Forums: Climbing Information: The Lab:
Stupid Simple Elette
RSS FeedRSS Feeds for The Lab

Premier Sponsor:

 
First page Previous page 1 2 3 4 Next page Last page  View All


Neoshade


Aug 21, 2011, 12:27 AM
Post #51 of 88 (3221 views)
Shortcut

Registered: May 11, 2009
Posts: 31

Re: [rgold] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

I see what rgold is saying as classic SOP (standard operating procedure)
As the leader, you say "Off Belay" once you're safe, and clipped into a personal anchor (The rope coming off your harness or a sling/PAS/daisy). Then your second begins dismantling the anchor below and prepares to climb. The second only removes the last 1 or 2 pieces (how little or much he leaves depending on the security of the anchor and his hanging/standing position to it), as the last act before climbing up with the shout from above of "On Belay!"

The idea is, when you're at a belay station, and ready to work the rigging, only 1 or 2 pieces of pro are needed to call it "safe" and switch out the belay. You're not going anywhere, and building or dismantling the rigging has to take place somehow. and overlapping the belays (building a rig on belay, or dismantling most of a rig on belay) is considered overkill by most experienced climbers.

However, (because it can't be emphasized enough) Never take even the shortest fall on a nylon or dyneema personal anchor sling. If you clip in with a personal anchor, and not the rope, cinch it tight to your belay, sit your a$$ down and don't climb even a foot above it. This is the exception to the rule that will get you killed. And violating this simple rule is really the only case for overlapping the belay as long as you and your second understand your anchors/pro.


patto


Aug 21, 2011, 3:34 AM
Post #52 of 88 (3212 views)
Shortcut

Registered: Nov 14, 2005
Posts: 1446

Re: [Neoshade] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Yep. All that is part of efficient multipitch climbing.

Getting your anchor set up should take only a few minutes. You should call Safe/OffBelay as soon as you feel appropriately secure. One bomber piece and I'll do that, two if I'm feeling paranoid.

As a second you should putting shoes and and dismantling the anchor. and be ready to go as soon as you hear the signal.

Interchanges are what make efficient multipitch climbing. I aim for a speed of 30mins per 50meters. But it all depends on the difficulty of the terrain.


mbrd


Aug 21, 2011, 9:06 AM
Post #53 of 88 (3197 views)
Shortcut

Registered: Aug 11, 2011
Posts: 67

Re: [rgold] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

sorry if i was unclear- turns out you've been clear enough for both of us.

we're on the same page here.


jktinst


Sep 2, 2011, 11:16 AM
Post #54 of 88 (3124 views)
Shortcut

Registered: Jun 29, 2010
Posts: 85

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Jim, please, do write that article.

In the swing resulting from an off-centre loading, some of the equalization systems make the tape or cord run through several of the biners or rings involved (Alpine equalizer, ACR, upper right part of Neoshade's latter system, etc.) whereas for others, it runs only through the central biner (eg the two-arm equalette with extension-limiting knots). I can feel that the former have more friction than the latter. On the other hand, in an off-centre loading situation, the pro closest to the direction of pull is initially going to get a greater portion of the load. So, without a more detailed description of the test, I keep wondering if the large difference you are reporting is really due solely to biner friction or if it might be due to a combination of friction and skewed loading.

Although this discussion has put a big question mark for me over the equalizing set-ups that involve friction over multiple biners, for now, I remain convinced that, even with less-than-perfect distribution of the load and the need to pick a compromise between the range of equalization and the limitation of extension, low-friction equalizing belays are still preferable to the alternatives that don't equalize at all under an off-centre load. Intuitively, I do buy into the belief that some equalization of multiple pros is better than loading each one sequencially.


patto


Sep 2, 2011, 2:54 PM
Post #55 of 88 (3096 views)
Shortcut

Registered: Nov 14, 2005
Posts: 1446

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

jktinst wrote:
Intuitively, I do buy into the belief that some equalization of multiple pros is better than loading each one sequencially.

A well configured statically equalised anchor will roughly equalise the load over the pieces. If dynamic rope is used it will equalise even more so.

