|
brundige
Jan 29, 2012, 6:09 AM
Post #1 of 23
(7170 views)
Shortcut
Registered: Apr 5, 2004
Posts: 305
|
So driving home from the crag today, my friend and I were having difficulty figuring out why shock loading a piece of gear as would happen using say a sliding x anchor is actually bad. It would be great if some of the more physics minded individuals could put some math in your response to justify your answer.
|
|
|
|
|
JimTitt
Jan 29, 2012, 8:23 AM
Post #2 of 23
(7132 views)
Shortcut
Registered: Aug 7, 2008
Posts: 1002
|
Like most things in life it might be bad or it might not.
Various testing shows the increased load from a failed sliding X over a fixed system is around 40% to 200% depending on the exact test. One could organise a plausible scenario which would give even worse results.
This is the force on the centre point of a 3-point anchor when one outer leg fails. The blue line is a sliding X with 16mm nylon tape, the others are fixed systems with various cords.
The graph is a bit small but you´ll get the idea!
(This post was edited by JimTitt on Jan 29, 2012, 11:01 AM)
|
|
Attachments:
|
3pointlegfailurecomp.jpg
(42.0 KB)
|
|
|
|
|
patto
Jan 29, 2012, 10:49 AM
Post #3 of 23
(7052 views)
Shortcut
Registered: Nov 15, 2005
Posts: 1453
|
Experiment:
Trial 1. Gently place a bowling ball on you toe. Observe the force felt on your toe.
Trial 2. Drop a bowling ball on your toe from approximately 4 feet. Observe the force felt on your toe.
Record observations and discuss.
|
|
|
|
|
brundige
Jan 29, 2012, 12:51 PM
Post #4 of 23
(7030 views)
Shortcut
Registered: Apr 5, 2004
Posts: 305
|
Jim thanks for the graph, Is there a study to go along with that graph that i might be able to find online. also what are the X and Y values for this graph. I always use a SRENE Anchor system, but from a physics/mechanical engineering perspective we were unable to account for the increase in force associated with "shockloading" (assuming a dynamic rope) we felt that the rope ought to be absorbing the force and that failure of one piece of gear in a dynamic system would not cause additional force greater than extending the distance of the fall by a few feet. Also does this graph take into consideration direction of force. For example a sliding x has the advantage of more or less equal distribution regardless of direction, where as a cordelette may weight pieces unevenly if the force is applied in any direction other than straight down
(This post was edited by brundige on Jan 29, 2012, 1:15 PM)
|
|
|
|
|
billl7
Jan 29, 2012, 1:26 PM
Post #5 of 23
(7018 views)
Shortcut
Registered: Oct 13, 2005
Posts: 1890
|
JimTitt wrote: Various testing shows the increased load from a failed sliding X over a fixed system is around 40% to 200% depending on the exact test.
Being a lazy person, the above is good enough for me.
And it is another reason to avoid an anchor that can extend very much. The first reason that comes to my mind is the belayer possibly getting catastrophically jacked around during the extension.
Bill L
|
|
|
|
|
JimTitt
Jan 29, 2012, 1:52 PM
Post #6 of 23
(7007 views)
Shortcut
Registered: Aug 7, 2008
Posts: 1002
|
There will be a paper one day when I finish writing it!
X is time and Y is force.
The various tests including mine assume the worst case which is the belayer is included in the system, that is he/she impacts the belay as well. Other systems like belaying off the belay itself will give different results but belaying this way is unusual and awkward for lead belaying.
The sliding X has nothing approaching equal load distribution and unless you have a huge angular discrepancy (which is hard to see could occur under normal circumstances) is just as likely to make things worse, not better. The cordalette incidentally isn´t generally any better at load distribution and often worse so you take your chances either way!
|
|
|
|
|
Urban_Cowboy
Jan 29, 2012, 2:19 PM
Post #7 of 23
(6994 views)
Shortcut
Registered: Feb 3, 2009
Posts: 219
|
patto wrote: Experiment: Trial 1. Gently place a bowling ball on you toe. Observe the force felt on your toe. Trial 2. Drop a bowling ball on your toe from approximately 4 feet. Observe the force felt on your toe. Record observations and discuss.
|
|
|
|
|
sittingduck
Jan 29, 2012, 2:35 PM
Post #8 of 23
(6988 views)
Shortcut
Registered: Aug 19, 2003
Posts: 338
|
Here is what some of the respected climbing literature says about systems with limiter knots to minimize extension and a dynamic lead rope:
Rock Climbing Anchors, a Comprehensive Guide
by Craig Luebben, 2007, page 53:
There is no truly super-elevated "shock-load" when you have a dynamic lead rope in the system."
