Offset knot for climbing

[Mobius]

I am still interested in finding what I consider to be a good offset knot for climbing.

The EDK simply does not meet certain fundamental requirements: It slips way too easily. That this controlled slippage is known about and accepted as normal by some is not a good reason to place undeserved virtue on the EDK. The argument goes that the EDK has been used for quite some time and has not been attributed to deaths (or a bunch of accidents), hence it is good. This is a subjective argument that is fallacious from my viewpoint. Good knots do not slip. So objectively, there are better knots to find: A knot that does not slip easily is a good criteria for any knot, let alone one you are climbing a mountain with.

The lohf8 version I first proposed here does an 'ok' job from my perspective: It does not slip in low elongation rope up to 300kg and up to 200kg in dynamic high elongation rope. This result is already way better than any of the trials I have read about (or conducted) on the EDK. Still, I do not think it is good enough and I am working on new knots.

One new knot I trialled recently is in the attached image. In 11.2mm dynamic rope it too worked 'ok'. However, about 250kg load it changed shape and slipped a bit, then becoming hard to untie. There are numerous knots to try and several ways to orientate either their tails or the loaded sides of the knot. Subtle dressing changes might dramatically effect different results.

At some stage I will find time to look at Dan Lehman's 9-oh and Rodger Callahan's 'snowdrop' knot (IGKT publication) in more detail. Other knots I have looked at seem too big to be strongly considered.

The image below is rubbishy... new camera on its way 

Cheers,

Ian

[Dan_Lehman]

The EDK simply does not meet certain fundamental requirements: It slips way too easily.
That this controlled slippage is known about and accepted as normal by some is not a good
reason to place undeserved virtue on the EDK. The argument goes that the EDK has been
used for quite some time and has not been attributed to deaths (or a bunch of accidents),
hence it is good. This is a subjective argument that is fallacious from my viewpoint.

Spitting into the wind is one thing,
mischaracterizing arguments is worse.
There is nothing "subjective" about pointing to the
quite objective point that the offset water knot has been
used for ages --in the very materials of issue-- without
failure.  And it's certainly worth pointing that out.

Yes, if one greatly overloads the knot, then it might
"roll"/capsize.  But I don't think that this behavior
is seen in abseiling.  And the simple remedy of tying
a back-up "EDK" snug to the first is one that even
I who found it so *ugly*/crude must admit has strong
factors in its favor --namely : it uses *known knot tying*
(just a repetition), it survives pretty much all ways
of "improper / not-so-well-dressed-&-set tying (vis-a-vis
thick+thin, soft+stiff joined ropes, and it's easy to tie
(esp. in severe conditions)).

Since you have some material in which you found
the popular knot to fail --"to slip easily"--, how about
trying the backed-up structure in that, as a severe
(& unrealistic) test of the knot.  We can maybe at
least advance to having this test case done.

For some, there might be yet the desire to tie knots
of greater material efficiency --of adding just another
full turn (for the offset 9-Oh) or tying off just the
appropriate (from choking turn) tail of the offset water knot--,
but these knots do require that one get certain things
right in orientation & loading whereas the "ugly" one
does not.   It's an old struggle between wanting folks
to know how ...., but wanting to anticipate errors --to
"dummy proof" things.

Good knots do not slip. So objectively, there are better knots to find:

The first is an assertion wanting support,
that latter is a non sequitur.
In fact, the tarbuck knot --which was a variation of the
rolling hitch was devised and advocated for rockclimbing
especially because it could slip : this slippage was supposed
to consume force and reduce peak impact force.  (That
was a theory more than fact.)  And the absence of something
doesn't imply that it must exist.  Ih HMPE cordage, the
venerable bowline slips, but it continues to be widely
used without issue.

At some stage I will find time to look at Dan Lehman's 9-oh and Rodger Callahan's 'snowdrop' knot (IGKT publication) in more detail. Other knots I have looked at seem too big to be strongly considered.

