Note: the question of which end is to be loaded is unanswered.
(I don't think that Mike had a preference or focus re this. Lyon Equip.'s
testing really doesn't explore this well, although at least acknowledging it
--their data seem to contradict their conclusions, and shows how limited
their testing is (e.g., several of the extreme values come in the mixed-form
case (where a loopknot is tied one way at one end, other way at other)!)
- Failure will occure at one of two points, either through compression (strangulation) as the loaded line enters the first loop twists or through tight radius as the load line makes its first tight turn around the load loops. As both of these structures are identical in both the single and double turn variants, then both knots should perform comparably.
I suspect one can see effects of dressing & setting variances here: one point to inducing
some bit of
cascading of the eye-bight legs from the SPart-entry point. Also, how big
the SPart's loop (u-turn) is affects how directly those eye legs can deliver tension to that
turn around the SPart's entry.
For the real world though, I have to agree with Dans conjecture that the extra 'meat' in this knot would make the knot more forgiving of falls or jerks (of both types).
Which is surprising, in that Dan made quite the opposite conjecture; rather,
he related Dave Merchant's assertion of test results: that on rapid loading,
the extra material of the larger knots in this series 8-9-10 seems to contribute
to frictional heat which
weakens the knot. Now, weaker might still be
stronger than some other knots, but there's a good deal of uncertainty in
what knot geometries are actually of issue in the various reports of data.
Here's a quote from Mr."Life-on-a-line" himself, in a caving forum:
I agree that in a controlled tensile test the F9 is 10% stronger than the F8,
but *only* in that configuration. The increased complexity of pinch points in a F9
means that in a drop test there's no clear winner, and a loosely-tied F9 can lose
to a loosely-tied F8. I've drop-tested loads of rope with F8 one end and F9 the other,
and gave up predicting the outcome years ago.
...
One thing it seems we're getting caught up in is the idea of 'strength' based on a number for tensile pulls
-- the '70% Figure 8' or the '50% overhand' are slow tests - unless you're lifting cattle, a sport caving knot
will see the biggest force in a fall, which is a short-lived dynamic 'insult' instead of a pull. Knots all act
differently when loaded this way, and the more complex knots show wider-spaced results. An overhand
is pretty much 50% at any speed, but a F9 or F10 strength can change by a third if you load it at high speed
as there's more frictional heating going on. To say a knot is 'strong' isn't all that useful unless you know what type of force you're going to put on it.
All this is good, but I remain skeptical that Dave Merchant (LoaL) has paid
close enough attention to knot geometry--and there are so many possibilities
of variation, testing thoroughly would be prohibitively costly without some
good, clever, sampling/
projecting strategy re results. But he's clearly
paying some attention to this, and did much looking in preparing his revised
version of the e-book
LoaL, 2nd (also available in hard form, I believe).
