Self-equalizing vs. static anchor

Ben,

Nope, but I have a survivors bias. There have been anchor failures over the years, though thankfully few. Given that the anchor is truly the last line of defense against death, it is worth putting a little too much thought into it.

John Long's writeup of the Equalette is pretty humble about the fact that there really have been VERY few anchor failures, and to my knowledge none attributed to the poor equalization of the Cordalette. However most anchors are NEVER stressed beyond body weight. How many WOULD fail if they actually had to handle a factor 2 leader fall? I'm guessing the number is larger than we'd like to admit to ourselves. So the pursuit of ever better anchors is worth rehashing endlessly.

Given some of the "anchors" I've seen other parties make, I'd argue this subject should not be downplayed. If you feel solid with your anchor skills, feel content to move on.

Edit:
These guys won't be posting up to answer your question:
cascadeclimbers.com/forum/u…

Moof,

The link to the actual accident report at the end of the thread you posted from CC is dead. The thread ended with speculation about whether or not it was actually anchor failure but someone was citing rock and ice as a source that it was anchor failure.

The thread also mentioned the fact that this curious belay that failed was set behind a flake. Not sure what to make of the story but I will say that no intricate amount of anchor rigging can make a flake stick to the wall any better. If that was indeed the cause for anchor failure.

Just curious if you know more details?

In the Gaines/Long book they tested equalization of a cordalette and found it did a very POOR job in vertically spaced anchors. It did a better job in horizontal anchors but not ideal by any means. Any off axis loading throws the whole thing out the door. Hence their equalette work.

Ya, I hear ya Moof. I'm not trying to be antagonistic here but rather simply posing a question that I'm curious about. I personally have never blown a piece in the 17 years I've been climbing and I am curious as to whether anyone else here has, how it happened and how it affected the anchor. Anyone??

Not a partial failure, but a 4 piece anchor that blew (all statically equalized using 4 clove hitches on a pair of double ropes). Only conjecture would answer whether the anchor would have held if it was better equalized:

friendsofyosar.org/rescues/…

9mm sounds kind of thick for double ropes. Also consider, that without any intermediate protection, those double ropes are acting like a single rope, effectively doubling the already significant fall forces.

Back in 2002 doubles went up to 9mm, and <10mm single ropes were still pretty rare. i would not be surprised if having two double ropes instead of a single made this situation worse, sounds plausible.

Another total anchor failure (about halfway down goes into details):
supertopo.com/climbing/thre…

Basically the dude placed a 2 piece anchor to save gear, and falls after placing his first lead piece. His lead protection pops. He knocks out his belayer. His anchor pops, but his belayer's limp body manages to stay on the ledge, and the friction of the unmanned ATC slows him down enough that he lives, but pretty messed up.

Morals of the story: Factor 2 falls happen. Pieces fail (lead and anchor). Build a real fricking anchor, not just 2 pieces. Don't fall.

Moof wrote:Not a partial failure, but a 4 piece anchor that blew (all statically equalized using 4 clove hitches on a pair of double ropes). Only conjecture would answer whether the anchor would have held if it was better equalized: friendsofyosar.org/rescues/…

i'm not sure if i'm visualizing it correctly but it doesn't sound to me as though the pieces were equalized, at least not all of them.

it sounds as though one lead rope was hitched to two pieces separately with one piece backing up the other rather than being equalized.

and, as for the other rope and two pieces, i can't tell if they're saying that the pieces were equalized on a sling or cord and then connected via clove hitch. or, if those two pieces were simply connected to the rope via the same clove hitch. there was no mention of a master point.

I agree with Mark and others that getting good pro and rigging it in a reasonable way is all that matters most of the time. But not absolutely all of the time, unfortunately. In the 50+ years I've been climbing, I've read of about five total anchor failures, all but one to very experienced climbers. I've also heard and read of pieces of an anchor blowing without the anchor completely failing.