Roger's knot isn't offset, and fails simple evaluation.
Now, the reverse is one that might be interesting
to see perform.


--dl*
====

[Mobius]

The EDK simply does not meet certain fundamental requirements: It slips way too easily.
That this controlled slippage is known about and accepted as normal by some is not a good
reason to place undeserved virtue on the EDK. The argument goes that the EDK has been
used for quite some time and has not been attributed to deaths (or a bunch of accidents),
hence it is good. This is a subjective argument that is fallacious from my viewpoint.

Spitting into the wind is one thing,
mischaracterizing arguments is worse.
There is nothing "subjective" about pointing to the
quite objective point that the offset water knot has been
used for ages --in the very materials of issute-- without
failure.  And it's certainly worth pointing that out.

--dl*
====

I am not mis-characterizing anything.

An objective measure of a knot is 'does it slip' at a certain load or % of mbs or other well defined criteria. Trying to convince anyone that the EDK is objectively good because it has been "used for ages" is plainly wrong. The "without failure" part is without qualification. How is the knot tied? Is it being embellished by a back-up knot? The EDK slips in the materials that it is used in at working loads, that is "a failure" to me. We have seen trials showing us exactly this.

To me, the EDK is a poor knot objectively, no matter how much I hear this "used for ages" argument. If you want to believe the "used for ages" argument I suppose a "granny knot" would be an even better knot to try mountaineering with, since that does not fail in everyday life (we would hear about all the accidents) and has been used by millions a lot longer than the EDK. Very few climbing accidents/deaths from it, so it must be good, right? I make a deliberately stupid argument, however not much more so than the EDK one I keep hearing in my opinion.

I have said this a number of times: Find a better knot. Anyone who wants to help, great. If not, I will find one on my own.

I am working towards a set of criteria I can practically trial offset knots to. Something along the lines of tolerable knot settling/slippage up to 300kg in 10mm high and low elongation ropes. Anyone is welcome to input into this, or suggest a knot to test.

Cheers,

Ian.

[Dan_Lehman]

The EDK simply does not meet certain fundamental requirements: It slips way too easily.
That this controlled slippage is known about and accepted as normal by some is not a good
reason to place undeserved virtue on the EDK. The argument goes that the EDK has been
used for quite some time and has not been attributed to deaths (or a bunch of accidents),
hence it is good. This is a subjective argument that is fallacious from my viewpoint.

Spitting into the wind is one thing,
mischaracterizing arguments is worse.
There is nothing "subjective" about pointing to the
quite objective point that the offset water knot has been
used for ages --in the very materials of issue-- without
failure.  And it's certainly worth pointing that out.

--dl*
====

I am not mis-characterizing anything.

Okay, you don't deny the facts of non-failure in
long-standing practice, but challenge that that
ipso facto is a good basis.
IMO, it certainly IS.  Consider it "trialling" and then
ask if anything else has been so thoroughly tested?!

An objective measure of a knot is 'does it slip' at a certain load or % of mbs or other well defined criteria.

Except that we've seen that even defining "slip"
is a matter of debate --is it really in any way doing
some dangerous "slipping" if it's just adjusting to
forces from a low-force setting?!  When a knot is
taken to break point, one might conclude that
slippage wasn't a problem (YMMV).

Trying to convince anyone that the EDK is objectively good because it has been "used for ages" is plainly wrong. The "without failure" part is without qualification. How is the knot tied? Is it being embellished by a back-up knot? The EDK slips in the materials that it is used in at working loads, that is "a failure" to me. We have seen trials showing us exactly this.

?!  On the contrary, while we might not know in
given detail even on most uses, we can know by
testimony and example how folks tie and use the
knot, and have otherwise to think how thousands
& repeated thousands of application-specific uses
have not brought anyone --even!-- to grief.  Again,
that is a vast amount of "practical testing" to deny!

