Cordelette Advise

Someone please kill this thread before it breeds

Jake woowrote:

This a good point and a general reminder that knots reduce strength and we should at least be aware of this when we tie them. Although I thought most knots reduce strength 30-40% opposed to 60+%. With this in mind, this is another great reason to use tech cord rated near 19kn or sewn slings rated even higher, so your safety margins are increased without losing function.

The amount amount that knots reduce strength is probably misunderstood in climbing.

From drop tests in pervious mentioned test report, in some cases knots do not decrease strength.  In the worst case it was a minor ~14% drop in strength.   These conclusions can be drawn from the infamous DMM drop test video also if I remember correctly.  

The mistake is the thinking that slow tension test represent a climbing fall.  

https://user.xmission.com/~tmoyer/testing/High_Strength_Cord.pdf 

I've used the web-o-lette a fair amount, and its lack of bulk is nice, but it can be really tough to untie mid-route when it has been cinched up. 7mm cord is bulky, but easy to untie, and I feel less bad when I have to cut some up for a bail anchor.

Seeing how other people use cordalettes reminds me of watching other people raise their kids

Preeti Pwrote:

The amount amount that knots reduce strength is probably misunderstood in climbing.

From drop tests in pervious mentioned test report https://user.xmission.com/~tmoyer/testing/High_Strength_Cord.pdf, in some cases knots do not decrease strength.  In the worst case it was a minor ~14% drop in strength.   These conclusions can be drawn from the infamous DMM drop test video also if I remember correctly.  

Well...

Most of the time people doing this sort of test don't actually test the strength of the cord without knots, because it's hard to do that. Instead, they just use the manufacturer reported MBS for the cord. Your link returns a 404, so I don't know for sure if that's what they were doing, but that would be my guess.

The problem is, that's not always a valid assumption: MBS isn't necessarily the strength of the cord without knots. Testing the strength of cord without knots is hard for manufacturers too. There are ways to do it with clamps and whatnot, but these methods are harder to make work with thinner cords (less surface area to grip). And ultimately, these methods aren't representative of how the gear is used in practice, so you have to question the validity of testing the MBS using a clamp anyway. In at least some cases, the way manufacturers test the MBS of cord is the same way it's used in the field: with end knots tied in.

What that means for backyard scientists is that unless we know how the MBS was determined, we can't really say how much a knot reduces the strength of cord. For example, if we're only seeing a 14% reduction in strength from tying in a knot, it may be because we're testing the strength of the knot that the manufacturer used to test the strength of the cord, with perhaps some minor variation. In that case, we're not comparing the strength of the cord with and without a knot, we're simply comparing the strength of the cord with two different knots.

There's a more philosophical question here, which is, what does it even mean for the cord to have strength in the absence of outside factors? A clamp also reduces the strength of cord--if you just pull cord with two clamps, it will likely break at one of the clamps rather than in the middle of the cord, because the clamp itself weakens the cord. The strength of cord in the absence of some weakening factor is a hypothetical thing rather than a phenomenon which exists in reality, because we don't have any way (at least that I know of) to test cords which doesn't weaken them.

The mistake is the thinking that slow tension test represent a climbing fall.  

Is that a mistake? As far as I can find evidence for, this hasn't really been tested. Obviously there are some differences here, but do you have any reason to believe that those differences create a significant difference in the forces at which a cord (or other gear) breaks?

Knot hard to test thin cord or even thread without  a knot or direct clamp.

Much like a knot less / tesion less anchor achieved by wrapping a rope around a tree and held loosely in place with a knot

Wrap the cord around the Biner sized fixture several times and then clamp or knot

David Kwrote:

Well...

Most of the time people doing this sort of test don't actually test the strength of the cord without knots, because it's hard to do that. Instead, they just use the manufacturer reported MBS for the cord. ...

The problem is, that's not always a valid assumption: MBS isn't necessarily the strength of the cord without knots. Testing the strength of cord without knots is hard for manufacturers too. There are ways to do it with clamps and whatnot, but these methods are harder to make work with thinner cords (less surface area to grip). And ultimately, these methods aren't representative of how the gear is used in practice, 

...

These are some good points.

If I were trying to test the MBS of a cord itself, I would probably have it wrap around a large diameter (let's say something the size of a soda can) pin on either side, wrapped several times then tied off (aka the friction hitch like gets used when tying ropes off to trees).

But yeah, since we will never use cords/ropes in the real world without knots, how useful is that information to end users? I think it would be rad to standardize a simple, real-world test to compare cord/rope MBS.  It could be: tie a followthrough figure 8 on each side of a strand, then slow pull to failure.

Ryan Jenks (How not to Highline) is in the final stages of building his drop tower, once he gets that thing rolling I think we'll learn a lot about how slow pull compares to drop testing.

