Offset knot for climbing

[agent_smith]

Hi Mobius,

I would like to include your creations in my 'Analysis of abseil rope joining knots' paper.

What I need most is the following data:

[ ] load threshold at which instability is triggered
[ ] load threshold where jamming occurs
[ ] force required to initiate translation of offset knot around 90 degree edge
[ ] tail draw-in at various load milestones (Note avoidance of the term 'slippage'. What we are observing is changes in the tail length as the knot compresses under load).

NOTES:
1. instability is defined as: A dynamic event where there is a sudden change in the knot structure (eg capsizing) - compression is not instability and neither is tail draw-in.
2. rope diameters tested must fall within the 8.5mm - 11.0mm range.
3. data obtained for both low stretch and dynamic ropes - must be identical diameters so comparisons can be drawn.
4. unequal rope diameters should also be tested - eg 9.0mm joined to 10.2mm - the variance should be more than 1.0mm.
5. wet rope test (it is assumed that all testing is with dry ropes) - it will be useful to look at data obtained from wet ropes (soak rope specimens in water for 1 hour to condition the ropes).
6. a sample set of 5 tests should be conducted for each test objective - this will enable some statistical analysis of results.

If you can obtain these data points, it will be greatly appreciated.

I would suggest that #1410 be used as a control.


Mark Gommers

[Dan_Lehman]

What I need most is the following data:

[ ] load threshold at which instability is triggered
[ ] load threshold where jamming occurs
[ ] force required to initiate translation of offset knot around 90 degree edge
[ ] tail draw-in at various load milestones (Note avoidance of the term 'slippage'.
 What we are observing is changes in the tail length as the knot compresses under load).

We'd be making far better progress were at least
an initial battery of tests done purely to see how
the various knots in various materials behaved at
forces that correspond to expected abseil forces
(either in direct materials expected to see such use,
or in scaled forces for other materials, based on some
kind of reasonable scaling estimation).

Notably, such testing would not be sacrificing test cordage
because of excessive loading!!!  It is much more helpful
to see some several tests of material & knot (put into
various orientations) at expected loading (or some small
excess of this) than to have break/jamming tests coming
from loading that would never be matched in application!

(I need to think up some easy way of applying and
removing such reasonable loading --done perhaps with
decent 2:1 pulley and approx. 100# load (which load
could be simply bumped by application of manual force
or attachment of a 2nd harnessed weight; or something
I stand in for initial load, and then can pre-load w/weight
before standing in it again).
)

Re the reddened assertion,]
no, I think that what Mobius has observed isn't compression
but in fact slippage as the prying-open forces rise (WAY past
expected forces!) and so pull out more tail(s) --at least in
some cases.  Compression doesn't suck tail in, just feeds
material out to loaded parts.
But it's worth noting that the slippage comes only with an
increase of load, not just given a loaded state and then
holding the load (suspended weight, say) and having some
creeping slippage.


--dl*
====

[agent_smith]

We'd be making far better progress were at least
an initial battery of tests done purely to see how
the various knots in various materials behaved at
forces that correspond to expected abseil forces

Yes - and one might also comment that we need meaningful data points - that can be replicated/reproduced by other testers. To me, this has been one of the biggest issues with historical testing. It is difficult if not impossible for anyone to try to reproduce others results (and we may well ask why...?). Part of the underlying reason (I think) is the fixation on pure 'MBS yield' break tests instead of looking at the salient features I have advanced.

The data points I am requesting are easy to understand milestones.
[ ] like what is the load threshold where instability is triggered?

Notably, such testing would not be sacrificing test cordage
because of excessive loading!!

We are talking about small quantities of cordage - not whole lengths of rope.
It is possible to work with 1.0m test articles - by simply uniting both ends with an offset knot - thus forming a circle/loop (a side-effect of testing a loop is that the forces must by definition be doubled - and so the test bed must be capable of achieving higher loads).

Compression doesn't suck tail in, just feeds
material out to loaded parts.

And yet, testing has demonstrated that tails do shrink in length as load is increased. I have documented this before - and I haven't seen a reversal of this - that is, I have never seen tails grow in length as load is applied. The reduction in tail length that I have observed has not been 'slippage' - it certainly was not a case of any sudden movements. The tail draw-in was always in a progressive manner in response to compression caused by increasing load.