If you want to believe the "used for ages" argument I suppose a "granny knot" would be an even better knot to try mountaineering with, since that does not fail in everyday life (we would hear about all the accidents) and has been used by millions a lot longer than the EDK. Very few climbing accidents/deaths from it, so it must be good, right? I make a deliberately stupid argument, however not much more so than the EDK one I keep hearing in my opinion.

This defies reason : the granny knot has NOT been used
AT ALL in the materials or application of rockclimbing, point
blank obvious, and in stark contrast to the offset water knot
which has long-standing usage.

(Now, as an aside, the "longstanding usage ..." reasoning must
be chary of all assumptions/contexts :: bring a new material
such as HMPE into the situation, and you might have other
than the longstanding happy behavior!  One had a sort of this
realization in seeing the ca. 2001 HSE (UK) study of using
the clove hitch for termination, where of their several tested
ropes, only the dynamic one held (IIRC), all of the tested
low-elongation ropes slipping at widely different loads ::
one could see a rockclimber who had confidence from
longstanding use of the knot in climbing ropes finding
unpleasant behavior in more inelastic stuff.
But you don't have any experience of granny knots climbing.)

I have said this a number of times: Find a better knot. Anyone who wants to help, great. If not, I will find one on my own.

And so you have --and have already done some testing of just
such a Found-Better Knot--, viz., the offset 9-Oh, which takes
the design goal of being offset (for smooth traverse of rough
services including edges), efficient tying and consumption
of material ("small footprint" to cite a term), and suitability
or joining dissimilar (within application bounds (6mm + 11mm?))
ropes, and began with the offsetting of the S.Parts and then the
sure nipping of the choke point with a full turn by one
of the two ropes --spec'd the thinner / more flexible one,
and thus the accommodation of dissimilar ropes.

And it works, as you've shown.

Even *hastening* the tying above, to yield the offset 8-Oh,
one gets a half-turn less choking but now has the new
(vs. EDK) aspect of tails going out of the nipping area
in opposite directions, so that there is an end on either side,
not both on one and connected parts on the opposite side.
That looks promising, but somehow it has seemed to me
that its resistance to capsizing looks less good, and with
the 9-Oh looking good and just a half-turn away,
I would favor that.

BUT, unlike the "EDK-backed EDK", both require that one
orient the thick-2-thin tying in just such a way, for ideal
and most sure results, at least --perhaps in reasonable
counter orientations per the application (e.g., 6mm tied
with 10mm), they still work, albeit with less of a margin
of surety!?

Ditto for simply tying off the offset water knot's choking-part
tail (which again should be the thinner/flexible rope if uneven) :
it works well in the specified orientation, but does require
that one note and achieve such orientation,
and should offer less security if botched in this regard.
IMO, that clumsy EDK-backed EDK might suffer not only
botched & varied per-knot orientation (e.g., thickier rope
choking) but also spaced back-up (which could enable
a "roll" of the primary knot before hitting the stopper)!?


--dl*
====

[Mobius]

I have not forgotten this thread, indeed the last 3 months have been full of trials involving knots tied in 10-11mm dynamic rope taken to at least 3 kN if they do not jam prior to that. I have found two knots I like, and one in particular. I believe 3 kN is significant in this application and is perhaps close to twice the tension needed after looking at force fall factor (FF) considerations I have read about. Others might confirm this.

My trial requirements were simple. Take a knot in 10-11mm dynamic climbing rope to 3 kN and see whether it slips and/or jams. I simply do not believe ''all knots slip" as I was told by an 'expert'. I do not (and never will) believe that #1410 is a good knot for this application. The EDK is a knot that has a donkey beside it in Ashley's book for a good reason in my opinion.

For the moment, I would kindly ask Dan Lehman to check the image below to see whether I tied the 9-oh to his satisfaction.

Cheers,

Ian.

[Mobius]

I have done a lot of  trials in various rope materials with the #1410. In particular here are some images of the knot tied in worn 11.2mm dynamic climbing rope. I have trialed the knot in both new 10.2mm dynamic rope and 10.2mm kernmantle climbing rope as well.