Ryan Jenks posted a video on this subject (more-or-less) on Aug 11:  https://www.youtube.com/watch?v=OZwY0LZS4qU

Even includes a Zeppelin Bend pull test.

Ackley The Improvedwrote:

Knot hard to test thin cord or even thread without  a knot or direct clamp.

Much like a knot less / tesion less anchor achieved by wrapping a rope around a tree and held loosely in place with a knot

Wrap the cord around the Biner sized fixture several times and then clamp or knot

And then it breaks at the "Biner sized fixture", as evidenced by tests which use this method with biners (which happen to be biner-sized fixtures).

I think you're missing the point here. Sure, you can come up with various methods of fixing the ends, but there isn't an escape from decreasing the strength of cord where the cord or another object exerts pressure against the cord, no matter how you fix the ends.

The best you can do is something like Sam mentioned, using a "friction knot" around a large-radius object (certainly a lot larger than a carabiner). That still decreases the strength, but with a large enough radius, you could reduce the decrease in strength to within the range of strength along the length of the cord, which would mean you'd start to get breaks that aren't at one of your large-radius objects, but rather at some imperceptible weakness along the length of the cord. In this situation I'd still expect a disproportionate number of breaks near the attachments, but it you could simply discard those results to get an idea of the strength of the cord.

But, obviously that is going to require more tests, and a break system that can pull apart giant cylinders--I don't even know how large of cylinders you'd need because I don't know of any experiments that have tried this. Maybe you don't consider that "hard" but I do.

And even if you go through the effort to test this, so what? Is there a real-world application for this? Even if you're using a friction knot on one end of your cord, you're probably putting more acute pressure on some other part of your cord that's going to make that the weak point.

PTRwrote:

Ryan Jenks posted a video on this subject (more-or-less) on Aug 11:  https://www.youtube.com/watch?v=OZwY0LZS4qU

Even includes a Zeppelin Bend pull test.

Ryan Jenks/ HowNotToHighLine, is a great example of why slow pull tests are not a great representation of a climbing falls and that is why he is building the drop tower.

Watching the stretch and the failure modes in those test are interesting.  

David Kwrote:

Well...

Most of the time people doing this sort of test don't actually test the strength of the cord without knots, because it's hard to do that. Instead, they just use the manufacturer reported MBS for the cord. Your link returns a 404, so I don't know for sure if that's what they were doing, but that would be my guess.

The problem is, that's not always a valid assumption: MBS isn't necessarily the strength of the cord without knots. Testing the strength of cord without knots is hard for manufacturers too. There are ways to do it with clamps and whatnot, but these methods are harder to make work with thinner cords (less surface area to grip). And ultimately, these methods aren't representative of how the gear is used in practice, so you have to question the validity of testing the MBS using a clamp anyway. In at least some cases, the way manufacturers test the MBS of cord is the same way it's used in the field: with end knots tied in.

What that means for backyard scientists is that unless we know how the MBS was determined, we can't really say how much a knot reduces the strength of cord. For example, if we're only seeing a 14% reduction in strength from tying in a knot, it may be because we're testing the strength of the knot that the manufacturer used to test the strength of the cord, with perhaps some minor variation. In that case, we're not comparing the strength of the cord with and without a knot, we're simply comparing the strength of the cord with two different knots.

There's a more philosophical question here, which is, what does it even mean for the cord to have strength in the absence of outside factors? A clamp also reduces the strength of cord--if you just pull cord with two clamps, it will likely break at one of the clamps rather than in the middle of the cord, because the clamp itself weakens the cord. The strength of cord in the absence of some weakening factor is a hypothetical thing rather than a phenomenon which exists in reality, because we don't have any way (at least that I know of) to test cords which doesn't weaken them.

Is that a mistake? As far as I can find evidence for, this hasn't really been tested. Obviously there are some differences here, but do you have any reason to believe that those differences create a significant difference in the forces at which a cord (or other gear) breaks?

Sorry, it looks like the link did not work with Mountain Project, I fixed it.

It was the same link shared a couple of times in this thread.  https://user.xmission.com/~tmoyer/testing/High_Strength_Cord.pdf

"Pull rates and fixtures were consistent with CEN standards." ... unknotted and knotted....

One thing that would have made this test report better is more pictures of the test setup to make it more clear.  Like I said before, I hope that test like these are repeated by independent tests with modern material.

Preeti Pwrote:

Ryan Jenks/ HowNotToHighLine, is a great example of why slow pull tests are got a great representation of a climbing falls and that is why he is building the drop tower.

No, even Ryan Jenks is pretty clear about the fact that he doesn't know whether slow pull tests are representative of falls. He's said things to the effect of "Do slow pull tests show us what happens in a dynamic or shock loading scenario or a fall? I don't know, so we're building a drop tower to find out!" in numerous videos.