In the vast majority of cases the load is only going to be in two possible directions the closest piece or directly down. So unless the closest pieces is at a significantly different angle from the down direction then you don't have a problem.

If its factor twos that you are designing for then you ALREADY know the direction.


JimTitt


Sep 5, 2011, 3:44 AM
Post #56 of 88 (3037 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

jktinst wrote:
Jim, please, do write that article.

In the swing resulting from an off-centre loading, some of the equalization systems make the tape or cord run through several of the biners or rings involved (Alpine equalizer, ACR, upper right part of Neoshade's latter system, etc.) whereas for others, it runs only through the central biner (eg the two-arm equalette with extension-limiting knots). I can feel that the former have more friction than the latter. On the other hand, in an off-centre loading situation, the pro closest to the direction of pull is initially going to get a greater portion of the load. So, without a more detailed description of the test, I keep wondering if the large difference you are reporting is really due solely to biner friction or if it might be due to a combination of friction and skewed loading.

Although this discussion has put a big question mark for me over the equalizing set-ups that involve friction over multiple biners, for now, I remain convinced that, even with less-than-perfect distribution of the load and the need to pick a compromise between the range of equalization and the limitation of extension, low-friction equalizing belays are still preferable to the alternatives that don't equalize at all under an off-centre load. Intuitively, I do buy into the belief that some equalization of multiple pros is better than loading each one sequencially.

The angles aren´t relevant, in practice you don´t have the option to change them anyway and the effect is what matters not the cause.
However since I´m working on the cause as well I did the easy thing to check and eliminated the angles altogether by arranging the points vertically above each other. You get the same results except that the narrower the angle between the pieces makes the sensitivity of the load direction more and more sensitive until the vertical case where the difference between the high and low on the pieces is physically impossible to detect and it is going to be impossible to manually equalise as well.

The big question is are there alternatives which give better results than equalising systems which often give results worse than just guessing. As the DAV decided the benefits of minimising extension are clear whereas the benefits of dynamic equalisation are effectively non-existent so redundancy is the choice giving clear benefits.

Jim


patto


Sep 5, 2011, 4:09 AM
Post #57 of 88 (3034 views)
Shortcut

Registered: Nov 14, 2005
Posts: 1446

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

JimTitt wrote:
jktinst wrote:
Jim, please, do write that article.

In the swing resulting from an off-centre loading, some of the equalization systems make the tape or cord run through several of the biners or rings involved (Alpine equalizer, ACR, upper right part of Neoshade's latter system, etc.) whereas for others, it runs only through the central biner (eg the two-arm equalette with extension-limiting knots). I can feel that the former have more friction than the latter. On the other hand, in an off-centre loading situation, the pro closest to the direction of pull is initially going to get a greater portion of the load. So, without a more detailed description of the test, I keep wondering if the large difference you are reporting is really due solely to biner friction or if it might be due to a combination of friction and skewed loading.

Although this discussion has put a big question mark for me over the equalizing set-ups that involve friction over multiple biners, for now, I remain convinced that, even with less-than-perfect distribution of the load and the need to pick a compromise between the range of equalization and the limitation of extension, low-friction equalizing belays are still preferable to the alternatives that don't equalize at all under an off-centre load. Intuitively, I do buy into the belief that some equalization of multiple pros is better than loading each one sequencially.

The angles aren´t relevant, in practice you don´t have the option to change them anyway and the effect is what matters not the cause.
However since I´m working on the cause as well I did the easy thing to check and eliminated the angles altogether by arranging the points vertically above each other. You get the same results except that the narrower the angle between the pieces makes the sensitivity of the load direction more and more sensitive until the vertical case where the difference between the high and low on the pieces is physically impossible to detect and it is going to be impossible to manually equalise as well.

The big question is are there alternatives which give better results than equalising systems which often give results worse than just guessing. As the DAV decided the benefits of minimising extension are clear whereas the benefits of dynamic equalisation are effectively non-existent so redundancy is the choice giving clear benefits.