The first image is simply the trialed knot. The tails were carefully set to 110mm each in length. I believe the knot was snugged appropriately. The legs were orientated as shown.

Image two has the knot at a modest 0.5 kN and the tails are now at 95mm.

Image three is at 1.5 kN and we now see the tails are at 85mm.

The last image is at 2 kN and the tails are now at 80mm.

This knot stopped slipping somewhere around the 2 kN mark and was jammed at 3 kN having first slipped a total of 30mm.

Other trials in newer dynamic rope materials suggested to me that if you were climbing with new rope you could probably expect less slippage, however when the knot stops slipping it starts jamming, hardly a bargain either way the way I see it.

Cheers,

Ian.

[Mobius]

Despite my misgivings about the #1410 from very early on, I have trialed it in different ways. The images that follow actually proved very useful to me in that it showed me that orientation changes to a knot (perhaps, particularly to an offset one) can make a quite a difference.

Image one is a #1410 with legs reversed, set carefully with tails at 100mm.

Image two shows how the knot slipped 10mm at 0.5 kN, and turned itself around to leave the legs in a position like the trial above. Will that always happen (rhetorical)?

There is not much difference in terms of mm slipped than the results in the previous post at 0.5 kN or later levels of tension, however it became clear to me that offset knots need to be trialed in both ways in terms of leg orientation. More on this later.

Cheers,

Ian.

[Mobius]

One knot I said I would trial earlier in this thread was "The Double Snowdrop Bend" as given in the September 2015 edition of KNOTTING MATTERS. Mr Roger Callan's misgivings about the EDK are perhaps even stronger than mine, that you could see for yourself if you get a chance to read his article and viewpoint.

Mr Callan's Snowdrop bend was immediately interesting to me, since he too had combined a simple overhand and figure 8 into a form that showed promise in being one that would not "unravel" (his term).

I have trialed his knot in 11.2mm dynamic climbing rope. I have not done much with this knot to date, here is what I found so far.

Image 1: the tying diagram (borrowed)

Image 2: the actual knot I trialed

Image 3: the knot at 2.1 kN. At this point I could see the tails being 'swallowed' and suspected jamming was not too far away. The actual knot jammed around 2.6 kN.

I can come back to trialing this knot if there is particular interest.

Cheers,

Ian.

PS. Roger is spelled with a d in the main article heading (Rodger), I think that is wrong judging from the signed diagrams

[Dan_Lehman]

One knot I said I would trial earlier in this thread was "The Double Snowdrop Bend" as given in the September 2015 edition of KNOTTING MATTERS. Mr Roger Callan's misgivings about the EDK are perhaps even stronger than mine, that you could see for yourself if you get a chance to read his article and viewpoint.

And sadly he shows little of comprehension and fair
consideration to the widespread usage of the knot
he faults, which greatly weakens his position.  (But
I cannot so easily dismiss his remarks about seeing
the knot in "climbing gyms," and yet cannot figure
out why this would be : climbers use single ropes
run floor-to-ceiling-area anchor sheave down to
climber (to my on-site awareness in USA), and no
end-2-end knots are required.  Climbers tie in,
belayers might clip in to floor-anchors if at all.
So that part of his statement puzzles me greatly!

His proposed solution only further exhibits his lack
of understanding --and you, who better understand,
thus took it backwards, where it makes more sense!

I have trialed his knot in 11.2mm dynamic climbing rope.

No, and I think I remarked about this point above:
you have tested the reverse of his knot.
HE shows SParts as the arrowed ends,
and his ends-adjacent side is thus the tails
(though you are right to see this as being
 like-other-*offset*-knots paired SParts;
 and *I* see it so because in my long-ago use
 of arrows/non I did the opposite to Roger,
  *pointing* the flow into & out of a knot body thus!)

PS. Roger is spelled with a d in the main article heading (Rodger), I think that is wrong judging from the signed diagrams

Yes, "spare the 'Rod' !" was my joke at this.
Roger took issue with it in the follow-up issue of KM.
But "D" is the highest grade we can give for his knot
as a solution.

