I set it inline with two separate pieces of WC tech sling, both tied with overhand loops at both ends. When pulled to failure, one of the WC tech webbing slings broke at the knot. There was no perceptible tail slippage.

Notes on Round 2:

7mm nylon has plenty of margin for me. The lowest rating I have seen for this cord is 9.3 kN (and the average rating is around 11.1 kN). When the strength is reduced 40%, this lends a 7mm fuse breaking strength of at least 5.6 kN (~1250 lbs breaking strength, knotted). I doubt anyone would hesitate to rap off a single strand of 7mm nylon tied with overhand knots. In the first two tests, the cord broke before the webbing knot failed (although there is a slim possibility that there was tail slippage in the second test). This is very reassuring.

In the third and fourth tests, the webbing knots were failing before the cord broke with no tail slippage. This was very surprising. Unless there is information that suggests that overhand knots in tech webbing runners reduces their strength by significantly more than the estimated 40%, then this is very very reassuring as to the use of spectra for rap slings. It suggests that for a single load test, the webbing will break before it slips, which is not the way I had envisioned it happening.

However, there is one possiblity that deserves mentioning here. To achieve the force necessary to break 7mm nylon and webbing, I would have to bounce the car (ie: accelerate in reverse, let off and the car springs back, and then reverse again even harder). This was nothing like a slow-pull test like most load-testing is, where they might pull a few inches a minute. Because the nylon cord stretches a lot before breaking, it may be able to handle these sudden loads better than tech webbing, which reportedly does not stretch. Perhaps if the loads were applied slowly, the cord would break before the webbing. However, in the context of rap slings, I still believe the overhand knots in spectra seem good enough. One might even suggest that the 'bouncing' of the car more closely mimics real-world situations in climbing than a slow-pull test. Unfortunately, I do not have the technical background to make claims here.

Another possibility that has been brought to my attention is the question of whether the slipperiness of tech webbings is not something that causes knots to fail by sliding through themselves (which does happen in tech cords with inappropriate knots), but rather a problem with accelerated knot creep over repeated non-failure-inducing loads (like use at an established rap station). Whether spectra creeps faster than nylon is something I do not have the ability to test. However, as long as you check the tails have not crept into (or near) the knot, this is not an immediate issue. Unfortunately, it is unrealistic to assume that all climbers will check all knots they rappel on.

Yet another problem with this amateur testing effort is the fact that the knots tested were overhand and figure-8 loops, not proper water knots and figure 8 follow-through knots. it is possible that the different loads on knots in a continuous loop (instead at the end of a single stand) act differently, creep quicker, etc. However, the testing I did is still relevant, as a lot (possibly most) of the tied rap anchors are built using the overhand on a bight (or whatever you wanna´call it) tested here, thereby getting more sling length out of a single cut tech sling.

End conclusion: for an isolated pull on an overhand or figure 8 loop, tail slippage just does not appear to be an issue with the 'slippery' slings tested. In the single instance where the knot may have slipped (it was probably my error in not setting the knot) I still broke 7mm nylon, and the knot can not have slipped more than ~1.5 cm on either strand. Assuming there is adequate tail on the knot, spectra webbing knots seem to be quite trustworthy for single-use rappel anchors. However, there are so many variables and unknowns that tied tech sling can not be reccommended at this point. More info or testing on creep is needed to be able to make any sweeping safety statements, particularly when considering outfitting established rap stations. The one thing this testing did teach me is that the failure mode for tied tech webbing is not necessarily the unraveling of a frictionless knot, which was news to me.