Jim

I like your work. That is what I have been arguing for 4 years on this site. But the sliding-x never seems to die around here.


livinonasandbar


Sep 5, 2011, 8:09 AM
Post #58 of 88 (3020 views)
Shortcut

Registered: May 3, 2003
Posts: 356

Re: [potreroed] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

potreroed wrote:
I've been climbing for 45 years and have never used a 'lette of any type. Why complicate things? I'm with patto on this one. Use the rope and the PAS which you'll have already girth-hitched on your harness if you're smart.

Ed, if you're not swapping leads, what do you do? Just untie and swap rope ends?


JimTitt


Sep 5, 2011, 9:12 AM
Post #59 of 88 (3008 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [livinonasandbar] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Since time immemorial or even before I and Ed started climbing it was standard practice that the first climber clips into a karabiner using a fig.8 to form a loop. The other climber does the same, slipping the loop up from underneath through the first climbers loop and clipping on.
Removal for the first climber to continue leading is then easy.

Jim


JimTitt


Sep 5, 2011, 9:21 AM
Post #60 of 88 (3006 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [patto] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

My pleasure Sir.

I´ve just been trying some tests with average punters to see how well they achieve equalisation and one of the most popular `equalising´methods got a load distribution of 85%, 15% and 0% and with no apparent way of pre-determining which was going to be which.

We were impressed!

Jim


potreroed


Sep 5, 2011, 9:51 PM
Post #61 of 88 (2966 views)
Shortcut

Registered: Sep 30, 2001
Posts: 1440

Re: [livinonasandbar] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

livinonasandbar wrote:
potreroed wrote:
I've been climbing for 45 years and have never used a 'lette of any type. Why complicate things? I'm with patto on this one. Use the rope and the PAS which you'll have already girth-hitched on your harness if you're smart.

Ed, if you're not swapping leads, what do you do? Just untie and swap rope ends?

I have done that a few times but not very often. I almost always belay off my harness facing the wall with a re-direct and stack the rope by butterflying it across the rope in front of me. When my partner is secure it's a simple matter of flipping the whole stack over onto him/her so it feeds out properly.


(This post was edited by potreroed on Sep 5, 2011, 10:39 PM)


qwert


Sep 6, 2011, 6:19 AM
Post #62 of 88 (2950 views)
Shortcut

Registered: Mar 24, 2004
Posts: 2392

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

JimTitt wrote:
My pleasure Sir.

I´ve just been trying some tests with average punters to see how well they achieve equalisation and one of the most popular `equalising´methods got a load distribution of 85%, 15% and 0% and with no apparent way of pre-determining which was going to be which.

We were impressed!

Jim
Any chance that we are going to get some more info on that?
Which popular method was it, how did you set up the test, etc.

Was this just you playing around, or will this eventually lead to a bigger report, and hence you cant publicise your intermediary results?

qwert


JimTitt


Sep 7, 2011, 1:19 PM
Post #63 of 88 (2898 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [qwert] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

I´m writing a bigger article but I´ve been touring around in a tent with my family for the last 3 weeks so not a lot of internet or writing time. Got back home a couple of hours ago so I´ll press on a bit!

Jim


Partner cracklover


Sep 7, 2011, 3:18 PM
Post #64 of 88 (2880 views)
Shortcut

Registered: Nov 14, 2002
Posts: 9935

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

JimTitt wrote:
I´m writing a bigger article but I´ve been touring around in a tent with my family for the last 3 weeks so not a lot of internet or writing time. Got back home a couple of hours ago so I´ll press on a bit!

Jim

Any chance you could toss a Mooselette in the mix if you're testing anchor methods? Just like the Dos Equis Man - I don't always use a cordelette, but when I do, it's always in Mooselette form.

Been using it for about 5 years, and am happy with it. But it sure would be nice to get real numbers.

Here's a link: http://www.rockclimbing.com/...rum.cgi?post=2512061 (3 posts down on that page)

BTW, it seems much happier if you use a big pear or HMS style biner on the powerpoint - wide side up. A small biner will pinch the strands together and make the friction bad.