--dl*
====

[Dan_Lehman]

Despite my misgivings about the #1410 from very early on, I have trialed it in different ways. The images that follow actually proved very useful to me in that it showed me that orientation changes to a knot (perhaps, particularly to an offset one) can make a quite a difference.

Image one is a #1410 with legs reversed, set carefully with tails at 100mm.

I don't see this as well set for sustaining this reversed
(reverse extreme : arc & loop <=> loop & arc)
orientation; one really needs to set it tight pulling
SParts together opposing knot qua stopper,
AND THEN
pulling SParts apart as desired and getting it to hold
the orientation.
That said, I think that one extreme --likely this one,
I'm not going to check now-- is less sure than the
other; I think your dual-orientation testing has that.

But I see more that you've turned --in this latter and
less stable case of arc-&-loop-- not to the other extreme
(which takes quite a swing!) but to the approximate
mid-range state (half-arc & half-arc),
WHICH
I think might be the LEAST secure/stable state.

In the extremes (arc-&-loop),
the tails exit roughly perpendicular to the axis of tension;
in the mid-range state,
 tails lie parallel.  NB!
(In the Autralian BushWalkers testing, they recommended
orienting the tails in a parallel way (they, thinking of which
way up or down they should point, missing these determining
distinctions of orientation being described & explored here.)

Will that always happen (rhetorical)?

To the non-rhetorical question, as noted above, I do believe
that the case of arc-&-loop where the loop is more *inner*
will tend to rotate some measure away from the extreme,
whereas the other extreme orientation is better sustained.

To the rhetorical question, GOOD POINT TO HOLD ::
that in any given testing that might be presented,
there is good reason do question how applicable the
supposed demonstrated behavior is to different combinations
of materials & forces & ... so on !!
I.p., in the mountain-project thread about the recent (May)
rappelling fatality, one poster gives a URLink to testing of
a WET offset water knot / EDK in which one sees the knot
roll & roll & continue until I think pulling out, with quite
an explosion of losing grip --and HUGE forces, vis-a-vis abseil
(maybe 3_000 pounds force at START of rolling, in old
climbing rope(s)).  Coming up after viewing this (Youtube?)
on a small page of other-videos-like-what-you-saw is one
that shows "EDK & EDK-8" testing.  Contrary the popular
viewpoint of those-in-the-know (i.e., that the latter knot
is LESS stable, more vulnerable to catastrophic flyping),
the EDK-8 sustains a well higher force; both knots hold
until rupture, IIRC --just viewed it once, so we need to
double-check this.  In any case, THIS particular presentation
would move one if anything to use what I think is better
understood as a more dangerous knot.  (E.g., those Bush-
walkers claim to have made if fail with just manual loading.)
To this all, we should note that what might be
common across inconsistent presentations
is NAME ONLY, less so the exact knot geometry!

however it became clear to me that offset knots need to be trialed in both ways in terms of leg orientation. More on this later.

Spot on.
AND in combination of materials,
thick & thin joined; new slick & old rough.
Preferably ELASTIC, to emphasize potential
deformation (but, no, not so much shock/bungee cord
--that's WAY elastic).

Cheers, Ian.

Close,
but first letter is better "D", gotten from 'RoDger',
which makes 'Dan' 

--dl*
====

[Dan_Lehman]

I have found two knots I like, and one in particular. I believe 3 kN is significant in this application and is perhaps close to twice the tension needed after looking at force factor (FF) considerations I have read about. Others might confirm thi

Or challenge it.
Expected abseil forces are WAY less than this:
your 1kn is already around double what ONE ROPE
(& hence, the knot) would get in the joined-ropes abseil!

Going beyond this is reasonable to try to find
any lurking issues, which might be tickled upon
some slip and more-dynamic loading.

