minimum diameter for nylon cordelette

[ek]

I know this is more relevant to rock climbing than to caving (or at least less relevant to US caving, where most people carry webbing rather than accessory cord), but I was wondering what the minimum diameter of nylon accessory cord is for use in tying cordelettes for rigging to anchors.

A cordelette, as most of you probably know, is a big loop of accessory cord used to tie load-sharing anchors between multiple (typically three, sometimes two) anchor points. The standard way is to load it in double-strand, and a "power point"/"master point" is tied with a figure-eight or, more rarely, an overhand.

7mm is the recommended diameter (for nylon), but I am wondering if 6mm (which some people use) is sufficiently strong.

[Scott McCrea]

5mm Spectra is often used for rigging. But you probably know that already.

[ek]

Yes, I have used 5mm Spectra for rigging. But I am specifically inquiring about the minimum safe diameter for a nylon cordelette.

[NZcaver]

You're a smart guy - look up the specs.

Figure out the MBS/MSWL for your particular nylon cordelette setup (single/double, knots, wet cord, etc), and decide for yourself. Am I missing something?

[chh]

I use 6 mil for multipitch. Even factored onto it once at a hanging belay. I'm not dead.

[Cody JW]

On this subject, I was wondering what you thought of using half inch tubular webbing for handline,I have some and have used it before.You can carry 50 feet of it alot more compact than rope.I think the strength on it is around 1200 lbs.

[NZcaver]

On this subject, I was wondering what you thought of using half inch tubular webbing for handline,I have some and have used it before.You can carry 50 feet of it alot more compact than rope.I think the strength on it is around 1200 lbs.

I have some of that stuff. The MBS is more than adequate for handline, but from a practical standpoint it has less surface area to grip. Especially noticeable if you have the smooth webbing rather than the ribbed stuff. But it's light, compact and it works in a pinch.

[ek]

You're a smart guy - look up the specs.

Figure out the MBS/MSWL for your particular nylon cordelette setup (single/double, knots, wet cord, etc), and decide for yourself. Am I missing something?

The cord will have an MBS for slow-pull without knots, but of course knots don't have established MBS values. My understanding is that, generally speaking, for a slow-pull (e.g. over-hauling), we take the knots to reduce the strength by a third (with wide error bars), and that for shock-loading, we take them to reduce the strength by half (with considerably narrower error bars, at least on the lower end). These are guidelines that I have been taught to apply to rope, and may or may not be applicable to thin accessory cord, but let's run with them and see what happens.

(The Manual of US Cave Rescue Techniques, Version 1.0 by Jeff Parnell--this is the official NCRC textbook, though perhaps a newer version has come out since 2005--says on page 18.3: "Almost all knots can reducce the strength of a rope by as much as 50% when shock-loaded. But under a static load various knots have a wide range of strengths[...]")

I must admit that I don't know how to take age and moisture into account. So for this first analysis, I'm going to ignore them. Therefore, if I conclude that some diameter cordelette is sufficiently strong, I could be mistaken (whereas, if the above assumptions are correct and I conclude that some diameter cordelette is not sufficiently strong, that conclusion can be trusted). NZcaver, can you give me some tips on how to account for age and moisture?

A cordelette is used doubled, but I am assuming the knot reduces the strength by (up to) half, so the strength of a tied cordelette is (at least, and not necessarily greater than) the MBS of the cord. Please note, again, that this does not take age and moisture into account.

A cordelette is typically attached to multiple anchors, and there is ideally some degree of load-sharing between them, and ideally none of them fail. But we can't rely on that--a cordelette is only strong enough if it withstands the highest fall force that we can expect, applied to a single arm (i.e. applied through the cordelette to a single anchor).

According to Climbing Anchors, 2nd Edition by John Long and Bob Gaines, no piece of rock protection rated to 10kN or higher has broken in a fall (though of course quite a few of them, poorly placed and/or in bad rock, have pulled out). This takes into account a very large number of falls, and a range of single person loads, and is probably the best indicator of the highest forces that lead falls can generate in rock climbing. Since the mechanics of ice climbing are essentially the same, 10kN should also be the highest force we have to worry about in ice climbing. Therefore, a cordelette used in rock climbing, ice climbing, or mountaineering is sufficiently strong if it can withstand a 10kN shockload.