GO


jktinst


Sep 10, 2011, 3:27 PM
Post #65 of 88 (2805 views)
Shortcut

Registered: Jun 29, 2010
Posts: 85

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

JimTitt wrote:
The angles aren´t relevant, in practice you don´t have the option to change them anyway and the effect is what matters not the cause.
However since I´m working on the cause as well I did the easy thing to check and eliminated the angles altogether by arranging the points vertically above each other. You get the same results except that the narrower the angle between the pieces makes the sensitivity of the load direction more and more sensitive until the vertical case where the difference between the high and low on the pieces is physically impossible to detect and it is going to be impossible to manually equalise as well.

The big question is are there alternatives which give better results than equalising systems which often give results worse than just guessing. As the DAV decided the benefits of minimising extension are clear whereas the benefits of dynamic equalisation are effectively non-existent so redundancy is the choice giving clear benefits.

Jim

I’m afraid that I didn’t understand any part of Jim's response in relation to my questions but I don't want to bug him for more explanations if a more complete article is coming.

This thread has sent me back to pore over the Ewing tests reported in the Long&Gaines book, converting the results to units that could be compared with this thread. Having done this, I thought that it might be helpful to this discussion to provide a synopsis. Apologies to those who are already familiar with the tests and sorry once again for a long post.

In the tests, a 100kg weight was dropped for 0.5 m and stopped by a 0.5 m length of dynamic rope attached to the central biner of a two-arm system (FF1). Each attachment point measured the portion of the load that it was subjected to. The rigs tested mimicked 2 placements in a horizontal crack (arms of equal length at 24 deg) or in a vertical crack (arms of unequal lengths at 0 deg). The tests compared the cordelette and the sliding X. The equalette was introduced later in the "unequal" configuration only. Many experimental protocol details were left out of the report (although nothing critical as far a I can tell) and a few points are a bit unclear; eg how come that, using slings of the same length, the "equal" tests ended up with the same arm lengths for both riggings (70 cm), whereas for the "unequal" tests, the arms of the cordelette rig were significantly shorter (55 & 12 cm) than those of the sliding X and equalette (100 & 45 cm for both). I'll also skip the comparisons of different cord and tape materials (nylon, high tensile) that don't seem essential for this discussion.

The results reported were the differences in load "felt" by each arm. What would have been useful for this discussion is the average dynamic load (sum of each arm's load) resulting from the drops. The closest the book comes to giving this is "about 1500 to 2000 lbs". I translated this as 8 kN +/- 1 kN and used 8 for the calculations of equalization ratios.

- Cordelette equal: mean difference of about 0.9 kN for an equalization ratio of about 1:1.25.
- Sliding X equal: about 0.25 kN for a ratio of 1:1.06.
- Cordelette unequal: about 3.3 kN for a ratio of 1:2.40 with high "regular" variability.
- Sliding X unequal: 1 kN for a ratio of 1:1.29 with generally consistent results but with occasional tests yielding a much greater difference ("outlyers"). These were considered to be caused by the "clutch effect" (resulting from one strand crossing over the other on the underside of the central biner's basket bar and occasionally effectively squeezing/braking it against the bar, limiting equalization).
- Sliding X unequal with the clutch effect manually eliminated (not an option while climbing): 0.5 kN (1:1.13).
- Equalette unequal: 0.5 kN (1:1.13).

The very poor equalization performance of the cordelette unequal was attributed to the fact that the shorter arm stretched a lot less than the longer one (even using static cord), causing it to bear a much greater portion of the load. I am not sure that using dynamic rope instead of static materials (as suggested by patto) would have made any difference: the longer arm would still have stretched more than the short. Given this explanation, it would have been useful for the tests to use the same arm lengths in both systems. If they had, the cordelette might not have looked quite as bad but there is no doubt that the sliding X would still have equalized significantly better. This also means that in drop tests with the equalette and the sliding X, some of the material from the long arm ran through the biner to compensate for the lesser stretch on the short side. The interesting thing in the context of this discussion is that it did so much more efficiently than the 1:1.5 ratio suggested by Jim for a single biner's friction.