Your "FF" ("fall factor") is something conceived of
to explain falls while climbing, and the UIAA drop
test uses one a little under 2, which could occur
if a climber falls before putting in any protection,
and hence falls the length of rope out to pass
the belay point and then of course that length
of now-slack rope until it takes hold --TWO times
length.  A rappeller would need to have some bit
of rope slack out at the start and then just fall
on that, to get a say 2m fall on 2m (doubled!)
rope and FF-1/2 on the knot.  And that's bizarre.

I do not (and never will) believe that #1410 is a good knot for this application.

I'm sort of okay with your not favoring it,
but I continue to view with dismay your
lack of appreciation that ACTUAL ON-SITE USAGE
has effectively *trialed* this dubious knot FAR
more times, in DECADES/thousands of uses
beyond what you have done.
AND NB :: your own testing show here could well be used
by advocates of the offset water knot / EDK as proof of its
suitability !!

So, it's fine that you seek something surer;
it's worrying that you dismiss the extensive
practice so fully.

BTW, I really do wish that you'd test the knot tied
with itself as back-up, pretty snug together
in that small cord in which it failed SO quickly !!
I want to see to what extent in that challenging
medium the backed-up structure can endure.

The EDK is a knot that has a donkey beside it in Ashley's book for a good reason in my opinion.

Ashley's reason, we should note, is re strength,
which is irrelevant to the abseiling application;
re security, he only remarks that it's better than
the (commonly used) sheet bend --though we might
question his basis for both judgements (I think that
the latter was based on his testing rig's check for
mohair(-like) fibre cord).

For the moment, I would kindly ask Dan Lehman to check the image below to see whether I tied the 9-oh to his satisfaction.

... instead of making all his other comments ... !   
Yes, that looks fine.
It is the full wrap made by the choking strand
that is believed to give the knot good resistance
to being pried open and then flyping.

Cheers,
I/Dan

[Mobius]

I have trialed his knot in 11.2mm dynamic climbing rope.

No, and I think I remarked about this point above:
you have tested the reverse of his knot.
HE shows SParts as the arrowed ends,
and his ends-adjacent side is thus the tails
(though you are right to see this as being
 like-other-*offset*-knots paired SParts;
 and *I* see it so because in my long-ago use
 of arrows/non I did the opposite to Roger,
  *pointing* the flow into & out of a knot body thus!)

--dl*
====

I have to laugh at myself for this one. I saw the arrows and fleetingly thought it a little strange to use arrows for the tails, however, reversed, the paired SParts were exactly the offset bend I was expecting.

Mr Callan's knot as intended is not offset so I won't be testing it in it's proper form. It did not do too bad a job in the reversed way I tried it, however I believe I know of better knot alternatives that I will share with you.

I'm debating whether to show you what I have now or get more material for testing. I have exhausted the supply of used indoor climbing rope I was given and my 'new' purchased rope has now seen a few tests. It doesn't look too bad, however one wonders how long new stays 'new' when I am trialling the same 2m sections of rope to 3 kN repeatedly.

I chose the 3 kN mark knowing (at least for this application) that it was quite a high mark to aim for. As Dan says here:

Going beyond this is reasonable to try to find any lurking issues, which might be tickled upon some slip and more-dynamic loading

This is part of the reason I choose higher than normal testing limits. Honestly, who is going to listen to a mathematician with a rig and a bent for knots telling you the knot he came up with slips 10% less at 1 kN than some other knot... whoop-de-doo-dah. On the other hand, I find a knot that does not slip at all up to 6 kN and maybe someone looks at that and says "this maths geek who has never climbed a mountain" might actually be on to something.

On top of that, doing trials for this site is no bargain, so I do them as well as I can.

Eg. Someone will agree with you in one breath and tell you something close to 'you are tying in noodle, or grandma's thread' when the trial result does not suit them. This last was just vitriloic nonsense, however it is the comments that follow from someone I respect that nearly made me give up trialing altogether. One was this: 'the slow pull tests have been done and done and I have a strong foundation of what works', followed later by 'I know these selected knots already pass because I have and continue to use them'.