OK, so cord has to be 10kN strong (or stronger) to be used for a cordelette. PMI's 7mm accessory cord has an MBS of 10.7kN, so it is sufficiently strong, whereas PMI's 6mm accessory cord has an MBS of 7.5kN, clearly not strong enough. Ergo, 7mm is the minimum safe diameter for a nylon cordelette, which explains the widespread view that 7mm is the right choice in diameter for such a cordelette.

However, there are some complications. First of all, my analysis has (as I've pointed out a couple of times before) ignored strength degradation due to age and moisture. (It has also ignored flex fatigue, but my understanding is that flex fatigue is negligible in nylon and polyester. I've heard that rappelling on nylon rope hundreds of times in rapid succession reduces its strength, suggesting that flex fatigue is a potential issue...but you don't do anything like that to your accessory cord.) Perhaps the strength reduction due to these factors is negligible--they both reduce elasticity in nylon, but while a bit of elasticity in the cordelette can potentially reduce the force of a fall onto it, the primary shock absorbing component of a climber's fall arrest system is the dynamic rope, and it's the dynamic rope, and not the cordelette, that has to remain elastic.

If a 7mm nylon cordelette is definitively strong enough, but not by a superfluous margin, then what about the equalette? Championed by Long and Gaines and Climbing Anchors, 2nd Edition, an equalette is a cordelette with two overhand knots tied near the middle. It functions like a big sliding-x with two load-limiting knots. You can use it to dynamically distribute the load between two anchors, while limiting extension if one fails. Since three pieces of hand-placed protection is considered the appropriate level of redundancy in most climbing anchor-building situations, a more common way to use an equalette is for dynamic load-distribution between a single anchor and a load-sharing configuration of two anchors (three pieces of protection), or for dynamic load-distribution between a two load-sharing configrations of two anchors each (four pieces of protection). The load-sharing is accomplished by clove-hitching the anchors to a single strand of cord (as two strands of cord come out of each extension-limiting knot--you clove hitch one anchor to each strand).

By the reasoning I have used to "conclude" that 10kN is the minimum acceptable strength cord used as a cordelette, I would have to conclude that 20kN is the minimum acceptable strength for cord used as an equalette!

Something is wrong here. I have the sense that I am overestimating the strength requirements for cordelettes...

[chh]

ek, I think you're forgetting that with the equalette each strand in between the knots gets a biner and then your power point/rope/whatever goes to both of them. So you essentially have a "doubled" cord there.
There is also a way to tie the equalette in such a way as to have three strands in between the knots if you prefer to use one big locker as your power point as opposed to the 2 (or possibly 3) you need with the regular equalette. You clip two and the third essentially acts like a keeper. In order to do this you make a big "S" with your cord and then tie the knots at the place you want them. You end up with a loop and an end on both sides of the knots, if that makes sense.
I used the equalette for a while but found it cumbersome as opposed to the regular cordolette. But, most of the time my belays offer the availability of solid gear that I'm not worried about equalizing as much. It's a good thing to know though. I think in the climbing world it's more handy to know how to place and equalize two pieces with one hand

edit:
You can also use the regular equalette with a single power point locker if you clip both the strands between the knots sliding x style. That is to say put a half turn in one of them. However this will load the one with the half turn in it more unless you account for that tiny bit of extra cord needed when tying the equalette. If you tie it loosely and then weight it with the power point clipped it usually sorts itself out.

[chh]

I reread your post ek and realized that i probably didn't even address your question. If you're concerned about the "single arm", than your reasoning could be considered correct. However, the whole reason behind equalization is that it will theoretically always load the two strands in an equalette in the same manner, thereby giving you 20 kns at your anchorpoint with 7 mil cord if it functions correctly. You are never on a single strand. I'm sure you could come up with some scenarios that would deprive the equalette of one of these strands, abrasion for example, but I would say that this is just one of the trade offs for equalization. And I think it unlikely, particularly if you are using the equalette climbing, where someone will typically be looking at it most of the time - roped soloing excluded.