As for jamming - this is a useful data point - and one could argue that it is useful to know for all knots.
I have thus far found that in the 9.0mm class of ropes, jamming appears to occur once the load reaches 3kN (approx 300 kg).
Once jamming occurs - there is no need to continue applying load - testing can cease at that point.
I define 'jamming' as a state where it is not possible to untie the knot by hand (without the use of any tools).
If a knot can be untied by hand - it is not 'jammed'.

Another point that I might make is that I have not been talking about loads to probe the MBS yield point. This is what virtually all other testers have focused their efforts on (which is pointless and proves nothing).
What I am requesting is more useful and meaningful.
[ ] for example - what is the force required to initiate translation around a 90 degree edge?
[ ] at what load is instability triggered?
[ ] at what load does the knot become jammed?

These are data points that others could stand a reasonable chance of reproducing in future tests.

Mark G

PS I can foresee a situation brewing where we get caught up in an endless debate about what to test - and the wheels just keep spinning and spinning in the dust...

[Dan_Lehman]

Yes - and one might also comment that we need meaningful data points - that can be replicated/reproduced by other testers. To me, this has been one of the biggest issues with historical testing. It is difficult if not impossible for anyone to try to reproduce others' results (and we may well ask why...?). Part of the underlying reason (I think) is the fixation on pure 'MBS yield' break tests instead of looking at the salient features I have advanced.

Given the many variables in such testing --of materials,
their orientation in the knot, the knot's orientation
(as has been previously discussed), the application
of force (which at its easiest might not be the best
for modeling actual behavior)--
it's not surprising that replication is difficult.
But before concluding that outright,
yes, it would be good were the many factors at least
recognized and reported.

Notably, such testing would not be sacrificing test cordage
because of excessive loading!!

We are talking about small quantities of cordage - not whole lengths of rope.

And yet some are talking about multiple test
repeats for the sake of Sacred Statistics, and
THAT even for a fixed-set-of-variable-factors!!
.:. Given all that, one is talking about hundreds
of these "small quantities" --not small at all,
and not small of required test effort!

Compression doesn't suck tail in, just feeds
material out to loaded parts.

And yet, testing has demonstrated that tails do shrink in length as load is increased. I have documented this before --and I haven't seen a reversal of this-- that is, I have never seen tails grow in length as load is applied. The reduction in tail length that I have observed has not been 'slippage' --it certainly was not a case of any sudden movements. The tail draw-in was always in a progressive manner in response to compression caused by increasing load.

"Not been slippage" :: however do you come to this
assertion?  How could compression possibly bring
material INTO the compressed object?!  It is necessary
in fact that material goes OUT of the knot, as compression
stretches & shrinks-in-diameter material (exp. such as is
used for abseiling; less so for hi-mod, static stuff).

As for jamming - this is a useful data point

FIRST, let's clear knots for expected loads PLUS
some bump for "just-in-case" excess; in short,
for actual USE,
not academic aspects to fill some table.

JUST to do this --for a variety of materials combinations
and knot orientations and loading variance-- will take a
LOT of effort; it need not take so much of materials.

--dl*
====

[agent_smith]

"Not been slippage" :: however do you come to this
assertion?  How could compression possibly bring
material INTO the compressed object?!

By observation Dan.

Here is data from a test:

Test date:    July 2011
Knot: #1410 with 1 additional binding turn (refer PACI knot study guide)
Tails initially set at 100.0 mm length.
Measured by digital load cell (dynafor 5 ton)
[ ] EN 892 dynamic ropes (9.1mm Beal Joker joined to 9.0mm diameter Edelrid rope)

Static load test:
[ ] at 0.0kN = 100.0mm tails
[ ] at 0.5kN = 90.0mm tails
[ ] at 1.0kN = 85.0mm tails
[ ] at 1.5kN = 85.0mm tails
[ ] at 2.0kN = 80.0mm tails
Test stopped at 2.0kN load.
Dynamic test drop also performed with 80kg mass + 1.0m free-fall.

The tails started at 100.0mm and ended at 80.0mm length.
That's a 20% reduction in length.
The tails did not 'slip' - as in, there were no sudden movements that one could discern as being slippage. Rather, it was a gradual and progressive response to increasing load.