So the first tells me that it doesn't matter what I come up with at least some members are unlikely to take much notice, they have made up their mind already. The second tells me that whatever knot someone uses is a 'pass' as long as it does not fail for them. Threre is no point in me finding any new, objectively better performing knot(s) if the mind of one's audience is closed. There are plenty of agenda's to overcome on this site. The last agenda I questioned here earned me a bunch of red writing and a tag of 'unprofessional'. Thank goodness my profession is not tying knots for a living

I do not believe slow pull tests have been done to death. There is so much to learn from watching knots unravel, twist, collapse, stretch, bulge, jam or hold firm on a rig. Here is a conjecture for you: I get to see what happens to a knot that might experience a sudden jolt, except in slow motion. Some who do not like slow pulls do uncontrolled fast pulls. Before my rig, I nearly pulled my front garden palm tree out with my 4 wheel drive doing a fast pull. What did I learn about the knot?... almost nothing. That is not to say fast pull tests are unimportant, there just has to be a controlled way of doing them to make them worthwhile.

I believe knot testing is in it's infancy and the backbone of it for present has to be slow pull tests. Later on working on a consistent way to do fast pull tests (without a laboratory) would be great. Do we expect a high correlation between slow and fast pull testing(?), well I do, however let us see.

So, although my attitude to the EDK may seem overly harsh, I am not immune to the statistical argument that Dan says I am not paying enough attention too. It is just that I cannot test the statistics, however I can look at the knot in terms of objective rig testing qualities, and try to do a better knot. That is what I have done.

Cheers,

Ian.

[edit: small changes so it reads better]

[Dan_Lehman]

You got a picture's worth of words there, and I read every
one of them --that alone separates you from the variable
X (or was that a constant  ) !

And not one word was about using that super-challenging
core in which your favorite villain of a knot fails strikingly
quickly (where in some other materials it yields only grudingly
or holds to break) with the backed-up variation.  This would
be --i.e., testing the EDK-backed EDK in that material--
going beyond expected conditions to find lurking issues!
We will be most grateful for this trialling --if the knot holds,
rising to such challenge!  As there is a LOT of reason to have
this compound knot in the toolbox : climbers can be in situations
entirely unwelcoming for doing math or much of any thinking
and matching careful ropes-threading, and have a recourse
that is doing something most simple times two (or more!)
is a significant asset (and noting as I did previously that the
compound knot I think allows a lot of the recommended
tying/orienting (dressing & setting) choices to go wrong
and STILL work.

doing trials for this site is no bargain, so I do them as well as I can

And we're grateful for them,
but your notion of "well" is also what makes
them --in terms of material expense, at least--
no bargain.  Tested to much lesser forces, which
STILL leaves them going well greater than expected
abseil forces (which let us stipulate would be about
0.5kN (on the single knotted half of twin lines abseil,
1kN the total load per anchor measurement)),
would let the ropes work longer for you.
--a material savings, but not time & effort ... .

You could regard this as some sort of "basic safety level"
and have in mind more severe cases beyond, to do at
some point ... .  The 1kN loading could sort out knots
to go further with, for your purposes; while qualifying
them for others'.  YMMV.  And, as you noted, there is
that aspect of how-SParts-depart-from-knot which isn't
on anyonElse's radar, even (but ours) !!  Which adds
one more factor to consider (and one more reason
to hope that the compound knot aforementioned
can endure any orientation!).

Here's some info from www.geir.com/mythbuster.html
regarding abseil forces --with humans, not elephants
(who take the long way down  )

It is difficult to produce loads of even 600 pounds on a rappel.
Jeff Fassett and I measured the force produced on an anchor
during a double-rope rappel using a dynamometer. Not surprisingly,
during normal rappelling the force on the anchor was simply body weight
(in our measurements, 150 pounds).  In this case, the load on the
[offset water knot is 75 pounds.  Even with rigorous 5-foot deadfalls
close to the anchor, we were unable to produce a load greater than 600
pounds on the anchor. In this case, the load on the [knot] is only 300 pounds.
(These tests were conducted with a heavily-used 10.2 mm Beal dynamic line
and a 150 pound person.)