[ek]

I potential point of failure on the equalette that I'm worried about is where it is clove hitched to protection, and not at the power point / master point.

The equalette can only distribute the load if there are anchors to distribute the load to. If an anchor on either side fails, all the load can go to one side, and then the cord is no longer doubled for increased strength.

Furthermore, even if anchors do not fail, the angle between the dynamically "equalized" arms would have to be very low for the strength to be twice the strength of the cord. With angles higher than zero degrees, each side takes more than half the load.

However, please note that while I believe that it is possible that the equalette is prone to failure in a way that I am first recognizing, I think it is much more likely that I am missing something and that I have calculated an overly high minimum safe strength for accessory cord used in cordelettes and equalettes. I am bringing up the equalette to demonstrate that something seems fishy with my analysis, and to request assistance in identifying what that might be.

[knudeNoggin]

There is several things to comment on in this issue. Perhaps Bob Thrun will
contribute some remarks about discriminating between rates of force upon
such things as rope -- I know that he often challenges the common notions
of "shock load" as more myth than substance. (We have on this forum some
testimony from Dave Merchant that a Fig.9 eye knot is weaken on dynamic
loading than trad. slow-pull, but also that it's a sort of "YMMV" situation as
knots differ (e.g., the Overhand eye knot has more uniform strength).)

1) Thanks for the URLink to that paradigm image of a cordellete.
Now, please note that is it sub-optimal, and really there should be but
ONE KNOT --viz., the "powerpoint"-- in the structure, not two; the two
free ends should be left free (or tied off with an Overhand stopper snug
to the powerpoint) on the clip-in (sort arms) side of the powerpoint
-- there is no reason to join the ends in some anchor leg! (This freedom
leaves the cord available for non-cordelette uses; it yields uniformly
strong anchor legs.) The clip-in eyes likely if anything make for a weaker
structure if there are three of them spreading the load on the 'biner than
just two; two eyes are way more strength than your body can handle
(and are double the single anchor leg, of course).

2) 5mm "spectra" , "strength" : mostly in the situations encountered where
cordelettes are used one is concerned about an impact load; static material
such as HMPE (Dyneema, Spectra) or aramid (Kevlar, Technora) fibres will
increase the peak impact force. Thus, whatever equal or greater tensile
strength they might show which looks so good to marketers of a product
are less good in practice. Moreover, knots weaken these materials more
than they do for nylon.

3) As has been presented in John Long's substantially revised, 2nd edition
of Climbing Anchors , the unequal legs of a cordelette make for unequal
load distribution (barring some clever anticipation of the effect by adjusting
each arm's tension at the tying). It should be pretty easy to see that it doesn't
take much of an angle shift with static material such as HMPE cord to have
just one leg loaded.

4) As for demanding that a single anchor leg sustain maximum expected load,
that too might be overdoing it: surely one must expect that the other two
legs are not worthless --i.e., they will require some significant force to fail-- ,
and as the expected loading is dynamic (what else?), it will be diminished by
the failing of those other arms; so, the maximum impact force to the overall
structure will be reduced before it hits an isolated leg (when loaded in the
expected angle to the cordelette).

*kN*

[ek]

There is several things to comment on in this issue. Perhaps Bob Thrun will
contribute some remarks about discriminating between rates of force upon
such things as rope -- I know that he often challenges the common notions
of "shock load" as more myth than substance.

OK...but so what? What effect does that have on this issue? What effect does that have on my calculation? An overhand knot tied to establish the power point / master point would still be considered to reduce the cord's strength by up to half, wouldn't it?

1) Thanks for the URLink to that paradigm image of a cordellete.
Now, please note that is it sub-optimal, and really there should be but
ONE KNOT --viz., the "powerpoint"-- in the structure, not two; the two
free ends should be left free (or tied off with an Overhand stopper snug
to the powerpoint) on the clip-in (sort arms) side of the powerpoint
-- there is no reason to join the ends in some anchor leg! (This freedom
leaves the cord available for non-cordelette uses; it yields uniformly
strong anchor legs.)