How do you account for this result Dan?

Mark G

[Mobius]

 What we are observing is changes in the tail length as the knot compresses under load).

Re the reddened assertion,]
no, I think that what Mobius has observed isn't compression
but in fact slippage as the prying-open forces rise (WAY past
expected forces!) and so pull out more tail(s) --at least in
some cases.  Compression doesn't suck tail in, just feeds
material out to loaded parts.
But it's worth noting that the slippage comes only with an
increase of load, not just given a loaded state and then
holding the load (suspended weight, say) and having some
creeping slippage.


--dl*
====

I agree with Dan here. What movement I see happening in my trials I largely attribute to slippage. If you increase the load the knot can slip some more, or jam. That later movement observed from increased load has little to do with compression in my opinion.

If compression is the culprit why does a knot like I propose in this application show little to no movement of the tails even at 3kN and beyond? I have images showing my knots under load and compressed, yet the tails have not apparently moved, this is what I strove to achieve in all the testing I did of this knot style.

This is also a point I tried to make a long time ago: I look at some test results and see slippage. When I look at a knot that does not slip I see results where the tails have hardly moved at all under increasing load.

Cheers,

Ian.

[agent_smith]

In puzzlement over test results and ones observations... I am thinking that the English word 'slip' or 'slippage' is being used in a way that belies what is actually taking place.

Oxford dictionary online:  Examples...
[with adverbial of direction] (Of an object) accidentally slide or move out of position or from someone's grasp
* The envelope slipped through Luke's fingers
* A wisp of hair had slipped down over her face
More example sentences Synonyms
* Fail to grip or make proper contact with a surface: the front wheels began to slip
* (as adjective slipping) a badly slipping clutch
* [with object] Escape or get loose from
* (a means of restraint): The giant balloon slipped its moorings.
* An act of sliding unintentionally for a short distance: a single slip could send them plummeting down the mountainside.
*[mass noun] Relative movement of an object or surface and a solid surface in contact with it.

I am unclear whether Mobius and Dan have actually taken notice of the test results I had posted?
I measured a 20% reduction in the overall tail length at a load milestone of 2.0kN. At each load milestone - the load was held for a short interval. During these holds, there was no observed 'slippage' of any tail segments.
It is important to note that this 20% reduction did not occur suddenly or within a very short time interval.
The shrinkage of the tails occurred progressively (ie gradually) in direct proportion to increasing load.

The English word 'slip' or 'slippage' is not an appropriate descriptor in my opinion.

I would also comment that in all the testing I have done, I have never seen tails increase in length as load is applied.
I have however, always seen tails decrease in length as load is increased.
I have also observed that as load increases, the knot core compresses.

I would also comment that if one was attempting to cast #1410 in a negative light - then such a person might be persuaded to use the term 'slip' to promote such a purpose. The average reader (layperson) might also be drawn to, and fixated on the word 'slip' - and therefore also form a negative view. Furthermore, the word 'slip' conveys feelings of danger - and since climbers/canyoners would be placing absolute trust in #1410 (or some other joining knot) - the word 'slip' further exacerbates any negative feelings.

Deleting the word 'slip' (or 'slippage') from the test observations and inserting "progressive tail draw-in", might cast offset knots such as #1410 in a different light.

The paradox here is that offset #1410 (aka 'Euro death knot' - a poor descriptor that some almost manically adhere to) is actually safe to use in competent hands. I personally have used #1410 on or about 200-300 occasions in my lifetime and I have never observed 'slippage' while trusting my life to it. I still exist - that is, I am still alive (I must be since I am typing this post). If on any of those occasions I would have observed 'slippage' - I would have ceased and desisted in using #1410. Note that the vast majority of my personal use is with EN892 dynamic climbing ropes in the 9.0mm range of diameters ('half' ropes - and more recently, triple certified ropes such as the Beal Joker).

Mark Gommers

EDITS: To correct grammar and typos...