Of course, this "stretch" explanation does not necessarily work to explain the poorer performance of the cordelette equal. It seems quite improbable that the difference reported could be ascribed to the very small differences in arm lengths resulting from the difficulties in making the two arms exactly the same length with the cordelette. A more likely reason is the difficulty in releasing the weight from a position exactly in line with the central biner. Most of the drops were probably inevitably very slightly off-center, resulting in somewhat uneven loading of the two attachment points. Of course, similar practical experimental constraints would have been encountered with the sliding X but, as the unequal tests suggest, this rig's dynamic equalization capabilities must have taken care of these small variations.

An analysis of equalization tests using 3-arm systems and off-centre drops, describing the experimental protocols clearly in addition to providing the results, would complement and expand on the Ewing tests very nicely and would be very gratefully received. By experimental protocols, I mean not just the rig configurations but also the steps involved in setting them up, whether, for example, the rigs were redone from scratch for each drop test or if several drops were performed after a single setup, etc. It may sound like too much information but these details could turn out to be essential for a proper interpretation of the results.

Despite my introductory sentence, I must say that I'm quite impatient to find out more about how the "85-15-0" test was performed and on what rig. I have a very hard time imagining a rig generally considered to be equalizing and a test protocol that would repeatedly yield "0" load on one of the 3 arms with no apparent way of pre-determining which. All I can think of would be a highly friction-challenged rig tested with an off-centre drop without having pre-stretched the 3 arms first (ie having left the arms randomly and unevenly slack after rigging it) but even then...

I am particularly interested in off-centre drop tests because not only do they probe the limits and efficiency of equalization of a particular rig but they are also much more representative of a real fall. Of course a fallen leader will always end up exactly at the vertical of the central biner with a cordelette or any other system that uses a central biner but it's where he starts from that matters and very rarely will he start falling exactly above the central point.

Eagerly awaiting complementary information.

(This post was edited by jktinst on Sep 10, 2011, 5:40 PM)


JimTitt


Sep 11, 2011, 1:23 AM
Post #66 of 88 (2783 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

Well I didn´t really understand your question!
If you test two points vertically above each other you (I got) get a load split of 61%/39% unless you are unbelievably skilfull. This isn´t substantially different from a 60° included angle between the points (64%/36%) or a 90° angle (61%/39% as well) so one can safely ignore the angles except the total load is being increased.

One has to differentiate between static and dynamic equalisation. When you set up a sliding X or whatever it is initially in the static mode where the climber decided where the centre karabiner was going to be. As the load direction is moved offcentre the equalisation stays static until the karabiner slips when we move into dynamic equalisation. For pretty well all the systems I´ve tested the load angle has to be about 13° before the dynamic condition is reached, until then you can just tie a knot in the middle and clip into that and one is not testing the dynamic equalising properties at all.

From your description it looks like the tests were not testing the dynamic equalisation but rather the equalising skills of the tester. The DAV drop test on a sliding X was offset but probably not enough and they got a ratio of 1.3:1.
How good we are at equalising is something else I´ve been checking. From what I´ve seen so far a 2 horizontal point system we can get fairly good, an unequal legged system is poor and a vertical sytem very poor. The 3 piece ones are more difficult!

Your last paragraph touches on one of the problems with restricting testing to one particular case, while most of the time the load may assumed to end up directly below the belay karabiner in other scenarios such as rescue, on traverses or where the rope snags the load may be angled from the vertica- or where the belay builder expectedl. So we are better off testing for the equalising properties of the system rather than how well the system equalised under any particular condition if you see what I mean!

Jim


(This post was edited by JimTitt on Sep 11, 2011, 9:51 AM)


patto


Sep 12, 2011, 3:29 AM
Post #67 of 88 (2727 views)
Shortcut

Registered: Nov 14, 2005
Posts: 1446

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

The irony in all this effort put into micromanagement of anchor design is that multipiece anchors almost always receive significantly lower loads than single piece protection running belays.