This is irrelevant to my question--it does not affect the strength of the cordelette as per my calculations. However, this is a useful thing to consider and discuss; thank you for bringing it up.

The way you're saying we should tie cordelettes is perfectly acceptable, but not demonstrably superior. Typically one's protection will be sufficiently far apart that there will be little advantage to having the excess cord available. The excess cord hanging down can get in the way, and one might be tempted to daisy-chain it...which would render the method significantly more cumbersome. Even in instances where you wouldn't have to use all the cord, most people usually would (unless the pieces of protection are alarmingly close to each other, which is potentially problematic in and of itself), because doing so would reduce the angle between the pieces of protection and thus put (somewhat) lower loads on the protection.

Furthermore, it's more cumbersome anyway, because when you have a pre-tied loop, you clip a single strand to each piece of protection, collect the cord between them, and pull down. I believe the method you are suggesting is one that most (though not all) people would find harder to use, in rendering the anchor load-sharing.

Finally, this only makes the cordelette stronger if the double-fisherman is where it breaks, rather than the overhand or figure-eight.

Have you actually used a cordelette in the way you're recommending? Have you used a cordelette in the traditional way as well? If yes to both, can you compare your experience with both, regarding ease of use? I have not used a cordelette in the way you're recommending, so I am not speaking from experience in criticizing it.

The clip-in eyes likely if anything make for a weaker
structure if there are three of them spreading the load on the 'biner than
just two; two eyes are way more strength than your body can handle
(and are double the single anchor leg, of course).

I agree that a single loop is strong enough, if the cord itself is strong enough. The simple reason for that is that you have single loops for all the pieces of protection. However, it is worth mentioning that it is not necessarily true that your anchor, to be safe, only has to be as strong as your body can withstand. Suppose that someone clips to the power point / master point of the anchor system as their first piece (rather than to one of the pieces). If they fall, the force on the cordelette is nearly twice the force on the falling climber, just as when you clip any piece of protection and fall onto it. You're "hauling" on your protection with a 2:1 (albeit an inefficient one) when you take a lead fall.

In any case, in my above analysis I work from the assumption that 10kN is the highest force that will be applied to any anchor in climbing.

The long-axis strength of a carabiner is far higher than the strength of just about any other anchor component, in just about any situation. Reduction of strength in the carabiner should still not render the carabiner the weak link in the system.

2) 5mm "spectra" , "strength" : mostly in the situations encountered where
cordelettes are used one is concerned about an impact load; static material
such as HMPE (Dyneema, Spectra) or aramid (Kevlar, Technora) fibres will
increase the peak impact force. Thus, whatever equal or greater tensile
strength they might show which looks so good to marketers of a product
are less good in practice. Moreover, knots weaken these materials more
than they do for nylon.

Agreed. Furthermore, the stretch in a nylon cordelette makes it distribute the load better than a more static material like Spectra (see below). However, this is not relevant to my analysis above. I am not telling people to start using a stronger material. I will continue to use 7mm nylon myself.

3) As has been presented in John Long's substantially revised, 2nd edition
of Climbing Anchors , the unequal legs of a cordelette make for unequal
load distribution (barring some clever anticipation of the effect by adjusting
each arm's tension at the tying).

If I recall correctly, Long and Gaines don't say anything about cleverly anticipating the effect. I think you've come up with that.

I thought of that too, when I was reading the book. But then I realized that you could only adjust them to be equalized at one particular tension. They would be unequally loaded at tensions smaller than that (and the protection could potentially fail at those smaller loads), and they would be unequally loaded at bigger tensions.

This issue is also mostly irrelevant to my analysis, though it arguably boosts in a bit because it shows that, with a very short short arm, you can have almost all the load on one arm of the cordelette even if other arms do not fail.

It should be pretty easy to see that it doesn't
take much of an angle shift with static material such as HMPE cord to have
just one leg loaded.

I don't think it takes much of an angle shift with more dynamic material like 7mm nylon to have just one leg loaded, either. But I do agree that this is one way in which nylon should fare better than Spectra and other extremely static materials.