[Dan_Lehman]

I am thinking that the English word 'slip' or 'slippage' is being used in a way that belies what is actually taking place.
...
I am unclear whether Mobius and Dan have actually taken notice of the test results I had posted?
I measured a 20% reduction in the overall tail length at a load milestone of 2.0kN.
It is important to note that this 20% reduction did not occur suddenly or within a very short time interval.
The shrinkage of the tails occurred progressively (ie gradually) in direct proportion to increasing load.

The English word 'slip' or 'slippage' is not an appropriate descriptor in my opinion.

I would also comment that in all the testing I have done, I have never seen tails increase in length as load is applied.
I have however, always seen tails decrease in length as load is increased.
I have also observed that as load increases, the knot core compresses.

But this is going silly : we know what "slippage" means,
but there can be the discrimination between you & Mobius
of slippage forced by transformation of the knot to some
more stretched or compressed form
and what might come at certain forces continuously,
the knot failing to hold, then --as is shown impressively
in the French, Beal video of "pure Dyneema" braided rope
just pulling through even a double grapevine/trip.fish.
(until it did break)!!

IMO, what Mobius sees is, yes, some "silppage", but it's
what comes in the knot opening wider (the particular one
pulling in some tail from choking strand to accommodate)
on increased load --upon which it holds, but then the load
is raised and ... <repeat>.
There is also the sort of "ratcheting" slippage that was
reported by both Tom Moyer and some other testers of
solid nylon & solid HMPE webbing in the water knot
which came at light, cyclical loading, where on the
relaxing of load there was this peculiar absorption of some
SPart but not the "exterior" [my term] tail, and so by
cyclic increments, it got pulled through --and slowly (to the
puzzlement of those who recall needing tools to untie such
knots after usage!).

"I have never seen tails get longer" :: well DUH!!  But you
have seen --or sh/could have done-- SParts lengthen.

.:. So I take A_S's point in resisting "slippage" as not seeing
the sort of it-just-keeps-flowing-out sort of thing one would
worry much about, but technically it is some bit of slippage
at a point (nipping the tail) in the knot in accommodating
forces that does happen.  (And one might wonder if it could
happen in the going-one-direction-only way that was seen
for webbing in that "ratcheting" --something to be worried
about for repeated loadings, e.g. !)

AND we can note that Mobius is seeing things at relative
forces WAY WAY beyond expected usage --costing him
materials, and me gasps of exasperation (air pollution!).

AND TO THIS POINT --of EDK tail slipping, and choking tail i.p.--,
I tried amending this knot by tucking the choking tail around
back under its SPart's initial turn into the knot.  This tucking
works okay with thin+thick, but is problematic (in being
rightly oriented and not botching dressing) in like-diameter
ends (meaning "do NOT do it"!).  Done rightly, the lock
looks good --that tail can't **slip** but only stretch a bit,
the nip coming w/high load and all.
SOOOO, that particular detail might be something to see
about incorporating into a *new knot*, avoiding the issues
of dubious dressing/positioning/stability.
(A stopper knot in the choking tail works similarly,
 at the cost of this added knot-bulk.)

And please stop grousing about "EDK" :: we have enough
confusion in this world --growing in leaps & Trumps, now,
courtesy of shoot quickly and broadly Net communications--
w/o furthering that :: best info is that the name arose from
dubious Yankee observers of Euro practice (a practice that
yes we hope was safe); with the knot's broad acceptance,
though still not wholly bereft of doubters, it becomes kind
of an inside joke.  --less funny when avoiding the safer
variety (overhand) a worse one (fig.8 ) is employed;
though that one, too, has been used safely much,
just sadly not as much.   "EDK-8" is something generally
understood when uttered; "offset fig.8" is the prettier term,
which we'll hope finds traction, as "offset" then gains legs
to go where needed.

Positive tests in this setting are less helpful than ones
that reveal problems.  And my urging Mobius's testing
of the material AND dressing that got such amazingly
quick failure w/the commonly used & working EDK
is wanted to see if the backed-up version can stand up
to that --yeah, a mere single positive, but one that
should impress us (if it works) into seeing how badly
tied things can be and yet be saved.
We won't so much do this all 'round,
and can recommend something *cleaner* and so on,
but for an ace-up-the-sleeve (or wildcard!), knowing
a clumsy but stupid-tired-proof solution is a lot.
(And SAR team might be working in deliberate ways
amenable to doing some careful things; climbers in
threat of weather after overreaching their abilities
can make mistakes.)