I credit Ryan Jenks a lot for that--good science starts from saying, "I don't know".

...which is why you should back off the idea that you know dynamic loading is different from slow pulls. Unless you have some evidence I don't know of, you don't know that. Maybe it is the same, maybe it's different, but so far there's no evidence I know of to say whether it's the same or different.

David Kwrote:

And then it breaks at the "Biner sized fixture", as evidenced by tests which use this method with biners (which happen to be biner-sized fixtures).

I think you're missing the point here. Sure, you can come up with various methods of fixing the ends, but there isn't an escape from decreasing the strength of cord where the cord or another object exerts pressure against the cord, no matter how you fix the ends.

The best you can do is something like Sam mentioned, using a "friction knot" around a large-radius object (certainly a lot larger than a carabiner). That still decreases the strength, but with a large enough radius, you could reduce the decrease in strength to within the range of strength along the length of the cord, which would mean you'd start to get breaks that aren't at one of your large-radius objects, but rather at some imperceptible weakness along the length of the cord. In this situation I'd still expect a disproportionate number of breaks near the attachments, but it you could simply discard those results to get an idea of the strength of the cord.

But, obviously that is going to require more tests, and a break system that can pull apart giant cylinders--I don't even know how large of cylinders you'd need because I don't know of any experiments that have tried this. Maybe you don't consider that "hard" but I do.

And even if you go through the effort to test this, so what? Is there a real-world application for this? Even if you're using a friction knot on one end of your cord, you're probably putting more acute pressure on some other part of your cord that's going to make that the weak point.

The CE test is wrapped 540° around a drum each end and clamped, 120mm dia if I remember rightly. Same for cord and tape, it's standard in the cordage industry.

David Kwrote:

No, even Ryan Jenks is pretty clear about the fact that he doesn't know whether slow pull tests are representative of falls. He's said things to the effect of "Do slow pull tests show us what happens in a dynamic or shock loading scenario or a fall? I don't know, so we're building a drop tower to find out!" in numerous videos.

I credit Ryan Jenks a lot for that--good science starts from saying, "I don't know".

...which is why you should back off the idea that you know dynamic loading is different from slow pulls. Unless you have some evidence I don't know of, you don't know that. Maybe it is the same, maybe it's different, but so far there's no evidence I know of to say whether it's the same or different.

The approach of letting the data do the talking is a great approach.

I'm not sure if I would say "I don't know" is part of the scientific method.    There should be some be at least a hint of logical hypothesis in there. Ryan Jenks is on the right path, though.

I think Ryan Jenks is upgrading to a drop tower because of solid criticisms of his slow pull test.  There is a reason why helmets are not tested by crushing them in a hydraulic press.  Helmets are tested in drop tests.   

There is good evidence on static vs dynamic in the links pasted in this thread.

From the frequented DMM anchor video, the anchors are with knots are performing better (less impact force) than the unknotted (sliding X).  

https://dmmclimbing.com/Knowledge/September-2013/Slings-at-Anchors

Jim Tittwrote:

The CE test is wrapped 540° around a drum each end and clamped, 120mm dia if I remember rightly. Same for cord and tape, it's standard in the cordage industry.

I believe you, but since the details of these tests are behind paywalls, there's still a lot of questions unanswered.

Where do the cords break when tested in this way? If they frequently break at or very near the drum, that would indicate to me that the drum is still compromising the cord slightly (though I wouldn't blame the industry for concluding that it doesn't compromise it enough to matter).

What was the motivation for choosing these parameters? Is the intention to get an accurate measurement of the strength of the uncompromised cord, or is it just to have a standard that isn't susceptible to variations in knot tying or other attachment methods?

David Kwrote:

I be2lieve you, but since the details of these tests are behind paywalls, there's still a lot of questions unanswered.

Where do the cords break when tested in this way? If they frequently break at or very near the drum, that would indicate to me that the drum is still compromising the cord slightly (though I wouldn't blame the industry for concluding that it doesn't compromise it enough to matter).

What was the motivation for choosing these parameters? Is the intention to get an accurate measurement of the strength of the uncompromised cord, or is it just to have a standard that isn't susceptible to variations in knot tying or other attachment methods?

No idea where it breaks but since it's probably been the industry standard test since before I was born I'll guess they are happy with it's results.

The manufacturers of cordage and tape don't know the application so logically want a test of the materials actual strength. For tape (webbing) there is a different test for sewn slings which are the only certified application apart from in harnesses, tying knots is out of remit and climbers are expected to do their  own testing or automatically de-rate the the product massively. Things like recommending knots in slings, cordalettes, doubling slings etc may lead to enormous loss in strength and the onus is on the promoters of these methods to test and warn the users.

