I am trying to find some testing on the strength etc. of an Alpine Butterfly knot (ABK)
when the loop is used as an end loop (i.e. only one side is loaded), ... .
It will be for rescue work so strength, lack of slippage etc. is important
(when aren't they?).
It will be tied in 11mm static kenmantle rope.
You can research the history of such usage and come
to a suggested answer to this question : that strength
is poorly understood and unlikely of much significance
in these applications --despite some rhetoric to the contrary.
The ABK is great for creating a mid line loop with minimal strength loss
in a line that will be loaded (i.e. both tails are loaded in opposing directions),
but for an end loop - e.g. attaching the load line to the anchor -
Still, one must realize that some intended uses of this mid-line
eye knot are expecting that it will become effectively just such
an end-line knot --e.g., in anchors where this eye knot forms
a "Y" to attach to a 2nd anchor, in case one fails.
It looks like there will only be one turn gripping the loop so I suspect slippage will be an issue.
You're half right : some slippage of the eye leg leading
to the side that isn't loaded can contract that collar
such that the knot is hard to loosen; but beyond this,
there is no worry.
You have not remarked --nor do most presentations show--
that the
butterfly knot is ASYMMETRIC : one half differs
from the other; this implies that one has essentially two
similar but slightly different end-of-line eye knots, depending
upon which end is loaded. There is also variety in the dressing
of the knot. (I strongly advise against the supposed end-of-line
orientation shown at the Layhands site.) Your posted images
are of low resolution, but it seems clear enough that you are
using a commonly shown dressing in which the eye legs (or
tails, if an end-2-end joint) do not
cross ; I'll post again
some images in which they do, which I think is better.
Have any of you seen any testing on the strength of the ABK when used in this way - only one side is loaded?
The
CMC Rope Rescue Manual gives test results for both
this and the
fig.8 used in both manners --eye knot & mid-line
impediment. You might be surprised to see the figures for the
latter case! ("Rhino" low-elongation 12.7mm nylon kernmantle)
qua eye knots :
fig.8 80% ;
butterfly 75%
qua mid-line ...: " " 65% ; " " " 69%
to which Dave Richards did --12.7 low-elong., 10.5 dynamic, &
7mm "accessory", resp. -- 76%/70%/74% vs. 80%/71%/72%
NB: we have no good indication of actual DRESSING/orientation
of the tested knots; my guess is that the
butterfly was NOT
with crossed legs, and that it would be stronger with them, for CMC;
and maybe Richards's results indeed show this --but we don't know
his orientation, either.
BTW, the
directional fig.8 was, resp., 75% & 59% by CMC.
There are some real questions to ask regarding such
mid-line knots as to what values they might have given
certain conditioning --e.g., if first loaded end-2-end,
and THEN loaded qua eye, or vice versa.
(MOST curiously, both of these testers tested the
fig.8
eye knot as tied-in-the-bight & re-threaded, as though
the test device should care how it got tied? --and without
any helpful indication of what actual differences one might
suspect via tying method : e.g., a field survey might show
that dressing corresponds (w/difference) to tying method
(whereas written guidance, such as it is, indicates that
either method should reach the same result --seldom is
that actual geometry specified w/clarity if at all). !!
Ahhh, but at least there are "numbers" to toss around!)
As Roo has suggested, there are knots similar to the
butterfly--i.e., ones of interlocked
overhand knots easily untied--
that should perform better; they do so by loading different
parts of the knot --i.e., the
butterfly essentially loads both
tails of the end-2-end knot, rather than one pluse the
opposing SPart. (Compare with
bowline &
sheet bend .)
I would favor
Ashley's bend #1452 here. Of course, one
could adapt the
butterfly similarly --and then it becomes
a choice of which side to load both ends of (the asymmetry).
Then, again, you could bypass this and use some
secured
bowline as a means to getting an easily untied eyeknot.
--dl*
====