--dl*
====

[agent_smith]

as is shown impressively
in the French, Beal video of "pure Dyneema" braided rope
just pulling through even a double grapevine/trip.fish.
(until it did break)!!

I haven't seen EN892 dynamic climbing ropes constructed from 100% 'dyneema' - have you? If you have, where? And by whom I wonder?
You can purchase this material in accessory cord form and in slings - but I don't think manufacturing has extended to EN892 certified dynamic ropes?

This then raises the question: Why test offset rope joining knots with dyneema? What will this prove?
Would it not be more relevant (and realistic) to test with commonly available EN892 climbing ropes?

One could use the term 'slip' or 'slippage' in dyneema cordage - as indeed this is what the ordinary person might witness in dyneema load testing. This ordinary understanding of the language is not applicable to EN892 certified dynamic climbing ropes - where the word 'slip' or 'slipping' is not what occurs within the operation loads of abseiling on offset joining knots.

Again, language such as 'slip' conveys the wrong impression to the layperson (as I have stated before) applying an ordinary interpretation. If ones objective was to attempt to convey mixed feelings or doubt about a particular joining knot (eg #1410) - then use of the word 'slip' would promote that purpose.

But this is going silly : we know what "slippage" means

Indeed! Ordinary understanding of this word by the layperson (in relation to life support knots) would convey feelings of doubt and apprehension.
I am careful in avoiding use of such a word - where its use is misplaced under a normal operational loading profile for an abseil descent.

And please stop grousing about "EDK" :: we have enough
confusion in this world

Indeed.
And imagine the use of this phrase to new learners. What feelings/emotions would it convey if an instructor chose to use this phrase in front of new learners?
This is casting #1410 offset joining knot in a controversial light right from the outset.

It is a sad and tired old joke - yes - and I concur with your historical musings - but under careful and close scrutiny, is #1410 worthy of such controversy? If you answer this in the affirmative, then yes the 'EDK' term should continue.

If you answer in the negative, then perhaps no, the 'EDK' term should be quietly dropped.
But who is driving this phrase? Is it the instructors who pass this on to their new climbing students? Is it forums like this on the internet?
Most climbers are not knotting theoreticians and simply latch on to popular topics they hear about and run with it. Perceived notions are obtained from those who choose to propagate them. Perception then becomes reality.

To advance my own theory:
New climbers/abseilers/canyoners have no pre-conceived notions about offset #1410. Their opinions must be derived from exposure to public forums, contact with other climbers/abseilers/canyoners, editorial/articles from books and magazines, and of course during instruction from professional climbing instructors and Guides.
If #1410 was presented in a positive manner with avoidance of the phrase 'EDK' and demonstrated effectively to new learners - the 'joke name' might quietly disappear (albeit slowly over time).
If on the other hand, those who are in positions of influence (eg Instructors/Guides/Editors and content creators of magazines and online forums) make a conscious decision to continue the use of the phrase 'EDK' - they are therefore deliberately choosing to propagate the use of the phrase. Laypeople and new learners will simply adopt and then repeat what others are telling them. If enough people are using the phrase - it reaches a 'critical mass' and it becomes self-perpetuating.

commonly used & working EDK
is wanted to see if the backed-up version

I do not support nor promote backing up #1410 with another #1410 (if that's what you had in mind). Although one can of course do this...but why would you if there is good data coupled with empirical evidence that #1410 is stable and secure within all expected operational loads (based on 1 person mass).

This is why we need some robust testing to establish reliable data points that can be reproduced by other testers.
For example: Probing for the load threshold where instability is triggered.

Mark G

[Mobius]

Hi Dan (for you  ),

Here are some images of a poorly tied back-to-back #1410 duo trialed in 11.2mm dynamic kernmantle (old climbing gym pieces I salvaged).

I kept the test loads 'low' so Dan doesn't have to remind me how I am testing WAY too high 

A missing image is at 1.5kN and if you want to see that, you can later.

Observations: I think the knot is so big it has not settled yet (even at 2kN). The top #1410 did quite a bit of moving (though no observable tail slip); hopefully you can see that from the images I provide. I thought the knot would collapse and perhaps it would eventually. However taking this knot double up to 5 or 6 kN does not prove anything much anyway.