4) As for demanding that a single anchor leg sustain maximum expected load,
that too might be overdoing it: surely one must expect that the other two
legs are not worthless --i.e., they will require some significant force to fail-- ,
and as the expected loading is dynamic (what else?), it will be diminished by
the failing of those other arms; so, the maximum impact force to the overall
structure will be reduced before it hits an isolated leg (when loaded in the
expected angle to the cordelette).

This might happen, but it very well might not, too. What if the arms fail before they are heavily loaded? (Sometimes you're in a situation where the protection is not so good...some may be strong enough to withstand the maximum possible loading, while others might be weak as hell due to instability of the rock or some other factors.)

Furthermore, what if all the pieces but one fail before the impact force reaches the peak?

Furthermore, what if the pull is off-axis so there is no equalization at all?

Furthermore, what if the short arm of the cordelette is way shorter than any other arm? Then the rigging still accomplishes minimizing extension should between one and all-but-one of the pieces fail, but little or no equalization is achieved (see above in this post).

It is unacceptable to have a cordelette that is too weak to take the maximum possible impact on a single arm.

EDIT: Changed "a more static material than Spectra" to "a more static material like Spectra."

[knudeNoggin]

... Perhaps Bob Thrun will
contribute some remarks about discriminating between rates of force upon
such things as rope -- I know that he often challenges the common notions
of "shock load" as more myth than substance.

OK...but so what? What effect does that have on this issue?

Goodness, this is just a couple posts up, and by your keystrokes:
"My understanding is that, generally speaking, for a slow-pull (e.g. over-hauling),
we take the knots to reduce the strength by a third (with wide error bars),
and that for shock-loading, we take them to reduce the strength by half ..."

An overhand knot tied to establish the power point / master point would still be considered to reduce the cord's strength by up to half, wouldn't it?

That's a simple rule of thumb (made risky for high-modulus cordage), but, no,
not a best guess, I think: CMC Rope Rescue Manual has the Overhand as being
the strongest of its tested knots (by a %pt over one of the two ways of tying
a Fig.8 eye knot); and at the powerpoint, well, I'd expect the bulk there
to help it (I think the "Big Honking Knot" was tested by one AMGA guy, btw).
And then one can find other data, though usually not much to explain it.

.:. "50%" is an easy factor for calculations where you want to err conservatively!

1) Thanks for the URLink to that paradigm image of a cordellete.
Now, please note that is it sub-optimal, and really there should be but
ONE KNOT --viz., the "powerpoint"-- in the structure, not two; ...

This is irrelevant to my question--it does not affect the strength of the cordelette as per my calculations. However, this is a useful thing to consider and discuss; thank you for bringing it up.

Yes, OT but a convenient point to remark, given the clear image.

The way you're saying we should tie cordelettes is perfectly acceptable, but not demonstrably superior. Typically one's protection will be sufficiently far apart that there will be little advantage to having the excess cord available. The excess cord ...

Um, you're going overboard in analysis. I don't know why you'll typically have wide
anchor placements --which if anything aggravate forces & equalization problems-- ,
but sparing the Grapevine gives you material to use. It also ensures that there will be
no jammed knot to wrestle with. You can knot the ends (with an EDK, EDK :o) on the
short-eyes side of the powerpoint qua stopper knot, if you'd like (but not clip them).

.:. There has been from the cordelette conception a presumption of a closed circle
of rope in which to effect it; I simply want to challenge that assumption, pointing
out possible benefits to NOT tying the ends, or what how to position such a knot if made.

*kN*

[chh]

ek, I think it is possible to load just one strand of the equalette if you were to lose an anchor piece in certain setups. For example, say a loop to one piece on the "left" side of the power point and two cloved strands to two pieces on the "right" side of the power point. If the piece with the loop fails you will load the other two pieces individually much like the arms in a traditional cordolette would be loaded if a piece popped, except obviously they would only be one strand, and I reckon there could be a little more of a pendulum effect allthough this would depend greatly on the configuration of the pieces themselves. If we assume that all the placements are equal, you would be in trouble. Is this what you're getting at?