The question needs to arise what scenario are you rigging the system for.
*high fall factors
*unexpected load directions
*one piece failing

There is no such thing as a worst case scenario of all aspects. For example you can't have a factor two fall from an unexpected direction.

As far as I am concerned my primary concern is about redundancy and limited/no extension in case of the rare event of piece failure. Additionally preventing a factor 2 fall is VERY important. Clipping the anchor and belaying lower than the anchor will achieve this.


As far as I am concerned two solid pieces is fine for many of my anchors. Tie them back to the belay and get on with the job. Wink


jktinst


Sep 15, 2011, 8:00 PM
Post #68 of 88 (2616 views)
Shortcut

Registered: Jun 29, 2010
Posts: 85

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

The question in my first message was based on my belief that off-centre drop tests inherently yield uneven loading no matter how good the equalization. When your earlier post mentioned that you got very uneven loading from 3-arm systems that are often considered to be equalizing and that the main reason for this was cord or tape friction over biners or rings, I felt that I had to ask whether some of the "unevenness" of the loading might not also be simply due to the "off-centredness" of the test. The fact that you did not understand the question made me wonder if in fact off-centre drop tests really do inherently yield uneven loading as I first thought.

A quick survey of climbing buddies with science and engineering training yielded a not-quite-convincing 2 in favor of "off-centre drop tests will inherently, albeit very temporarily, put a greater load on the anchor that is closest to the direction of the fall" and 1 in favor of "efficient 2-arm equalizing systems equalize the load no matter how off-center the drop" (disregarding friction at this point).

Maybe you (and other engineering-minded contributors) would care to shed some light on my dilemma.

With regards to your latest reply, I understand now that, with an experimental protocol yielding a given distribution split among two unequal arms, the split remains essentially the same regardless of the angle between the arms. Only the overall load increases with the angle. My question was not about this but thank you for the clarification.

For the DAV offset drop test on a sliding X yielding a 1:1.3 equalization ratio that you mentioned, was the offset less than the 13° you consider to be the cut-off for visible biner movement? Was it done on unequal vertical arms? It's interesting that it yielded a result that is essentially identical to the Ewing/Long/Gaines tests on the unequal but centered sliding X. Your 40:60 split means a ratio of 1:1.5 but, without the experimental protocols, comparing your, the DAV and the Ewing/Long/Gaines tests together makes no sense at all.

I was initially frustrated with the Ewing/Long/Gaines tests because I felt that a drop from the exact mid-point of a 2-arm system was unrealistic. However, their results with the unequal systems clearly showed efficient dynamic equalization despite the centered drop. Of course it is not the same as an off-centre drop test but, in both cases, equalization means that the cordage must slide through the biner. This is why I have a hard time with your assertion that, since the biner is not seen to move unless the drop is off-centre by more than 13°, you’re effectively in a static equalization situation and you might as well put a knot in the sling. It seems to me that the Ewing/Long/Gaines tests clearly demonstrated that you don’t need obvious biner movement to have dynamic equalization but that putting a knot in the sling quite effectively turns dynamic equalization into static non-equalization.

(This post was edited by jktinst on Sep 15, 2011, 8:01 PM)


JimTitt


Sep 16, 2011, 12:03 AM
Post #69 of 88 (2593 views)
Shortcut

Registered: Aug 7, 2008
Posts: 953

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

If the system is truly frictionless then an off-centre drop will load the pieces equally since by definition there is no opposing force on the centre point, the load difference must be infinitely small since the moment one force is greater than the other the central point will move and make them equal.
Since however there is friction the pieces will be unequally loaded and as the friction remains then the unequal loading remains as well.

You´ve misread what I wrote, the difference in the DAV test was 1.3:1 which is considerably different to the tests you quoted.

If the centre point karabiner doesn´t move for whatever reason there can be no dynamic equalisation only the original static equalisation. It doeasn´t matter why the karabiner doesn´t move, it can´t tell the difference between a knot and friction and simply remains in the same place.