Cheers,

Ian.

[Mobius]

What follows is off topic, though still relevant to recent discussions here I think.

The nomenclature of knots is very difficult. I have loved knots and read about knots since I was a boy. I have never studied them in earnest, until relatively recently. Despite that, I really thought I knew what words like 'stable', 'slippage' and 'secure' (for example) meant when it came to a knot.

Later I learned to qualify such words and think more in terms of statements like "xknot does not slip at 50% mbs", or, "xknot does not slip at the working load of the rope". That be may fine, however I think that hides the real problem: The problem seems to me that words like 'slip' mean different things to different people. The lexicon is in doubt.

Additionally, I have tried searching words like 'stable', or, 'instability' (for example) in Ashley (and the internet wrto knots) and got nowhere.

I suspect at least some of the reason that progress is very slow with regard to knot technology is knot nomenclature. We have not got the lexicon right yet. In some cases we do not speak the same knot language, so it seems.

If this is something others are willing to discuss, our moderators might fresh start this topic (post) for us.

Cheers,

Ian.

[Mobius]

Here are some images of back-to-back #1410's that are tied properly. Dan might be interested in these too.

This knot duo behaved better than two poorly tied #1410's, as one would hope. There was far less movement of the top #1410 with increasing load and there was no discernible tail movement.

Cheers,

Ian.

[Mobius]

Hi Mobius,

I would like to include your creations in my 'Analysis of abseil rope joining knots' paper.

What I need most is the following data:

[ ] load threshold at which instability is triggered
[ ] load threshold where jamming occurs
[ ] force required to initiate translation of offset knot around 90 degree edge
[ ] tail draw-in at various load milestones (Note avoidance of the term 'slippage'. What we are observing is changes in the tail length as the knot compresses under load).

NOTES:
1. instability is defined as: A dynamic event where there is a sudden change in the knot structure (eg capsizing) - compression is not instability and neither is tail draw-in.
2. rope diameters tested must fall within the 8.5mm - 11.0mm range.
3. data obtained for both low stretch and dynamic ropes - must be identical diameters so comparisons can be drawn.
4. unequal rope diameters should also be tested - eg 9.0mm joined to 10.2mm - the variance should be more than 1.0mm.
5. wet rope test (it is assumed that all testing is with dry ropes) - it will be useful to look at data obtained from wet ropes (soak rope specimens in water for 1 hour to condition the ropes).
6. a sample set of 5 tests should be conducted for each test objective - this will enable some statistical analysis of results.

If you can obtain these data points, it will be greatly appreciated.

I would suggest that #1410 be used as a control.


Mark Gommers

Hi Mark,

Firstly, you are welcome to use any knots I come up with that may be useful to your paper.

With regard to the one other point raised above:

The triggering of instability (as Mark defines it) is a tricky one for me to determine. My personal approach would be to first take a proposed knot up to a level of 50% mbs (~1 kN) in the 3mm polyester braid I often still use for preliminary trials. If the knot could go to that level without seeing a collapse (looking for no jamming, acceptable tail movement as well of course) than I would then feel pretty confident that the same knot would perform well in (say) 10.2 mm static kernmantle to very high load levels.

I would be much less confident about the performance of (say) 10.2mm dynamic kernmantle. Dynamic rope in climbing is a tough material to trial in, especially in terms of finding a knot that meets the standards I set. I have sometimes been tempted to trial knots in bungee cord; if the knot worked for me in that, dynamic climbing rope would be no problem  

The real issue here is I suspect I will not see any instability (or jamming, tail movement) in the knots I have proposed at the load level I can take 10.2 mm climbing rope to. The rig I have worries me if I take it above 6kN. I have seen my work bench start bowing at around 7kN and my winch (though rated above 10kN) starts makes some disconcerting noises.

I could do a battery of trials with 10.2mm static and dynamic ropes to 6kN and report what I see. Is that useful?

Cheers,

Ian

[Edit: normal typos and grammar fixes]

[agent_smith]

I'll be thankful for any quality data on tests involving offset rope joining knots.

The triggering of instability (as Mark defines it) is a tricky one for me one for me to determine.

