Accidental release of “releasable” anchors, forces involved
(relevant to canyoneering)
(Takeaway: The CEM is harder to release by
accident than the Macrame or fiddle-stick type devices, and the CEM is
easier to release after weighting, than the Macrame.)
This report discusses three releasable anchors: 1) the
“southwestern” CEM knot, 2) the macrame (also called equivocation hitch), and 3) the smooth
operator
(a variant of the fiddlestick). There are "none to few" reports of
serious accidents from unintentional release of these three, perhaps
because
canyoneers are well-disciplined and trained (I'm not a canyoneer). In
particular, fiddlestick variants have been used for thousands of
descents with complete safety. I've done these tests many times before
(e.g. for smooth operator), but for this report, I try to keep the pull geometries consistent and compare the full forve vs. time profiles.
A releasable anchor is useless if it cannot be released.
There is a fine line between those that can be released easily by mistake,
killing the rappeler, and those that just might cinch up and keep the rope (that you
were expecting to retreive), thus putting your life at
risk for other drops down a canyon.
Three forces must be considered; A) the amount of force
needed to pull the release by mistake, if you happen to accidentally grab, or
get tangled in the release cord; B) the amount of force needed (on the rap side) to cinch the
release so tightly that it cannot be undone at the end of the rap, so the rope
remains stuck above you; and C) the force required to break the rap side. While
the last is often measured, it is largely irrelevant, unless one is rapping on
marginal cords.
I’ve tried many releasable anchors over the years.
This
issue is personally very relevant, since my brain damage makes it very
easy to be
distracted by extraneous conversation. I now think the release
cord
should be kept bagged and failsafe till the LAPAR (last person at
risk)— or it should
look very different (e.g. Amsteel with a fiddlestick) and be clearly
out of the
way for the LAPAR.
(A)
Force needed to pull the release cord accidentally.
These tests involve a free-hanging rope in my stairwell, with
me suspended from the "rap side" in my harness. The load cell is attached to the
pull line, to measure the force on the pull line when it is weighted
accidentally. In a slow-moving world, the force on the rap side (which is
providing some stability to the release) decreases as the force on the pull
side increases, following:
body weight = (force on rap side) + (force on pull side).
This equation breaks down as the accidental release becomes
very hard, and is buffered by the springiness of the rope and amended by energy
from body muscles. We previously tested with
one person hanging from rap side, and another person pulling on release
cord; that is a less severe case and requires a harder pull for
failure.
The three cases are shown below for CEM, Macrame, and Smooth Operator (click small images to see full-sized images):
 CEM never failed after multiple
bounces to near body weight. Rope pulled down at end. NB: I weigh <150 lbs with gear. |
 Movie (5MB) |
 Macrame failed in stages, as each half-hitch popped. The harness force is body weight minus pull side force. |
 Movie (3MB) |
 Smooth operator failed at ~1/3 body weight. |
 |
We cannot model two important and vaguely defined
situations. First, the most dangerous time for accidental release comes just as
one leaves the anchor, when the rap line is not fully weighted, and the release
(pull) line may be near at hand and easy to confuse with the rap line. Second,
if one reaches an intermediate point on the rap (say comes to rest on a ledge),
unweights the rap side, and then somehow pulls the release cord, results can be
very bad, especially with fiddlestick-type devices.
(B) Force on rap side needed to cinch up anchor (make it unretrievable)
The tests were just on the CEM and macrame. The test setup is shown in this movie for the CEM, which also shows the CEM knot was easily pulled after loading. The rope was 100% PET 8 mm (really about 8.2mm) NER Maxim. Briefly, the knots were tied to a hardshackle anchored on a tree, then pulled with a powerpuller, with the linescale 3 measuring the forces at 40 Hz. I've found the scale has a reproducibility of about 20 lbs at these forces. The Macrame was pulled to 738+/- 20lbs, and CEM to 736 +/- 20lbs. The Macrame force profile for the power puller is below; the final drop to zero force occured when I manually released the tension on the power puller (i.e., was NOT a break of the rope). After these pulls, the Macrame was cinched very tight and had to be cut free with a knife. In contast, the CEM pulled easily.
(C) Force on rap side needed to break anchor
Below are the forces on pulls of the Macrame and
CEM knots, when they were pulled to breaking behind my jeep. Neither
broke in the anchor knot; both broke in the figure 8 knot at the jeep.
This rope is reported to have a tensile strength of 12.8 kN (2878 lbs)
which (from experience breaking other maxim cordage) is likely an
underestimate. In any case
the forces needed to break these knots are still unknown, but
definitely above 2196 lbs (9.8 kN), enough to seriously injure the
rappeller.
For the CEM, the knot popped in several stages, and each half-hitch failed. The CEM movie is here.