Anchor failure is extremely rare, but you have to remember that events that test anchors significantly are also extremely rare. How many of our real-life anchors are actually up to withstanding all possible loads is unknown, and since no climbers that I know of build anchors and then drop-test them, it is fair to say that even highly experienced climbers know little about the security of their anchors. Comments of the sort "I've been climbing for X years and I've never had..." sadly mean almost nothing.

Although all these things have been discussed ad nauseum, it is clear, even from some of the posts on this thread, that the situation is still poorly understood. Here are a few things to ponder:

Because the arms in an anchor are elastic, the tension in them is going to be roughly inversely proportional to their length (roughly because I'm not considering the arm angles here), assuming the anchor was properly tied to begin with. This "theoretical" prediction has been tested and confirmed. The result is that longer anchor arms will develop less tension, and so equalization in a static anchor is generally impossible. This effect is most evident when the anchor pieces are vertically arrayed.

Equalizing systems require that the rope run around the anchor points to equalize strand tension. Unfortunately, friction in the system inhibits the equalizing effects. No one, to my knowledge, has done dynamic testing to see how three-anchor equalizing systems will behave under fall impacts. Static testing may not be relevant because of the difference between static and sliding friction. So we really don't know much about whether self-equalizing systems are much better at load distribution.

The idea that shock-loading for extendable anchors is a myth is itself a myth. The tests in Long's book do not properly represent anchor failure considerations and so cannot be taken as providing evidence for the irrelevance of shock-loading.

Simple theoretical considerations of the sort that have, in the past, been shown to give decent order-of-magnitude approximations suggest that shock-loading is potentially a serious issue for extending anchors. The critical ingredient is something analogous to the fall-factor, namely the amount of extension in the anchor divided by the length of the belayer's tie-in. The practical take-away is that you should tie in (with the rope of course!) as long as is feasible for extending anchors.

Three-point static anchors, with the three anchor points more or less horizontally arrayed, can fail in a cascade fashion if one of the outer arms fails. In such a case, the entire load is transferred to what was the middle piece, and if this fails that load goes to the remaining piece. I suspect (with almost no evidence) that this is what happens in the cases of known static anchor failure.

What to do in practice? If you believe your individual pieces are good, a classical static anchor with good angles should be fine. If some of the anchor arms are much longer than others, it might make sense to use some dyneema slings on those anchors to make the more elastic cordelette or rope components of approximately equal length.

If you can't get good anchors (for example, finger-size cams in a sandstone crack, the last such anchor I considered questionable), and if protection for the leader is not almost immediately available (it was not in the situation just mentioned), then you might want to try stacking the odds in your favor with an equalizing system. But just as importantly, you should be thinking about a dynamic belay if the leader does fall, regardless of the additional danger to the leader, because there is no advantage to stopping the leader more quickly and having the anchor fail.

Crag Dweller wrote: i'm not sure if i'm visualizing it correctly but it doesn't sound to me as though the pieces were equalized, at least not all of them. it sounds as though one lead rope was hitched to two pieces separately with one piece backing up the other rather than being equalized. and, as for the other rope and two pieces, i can't tell if they're saying that the pieces were equalized on a sling or cord and then connected via clove hitch. or, if those two pieces were simply connected to the rope via the same clove hitch. there was no mention of a master point.

I can't vouch for it actually being equalized, but in general when you use the rope to clove pieces together you try to statically equalize them, though this is one of the poorest equalization schemes out there, and under high load I can easily imagine a cascading failure where the lowest pieces would take most of the load, with one likely seeing more if the fall was not purely straight up and down. As each popped the next would be carrying most of the load. The gear described was not the biggest (stopper, blue alien, 0.5 and 0.75 camalot Jr's), but I've belayed off of less than at least told myself it was bomber.

Moof, Both instances you site admit to having poor and inadequate anchors. My question stands.

DannyUncanny wrote:that without any intermediate protection, those double ropes are acting like a single rope, effectively doubling the already significant fall forces.

That’s not how it works. Half-ropes treated as twin ropes do not double the rated impact force, it’s more like a 15 - 30% increase. The reason being is that when the load is carried by two ropes, the modulus of elasticity of the ropes change. Take the Beal Joker for example, its rated for 8.2 kN when used as a single and 9.5 kN when used as a twin. That’s only a 16% increase in impact force when you clip two versus one.