If you look at historical test results floating around on the internet - the bulk of it is fixated on MBS yield break tests.

You will have trouble finding data on 'instability threshold' and 'jamming threshold'. The other data that is lacking is on the force required to initiate translation around a 90 degree edge. As far as I can see, Dave Drohan looks like the only person to have seriously investigated this aspect.

I strongly recommend that you read Mr Drohan's report here: http://www.bwrs.org.au/sites/default/inline-files/1%20main%20paper.pdf

[ ] page 7: illustrations of knot specimens (note his naming conventions)
[ ] page 12: test rig for measuring force required to initiate translation of knot over 90 degree edge (see also pages 22-26)
[ ] page 13: reference to tail 'slippage' (see also page 8 )
[ ] page 18 (table 2): knot strength comparisons.... (Note that I prefer to use the term "Knot MBS yield point")
[ ] page 19 (last para): another reference to 'slippage'
[ ] page 27: conclusions

A note on Mr Drohan's use of the term 'slippage'.
My personal view is that what Mr Drohan was observing was not 'slippage' or 'slipping' per se. What he observed was what I refer to as "compression induced tail draw-in".
I again make reference to the fact that when #1410 (offset overhand knot / offset water knot / offset overhand bend) is used in the field in real situations - we do not see any 'slip' or 'slippage' of the tail under nominal loads. The tails do not 'slip' - if climbers/canyoners ever observed any 'slippage', you can be assured that they would immediately stop and abandon using #1410 due to fear of death. In artificial test situations in the lab - where we deliberately load offset knots to study their behaviour - we also don't see tails slip, we see a gradual tail draw-in which is induced by compression of the knot core. I have observed this myself when applying load to offset knots - at various load milestones I paused and held the load constant for several minutes. During these temporary 'pauses / 'holds' - I never observed any evidence of 'creep' of 'slipping' of tails.
The opposing SParts are pulled and stretched away from the core while the core is also being compressed. In response to the stretching of the SParts and the compression of the core, the tails have a corresponding 'draw-in'. I would not characterize this 'compression induced tail draw-in' as slippage. Mr Drohan was not looking at knots in the way some theoreticians do in the IGKT. He certainly was not aware of technical terms such as 'offset' and what the term 'offset' actually means (the core of the knot is displaced from the axis of tension and the SParts converge along a parallel pathway).

What I like about Mr Drohan's research is that he is one of the first to look at this issue in a different way than his predecessors. His examination of the force required to initiate translation of the knot over a 90 degree edge is excellent and provides us with an opportunity to replicate his results. And this is one of my biggest gripes with much of the knot testing done around the world - it is difficult if not impossible to reproduce the test results of others.

He had a chance to examine 'instability' - but did not. I think he was still caught up in an era where the focus was mostly on knot MBS yield points.
he also did not examine the jamming threshold of offset knots.

I define 'instability' as: A dynamic event, where a sudden change occurs in the knot geometry in response to load. The sudden change can be 'capsizing' or some other tipping point where structural change is triggered.
I would urge future testers of 'offset' knots to probe the load threshold where instability is triggered.
...

The nomenclature of knots is very difficult.

With regard to nomenclature, others have tried but in virtually all cases - it seems to bog down and evaporate. Derek Smith realised early on that terminology, definitions and knot components needed to be clarified...link: http://igkt.net/sm/index.php?topic=5594.0

Mark Gommers

[Dan_Lehman]

Hi Dan (for you  ),

Here are some images of a poorly tied back-to-back #1410
duo trialed in 11.2mm dynamic kernmantle (old climbing gym pieces I salvaged).

I kept the test loads 'low' so Dan doesn't have to remind me how I am testing WAY too high 

Thank you.  And so as to satisfy your thirst for high
load values, you could do multiple low-loadings and
add them (1kn + 1kn + 1kn...), having fun just throwing
together intentionally careless tyings of this EDK-backed EDK.

It won't do well in general practice to have quite loose
knots, as those might disappear prior push coming to shove!
But somewhat so, and with some space between knots,
I think gives a fair idea of how robust this knotted structure
is, if it can handle a case that might arise in fatigued hurry
and stiff ropes (looseness somewhat modeling knotted
stiff cordage which by stiffness hasn't set well).