Climbing Anchors, Second edition
By John Long and Bob Gaines, 2006, page 191:
Here extension simply means that the next anchor that holds will be subjected to the initial load, minus what the first leg/piece absorbed. The load does not increase or multiply between the initial failure of a leg and the resultant loading on the remaining leg(s).
|
|
|
|
|
patto
Jan 29, 2012, 6:53 PM
Post #9 of 23
(6913 views)
Shortcut
Registered: Nov 15, 2005
Posts: 1453
|
brundige wrote: I always use a SRENE Anchor system, but from a physics/mechanical engineering perspective we were unable to account for the increase in force associated with "shockloading" (assuming a dynamic rope) we felt that the rope ought to be absorbing the force and that failure of one piece of gear in a dynamic system would not cause additional force greater than extending the distance of the fall by a few feet.
Completely true if you assume that there is zero mass incorporated in the belay.
But as Jim says. When the belayed is attached to the belay and gets displaced as well the the belayer "shock loads" the system.
|
|
|
|
|
jt512
Jan 29, 2012, 7:04 PM
Post #10 of 23
(6908 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
JimTitt wrote: The various tests including mine assume the worst case which is the belayer is included in the system, that is he/she impacts the belay as well.
In this test was the belayer connected to the anchor with a dynamic rope, or not; and, whichever is the case, do you also have results for the other? And if so, can you explain how the results differ—without giving away the whole paper before it's published?
Jay
(This post was edited by jt512 on Jan 30, 2012, 6:40 AM)
|
|
|
|
|
JimTitt
Jan 29, 2012, 7:18 PM
Post #11 of 23
(6902 views)
Shortcut
Registered: Aug 7, 2008
Posts: 1002
|
Even more of a bummer if your buddy gets to the belay and clips in and then a piece fails or he uses a piece as a Jesus nut and it doesn´t hold or three of you are on the belay when the haul-sack cuts loose or............
Bolts are the answer!
|
|
|
|
|
Sheldon
Jan 29, 2012, 9:06 PM
Post #12 of 23
(6862 views)
Shortcut
Registered: Jun 30, 2011
Posts: 27
|
JimTitt wrote: Bolts are the answer!
16mm twisted bolts up the compressor for 2012, who's with me?
|
|
|
|
|
brundige
Jan 30, 2012, 2:01 AM
Post #14 of 23
(6764 views)
Shortcut
Registered: Apr 5, 2004
Posts: 305
|
JimTitt wrote: There will be a paper one day when I finish writing it! X is time and Y is force.
That's what i was assuming, so based on your test results the sliding x did have less force associated with it except for at its maxima, What im trying to figure out is if the forces at the maxima are greater then to or equal to the distributed load prior to failure. if they are greater. Why? if you have an equation that would be awesome.
|
|
|
|
|
jt512
Jan 30, 2012, 3:00 AM
Post #15 of 23
(6738 views)
Shortcut
Registered: Apr 12, 2001
Posts: 21904
|
brundige wrote: JimTitt wrote: There will be a paper one day when I finish writing it! X is time and Y is force. That's what i was assuming, so based on your test results the sliding x did have less force associated with it except for at its maxima, What im trying to figure out is if the forces at the maxima are greater then to or equal to the distributed load prior to failure. if they are greater. Why? if you have an equation that would be awesome.
I think you're misinterpreting the figure. My interpretation is that the distributed load prior to failure is the more-or-less constant part of the graph, showing that the load was about the same for all non-failing anchor systems. The failure occurs at the little peak toward the left side of the figure. Then the load drops below baseline until the "shock load" occurs at the big peak. The load then oscillates after that.
I could be wrong, though.
Jay
|
|
|
|
|
brundige
Jan 30, 2012, 5:51 AM
Post #16 of 23
(6686 views)
Shortcut
Registered: Apr 5, 2004
Posts: 305
|
In reply to: I think you're misinterpreting the figure. My interpretation is that the distributed load prior to failure is the more-or-less constant part of the graph, showing that the load was about the same for all non-failing anchor systems. The failure occurs at the little peak toward the left side of the figure. Then the load drops below baseline until the "shock load" occurs at the big peak. The load then oscillates after that. I could be wrong, though. Jay
Jay, I think your right, thanks for helping to correct my error. My question then to JimTitt is the same as yours, in what way was the belayer attached to the anchor.
|
|
|
|
|
patto
Jan 30, 2012, 8:03 AM
Post #17 of 23
(6657 views)
Shortcut
Registered: Nov 15, 2005
Posts: 1453
|
brundige wrote: Why? if you have an equation that would be awesome.
Go back to the bowling ball example. Or consider the operation of a seat belt. Continuously restrained objects are going to exert less force that one that is allowed to free fall (accelerate) and then impact.
F=ma, an understanding of the kinematics equations , and a reasonable assumption stopping distance of the shock load is all that you need.
jt512 wrote: In this test was the belayer connected to the anchor with a dynamic rope, or not; and, whichever is the case, do you also have results for the other? And if so, can you explain how the results differ—without giving away the whole paper before it's published?