What you have to remember is that as the load moves off-centre the proportion taken by each leg merely changes until either the karabiner moves or if it is knotted the load on one leg becomes zero.
For an 60° included angle system and a single cord the load proportion changes like this as you offset the load (these are test results so a bit erratic, I´ll work out the theoretical ones sometime):-

0° 50%/50% 1:1
2,5° 48%/52% 1.08:1
5° 44%/56% 1.27:1
7,5° 42%/58% 1.38:1
10° 38%/62% 1.63:1
12,5° 37%/63% 1.7:1
13° Slip
15° 36%/64% 1.78:1

After the slip point the split remains the same in a sliding system but with a fixed karabiner the split continues to increase until 30° when the load on one leg is zero. Obviously the split and the point at which it becomes 0 on one leg varies with the included angle of the pieces with narrower angles naturally being worse. With the pieces vertical it´s not only virtually impossible to equalise by hand but the slightest angular change completely unloads the upper piece immediately whereas a wider angle is much less sensitive for both initial equalising and angular movement, shame it loads the gear more though!

As you can see a degree or two error in the initial equalisation would make a considerable difference (easy for me as I just watch two readouts but still a fiddle, with the knotted set-ups its a real nightmare).

While the offset drop tests are representing one particular scenario which might occur (though selected on no particular basis as far as one can tell) they certainly aren´t testing the dynamic equalisation of the system, one needs to move the load angle a lot more than they did.
If the offset the DAV and the others used is considered the real life maximum then clearly one save the effort of trying to get dynamic equalisation, however since scenarios could (do) exist which can give far greater offset then either one has to improve the equalisation or avoid them or do something different again.

Jim


LostinMaine


Sep 16, 2011, 3:44 AM
Post #70 of 88 (2582 views)
Shortcut

Registered: May 8, 2007
Posts: 539

Re: [JimTitt] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

I'll start off by saying that I am no engineer and that I haven't picked up the Long and Gains anchors book in a while. If I recall, they had a series of box and whisker plots describing a fair amount of variability in some of their tests.

In my world of statistics, it is often the variability, rather than the mean or theoretical "true" value, that is of interest. On the tests that some of you have run, do you also find wide variability in the drop tests with off-center loading? I suspect the loading variability among successive drop tests would be much higher with "equalizing" configurations like the sliding X rather than the fixed powerpoint of something like a cordelette.

If that is the case, I am much more concerned about potential spikes in unequal loading distribution rather than mean values of how anchors handle force distribution the bulk of the time. For example, if the sliding X would somewhat equalize much of the time, but bind unexpectedly on occasion (occasions which may not have been predictable or preventable beforehand), those occasional spikes would be enough for me to choose a static anchor instead.


Partner cracklover


Sep 16, 2011, 8:09 AM
Post #71 of 88 (2569 views)
Shortcut

Registered: Nov 14, 2002
Posts: 9935

Re: [LostinMaine] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

LostinMaine wrote:
I haven't picked up the Long and Gains anchors book in a while.

Um... go do that. It does not say what you think it says.

In reply to:
I suspect the loading variability among successive drop tests would be much higher with "equalizing" configurations like the sliding X rather than the fixed powerpoint of something like a cordelette.

Nope. Look at the chart in the back, and the data series labeled cordelette unequal arms.

GO


LostinMaine


Sep 16, 2011, 5:58 PM
Post #72 of 88 (2540 views)
Shortcut

Registered: May 8, 2007
Posts: 539

Re: [cracklover] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

cracklover wrote:
LostinMaine wrote:
I haven't picked up the Long and Gains anchors book in a while.

Um... go do that. It does not say what you think it says.

Just reread, and it says exactly what I thought it said. To restate it differently, I'm not so much concerned with the load distribution, rather the unpredictable variation that cannot be accounted for in an anchor setup.

cracklover wrote:
In reply to:
I suspect the loading variability among successive drop tests would be much higher with "equalizing" configurations like the sliding X rather than the fixed powerpoint of something like a cordelette.

Nope. Look at the chart in the back, and the data series labeled cordelette unequal arms.