Jim Tittwrote:

No idea where it breaks but since it's probably been the industry standard test since before I was born I'll guess they are happy with it's results.

The manufacturers of cordage and tape don't know the application so logically want a test of the materials actual strength. For tape (webbing) there is a different test for sewn slings which are the only certified application apart from in harnesses, tying knots is out of remit and climbers are expected to do their  own testing or automatically de-rate the the product massively. Things like recommending knots in slings, cordalettes, doubling slings etc may lead to enormous loss in strength and the onus is on the promoters of these methods to test and warn the users.

That all makes sense. I don't think we can conclude that the 120mm drum is large enough to not significantly affect the material strength from this. My guess would be that it does affect the material strength, but not enough to matter for any pragmatic purpose.

To be clear, even if it does affect the material strength that doesn't make the measurement "invalid" or "unscientific" in any way, it just means that the test answers a different question than "What's the strength of this cord across its length?" As I've said a few times, that question isn't particularly relevant to any applications, so manufacturers are right to test the materials the way they do. It wouldn't make sense for manufacturers to spend a lot of time, money, and effort to answer a question when nobody cares about the answer.

My only criticism of manufacturers is that they (usually) don't tell us how they tested their equipment. This leads to situations where we can't compare numbers from different manufacturers because we don't know how the numbers were obtained. But for the average climber (not a nerd like me) that's a minor criticism because almost all gear rated by climbing organizations is super good enough (as Ryan Jenks is fond of saying).

If it's certified then it's tested as required by the relevant EN, if it's not certified then who cares?

David, the inherent strength of a material across its length IS important to many uses. It baffles me how you can not see any possible way to use a cord or webbing without knots and how many posts you have made saying that having a number doesn't matter because its not real world or whatever. Having a very close idea to what that number is, means you can calculate the strength loss from various termination methods and determine for what use a particular material is suitable... You even said that if we don't know how the manufacturer calculates MBS, it complicates things for us.

Slacklines / highlines with all sorts of nifty webbing locks and grabs, sewn slings, sewn rope terminations, spliced terminations, rope termination plates, tensionless hitches used to span long distances, etc... Knowing the strength of the material without knots is incredibly important. There are so many ways in which materials are used without knots, in climbing, and in other markets. Having the theoretical strength of a material and having a good idea how much a knot will compromise its strength is a very good starting point... but how useful would that really be to like the sailing market using hollow braid dyneema where everyone is splicing? Do you think they would rather be given an MBS of their cord assuming figure 8 knots?

Desert Rock Sportswrote:

David, the inherent strength of a material across its length IS important to many uses. It baffles me how you can not see any possible way to use a cord or webbing without knots and how many posts you have made saying that having a number doesn't matter because its not real world or whatever. Having a very close idea to what that number is, means you can calculate the strength loss from various termination methods and determine for what use a particular material is suitable... You even said that if we don't know how the manufacturer calculates MBS, it complicates things for us.

1. It baffles me that you think I can't see any possible way to use cord or webbing without knots, given I've mentioned a few such methods in this thread. I've tried to be careful to NOT say "without knots" because I'm well aware there are other ways of using cords and webbing. If I somewhere mistakenly said "without knots", please point out where and I'll edit the post to correct my mistake, as that was never my intent.
2. It seems we agree that what is important is the strength of the termination method. The inherent strength of a material across its length isn't the only way to calculate that. For example, the testing method Jim Titt described may or may not be the inherent strength of the material across its length, but it's de-facto useful for getting an idea of the strength of the material when used with knots or other attachments, regardless.

Slacklines / highlines with all sorts of nifty webbing locks and grabs, sewn slings, sewn rope terminations, spliced terminations, rope termination plates, tensionless hitches used to span long distances, etc... Knowing the strength of the material without knots is incredibly important. There are so many ways in which materials are used without knots, in climbing, and in other markets. 

1. True, but this is a climbing forum, and more specifically, this is a thread about cordelettes. There are some sewn cordelettes/webolettes on the market, but I don't think we need to calculate the MBS of the sewn termination in that case because I think we can assume the MBS of the sewn termination is included in the MBS of the sewn cordelette/webolette given by the manufacturer.
2. See my point 2 above.

Having the theoretical strength of a material and having a good idea how much a knot will compromise its strength is a very good starting point... but how useful would that really be to like the sailing market using hollow braid dyneema where everyone is splicing? Do you think they would rather be given an MBS of their cord assuming figure 8 knots?

Nope. Which is why I didn't say that.

I'm aware that I'm nitpicking already, so I'm not going to criticize you for nitpicking. But if you're going to nitpick my nitpick, I do insist you respond to things I actually said. These nerdy discussions are already tedious enough when people read what you actually say. :)