(This post was edited by patto on Jan 30, 2012, 8:05 AM)
|
|
|
|
|
JimTitt
Jan 30, 2012, 11:43 AM
Post #18 of 23
(6623 views)
Shortcut
Registered: Aug 7, 2008
Posts: 1002
|
The belayer (a 80kg test block) was directly attatched to the centre point of the system and then one point on the side released. I´ve got to find a higher building to start including rope into the system since with the stretch the drop length is a problem!
Anyway including rope isn´t worst case so doesn´t really change things, one can evade the issue of extension by adding 100m of rope below the belay or even going back to the ground. I already helped things by using nylon tape not dyneema which gives considerably higher impacts. Fortunately McKently et al have already done the 3-point X with both materials (nylon and Dyneema/Spectra) and get 27% higher impact for the dyneema so we can eliminate that as a sane choice for belay building straight away.
The only other thing would be to change from tape to a rope based system but realistically who uses a sliding X in climbing rope, normal 8mm cord and 9mm dynamic rope appears to be about the same as tape in the impact tests we´ve done anyway so I stuck with 8mm cord for most tests.
As you surmised the initial constant part of the graph is the equalised load in all cases (which I adjusted to be as near the same as possible) and the release point the initial smaller spike. Then there is the free-fall, impact, rebound and then plenty of bouncing and swinging around.
The tests by McKently also give a similar picture to the graph I posted;- I get an increased in impact in nylon from a cordalette to a sliding X of 236% and he saw worst case 260%, for Dyneema he got 330% increase.
As regards those who want a formula, well in a scenario where you have an effectively infinite number of possibilities for anchor position and anchor leg lengths plus variable force and material extension it is going to be complicated!
And limiter knots? Well if they are close enought together to limit the impact
to an acceptable level (whatever that might be) then the sliding part of a sliding X serves no pupose anyway, depending naturally on the exact configuration of the belay. A vertically orientated system can use a sliding X with very close limiting knots to advantage whereas a horizontal one generally is better off without, and randomly orientated naturally may or may not benefit but that is yet another topic!
Jim
|
|
|
|
|
MomentSurf
Feb 8, 2012, 9:42 PM
Post #19 of 23
(6311 views)
Shortcut
Registered: Nov 25, 2010
Posts: 13
|
Urban_Cowboy wrote: patto wrote: Experiment: Trial 1. Gently place a bowling ball on you toe. Observe the force felt on your toe. Trial 2. Drop a bowling ball on your toe from approximately 4 feet. Observe the force felt on your toe. Record observations and discuss. I don't have a bowling ball available, could someone please complete this experiment and post the results. I'm very courious, so please hurry.
I had one. I tried. It hurt. I learned a lot.
|
|
|
|
|
Rmsyll2
Feb 12, 2012, 3:50 AM
Post #20 of 23
(6179 views)
Shortcut
Registered: Oct 6, 2010
Posts: 266
|
Is "nylon tape" what I was taught to call 1" tubular webbing?
Also, one function of using a Sliding X twist in an Equalette (two strands per side and two stop-knots) is to utilize the stop-knot feature with a single carabiner for the rope. Two carabiners can be used one per strand at the bottom part.
You seem to be making a blanket case against Sliding X. But I don't understand a general case for Long's Sliding X rigging, since the extension is to half the length of the loop, which can be whatever one used as sling and longer being worse.
.
|
|
|
|
|
majid_sabet
Feb 12, 2012, 5:12 AM
Post #21 of 23
(6163 views)
Shortcut
Registered: Dec 13, 2002
Posts: 8390
|
All depends on what type of anchor you got and how its made.
(This post was edited by majid_sabet on Feb 12, 2012, 4:22 PM)
|
|
|
|
|
JimTitt
Feb 12, 2012, 9:17 AM
Post #22 of 23
(6126 views)
Shortcut
Registered: Aug 7, 2008
Posts: 1002
|
As it says, 16mm nylon tape. This is a flat woven tape from Singing Rock, not tubular which is rarely used in Europe these days. From experience 1" tubular won´t give any meaningful benefits except any knots will be even harder to undo.
Adding a twist in a sliding X increases the friction making it even worse at equalising though since we are talking about the difference between terrible and appalling I guess it´s pretty irrelevant which way you do it.
If you think the extension is half the length of the sling (or half the distance between the limiter knots) then you have fallen into the trap of assuming the points are symetrically orientated at right angles to the direction of load. With other arrangements the extension can be the full length of the sling or nothing depending on which part fails. In some circumstances the sliding X is better than a fixed system, sometimes the picture is reversed.
Neither is as desirable as a general principle of building belays as using good pieces and redundancy, the cordalette generally being a complicated and clumsy way of achieving this compared with the traditional use of slings and/or the rope.
|
|
|
|
|
|
|
|
|