20 kN wrote: That’s not how it works. Half-ropes treated as twin ropes do not double the rated impact force, it’s more like a 15 - 30% increase. The reason being is that when the load is carried by two ropes, the modulus of elasticity of the ropes change. Take the Beal Joker for example, its rated for 8.2 kN when used as a single and 9.5 kN when used as a twin. That’s only a 16% increase in impact force when you clip two versus one.

Beal goes into a little more detail here:

beal-planet.com/sport/angla…

What I tried to make clear in the last anchor book is that isolating the anchor and trying to make it as bomber as possible is secondary to having a strategy that never allows the leader to fall directly on the anchor. As Rich pointed out, this is very rare, and should be avoided at all costs. The most important piece in the roped safety system is the top piece, because it absorbs direct loading in a fall, and safeguards the dreaded Factor Two whipper. Shoring up the anchor to withstand a Factor Two falls is sound practice but the focus should be on the top piece, the so-called Jesus Nut.

That much said, I tie in with the rope most of the time. The elaborate rigging systems are needed rarely, only when the individual pieces are crap. This is very rare on most popular routes. I figure things like an Equalete are needed only about 1% of the time, but they can be godsends when required.

Why everyone does not use a Quad on all two bolt anchors sort of baffles me since it is so light and pretty much achieves ideal equalization if a big biner is used.

JL

I think where testing may be beneficial in the future will be moving away from whether attaining equalization is important or not, and looking more at the resiliency of the overall system. From what I've seen in overall system tests, simple imperfect load distribution from anchor rigging using solid placements and good angles is fine & regardless of the material used -- cord, webbing, slings, rope, etc.

To rebut a point of Rich, the rescue community has performed several types of dynamic hits on systems and anchors and found anchors can hold and with more mass than what climbers are limited to.

Further testing should probably be looking at the effect of the main line/climbing rope as the weak point in continued energy. Notably, the first placement holding a portion of the fall then failing, resulting in material that is not as resilient once it hits the anchor but still has energy.

bomber bolts? ... i just use a sliding X ... KISS

good gear ? ... i just tie off with a fig 8 or clove master point ... KISS

weird shiet? ... plug in as many pieces as you can and pray ...

there are many ways to die climbing ... anchor failure i suspect is fairly rare ...

id be more concerned with running into a hungry bear on the approach and becoming sunday brunch for the bear family ....

bearbreeder wrote: id be more concerned with running into a hungry bear on the approach and becoming sunday brunch for the bear family ....

Well then, maybe you shouldn't breed so many bears...

I like the acronym NERDS better than SRENE
No
Extension
Redundant
Distribution
Solid

It's mostly the same except distribution replaces equalization. In the real climbing world you sometimes have points in your anchor matrix that may not be able to hold as much weight as other pieces, for a variety of reasons. For example: A well placed gold C4 will most likely hold more force than a #2 stopper. The reasoning is that you might want to put more weight on the C4 and less on the stopper because it is more solid.

Mark Nelson wrote:To rebut a point of Rich, the rescue community has performed several types of dynamic hits on systems and anchors and found anchors can hold and with more mass than what climbers are limited to.

Hey Mark, I'm always happy to be corrected, but would you mind explaining which of my points you are rebutting here? And while you are at it, do you have any references to the tests you mention?

rgold wrote: Hey Mark, I'm always happy to be corrected, but would you mind explaining which of my points you are rebutting here? And while you are at it, do you have any references to the tests you mention?

I'd also be interested in reading about these tests. In my experience, Rescue anchors are only vaguely similar to climbing ones. In SAR-HAR, EVERYTHING is MUCH MUCH stronger, over engineered and redundant. If these tests were with RESCUE anchor systems I'm not at all surprised they held.

What you want tested are the FUNKY anchors climbers run into and are forced to use since it's the only option. Weird Rock, small nuts, placements in non-ideal directions etc etc.