GO

Looking on page 187, under the heading "sliding X unequal length" second paragraph is the statement "In terms of consistency in equalization from test to test, the sliding X unequal length configuration was as consistent in the absolute differences in load generated across repeated falls. Interestingly, though, while it tended to be very consistent across repeated measures, this anchor was also the most likely of all configurations to produce an occasional extreme difference in load. In other words, although tests showed this configuration to be generally consistent, it generated some unpredictably dreadful equalization." This was attributed to what they referred to as the clutch effect and was manually removed in successive tests.

However, the box and whisker plot on page 189 shows that the cordelette unequal had the greatest loading difference between legs and the greatest variability as well. This data set does not even remotely back up the words in the text. "Extreme" difference in load is not at all indicated in the box plot for the sliding X. I wonder if they chose to only report data for the sliding X where Jim Ewing manually separated the strands.

I was simply curious if others had noted this variability in similar tests conducted.


jktinst


Sep 16, 2011, 8:34 PM
Post #73 of 88 (2527 views)
Shortcut

Registered: Jun 29, 2010
Posts: 85

Re: [LostinMaine] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

I did note that and used the term"outlyers" in my synopsis to identify the issue even though I wasn't prepared to make my post much longer by discussing it. The narrow error whiskers on the sliding X unequal suggest that these outlyers were not included in the stats, only discussed in the text. However, those stats must be those of regular sliding X tests because, for the sliding X with clutch effect manually eliminated, the average result is lower (as low as the equalette).

Of course, not having the raw data, it's hard to gauge what an "occasionally extremely high" sliding X unequal result might be compared to the "regularly low" ones and compared to the high and "generally highly variable" cordelette unequal results.

Interesting that you should say that this question mark over the sliding X results might send you running back to a static equalization option when the equalette was clearly shown to have superior equalization with no outlyers.


patto


Sep 16, 2011, 9:36 PM
Post #74 of 88 (2523 views)
Shortcut

Registered: Nov 14, 2005
Posts: 1446

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

OK. We all know the desired solution that people seem to be trying to solve. PERFECT EQUALISATION ALL THE TIME.

However, do we even know what is the problem that is being attempted to be solved?

As far as I am concerned when I am climbing I don't normally face problems that need perfect equalisation or even 'good' equalisation.

It seems to me we are after an unachievable solution to a problem that doesn't even exist.


(This post was edited by patto on Sep 16, 2011, 9:37 PM)


LostinMaine


Sep 17, 2011, 6:31 AM
Post #75 of 88 (2484 views)
Shortcut

Registered: May 8, 2007
Posts: 539

Re: [jktinst] Stupid Simple Elette [In reply to]
Report this Post
Average: avg_1 avg_2 avg_3 avg_4 avg_5 (0 ratings)  
Can't Post

jktinst wrote:
I did note that and used the term"outlyers" in my synopsis to identify the issue even though I wasn't prepared to make my post much longer by discussing it. The narrow error whiskers on the sliding X unequal suggest that these outlyers were not included in the stats, only discussed in the text. However, those stats must be those of regular sliding X tests because, for the sliding X with clutch effect manually eliminated, the average result is lower (as low as the equalette).

Of course, not having the raw data, it's hard to gauge what an "occasionally extremely high" sliding X unequal result might be compared to the "regularly low" ones and compared to the high and "generally highly variable" cordelette unequal results.

Interesting that you should say that this question mark over the sliding X results might send you running back to a static equalization option when the equalette was clearly shown to have superior equalization with no outlyers.

The reason is that I don't care much at all about equalization - especially when it is so variable as to be unpredictable. I would much prefer to identify the strongest pieces in the anchor and force the load onto those, thereby reducing the likelihood that an unpredictable, extremely high load shifted onto a supposedly equalized anchor component that was suspect from the start.

First page Previous page 1 2 3 4 Next page Last page  View All

Forums : Climbing Information : The Lab

 


Search for (options)

Log In:

Username:
Password: Remember me:

Go Register
Go Lost Password?
$13.46 (10% off)
$7.16 (10% off)
$2.66 (10% off)
$16.16 (10% off)



Follow us on Twiter Become a Fan on Facebook