Now, something else :

what happens to a knot that might experience a sudden jolt, except in slow motion.

You might be misleading yourself, here :
i.e., what happens in slow(-pull) motion does NOT
model dynamic loading, which doesn't let up on the
knot's rolling (say) but sustains its force --there is
no lag time of a slow-pull device catching up to
re-tension the line!  --there might be even a little
shock load?!  (I've often wondered at this.)  A suspended
mass could be accelerated a little by a capsizing then its
resultant load would not only come more quickly but be
greater for this acceleration, unlike a test machine's slow
take-up of capsized slack/reduced tension!?

On the other hand, I find a knot that does not slip at all up to 6 kN and maybe someone looks at that and says "this maths geek who has never climbed a mountain" might actually be on to something.

Except that you first have to convince this person
that the used-by-everbody-to-do-what-is-needed knot
"slips" and that this slipping is somehow of concern,
even though the knot has been used for 50 years
or more w/o problem!  (And that doing some clever
reaving just so in order to get a non-slipping knot
is a winning proposition --the learning curve, the
remembering-when-in-duress odds ... (though for
half a century the rumored slippage hasn't troubled us).)

I have made that pursuit,
and the offset 9-Oh and some other knots resulted
from the exploration,
along with the tie-the-choking-tail-of-EDK securing,
and orienting mis-matched lines to choke with thinner,
and ...
I must admit that each such *instruction*/requirement
amounts to a chance to go wrong!  But I'd likely use one
of these more-secure knots.

Still, having that super-safe/-resistant compound knot
in the toolbox adds assurance.

Cheers,
--dl*
====

[Mobius]

Thank you for your feedback as always Dan.

And not one word was about using that super-challenging
core in which your favorite villain of a knot fails strikingly
quickly...

I am going to retrial the #1410 and a back-up to it tomorrow in my 'challenging' material. I think I know the answer to what worries Dan already and it is to do with Ashley's diagram (reproduced below). The way it is shown is the way I tied it and trialed the #1410 initially. Later all my trials were in climbing rope and tied without the funny twist Ashley shows.

When I first trialed the #1410, tying it the way Ashley shows it, did not seem like a bad idea. I have to look again at my original post(s) on the #1410 results, however the amount of knot movement I saw then I think is out of keeping with what I have seen in low-elongation kernmantle later on. Huge movement is not typical from my experience with the EDK. Dynamic rope is a different story, you see more movement.

what happens to a knot that might experience a sudden jolt, except in slow motion.

You might be misleading yourself, here :

Possibly, it was a conjecture. However some behaviour elements might be in common. It sure would be nice to see some slow motion photography of a knot I have studied under sudden load. For example, is it even clear that a knot behaves worse (the usual assumption) under a sudden load? Some might behave better!

Cheers,

Ian.

[Mobius]

First image: The top knot is Ashley's. If the #1410 is tied like this when climbing... good luck At 30 Kg in this material (mbs ~ 2 kN) the knot had rolled and collapsed with a slippage of 15mm. 15mm when we have a 3mm diameter is hard to ignore, it is a lot. The words 'rolled' and 'collapsed' are deliberate, that is what I saw.

The bottom knot is the flat version. I tied all my climbing rope trials like this (see other images for yourself). This time the #1410 has slipped about 2-3mm at 30kg. Scale things up and think kernmantle 10-11mm and 3mm movement turns into ~10mm with larger diameter rope and equivalent load in mbs terms. That is not far from the results I actually saw.

How close my trial material (3mm polyester braid, ~10% elongation at 50% mbs) replicates (10-11mm kernmantle, ~7% elongation) is not the point here.

The issue was tying the #1410 in the wrong way initially. Rest assured, any climbing rope trials I have done have been the flat version, not with Ashley's twist.

Cheers,

Ian.