I tie my cordalettes exclusively with single EDK's in 6 and 7 mil nylon cord. I use my cordalette this way during the 200 or some odd days I spend climbing outside. In the several years I've been using this configuration I have not had a problem. That does not mean it is without limitations.

I find it helpful to untie it on alpine climbs where I may be threading around boulders or rock pinches.

I find I untie it less often for setting up topropes or any single pitch stuff.

There are plenty of other knots for joining ropes, depending on their application and your needs some may be better. Make sure it's dressed and you have decently long tails.

Assuming you build bomber anchors then the capsizing of the knot is irrelevant because it is isolated from the rest of the anchor legs by the masterpoint knot. If you are not isolating the knot in your anchor you may consider a different knot.

I agree with jmeizis - I've been using an doubled overhand knot to join my cordellette for 13 years. Because I'm paranoid about Jay's scenario (the factor 2 fall on the anchor, causing the overhand to roll), I back it up with a second overhand, and three inches of tail (about the width of my hand).

I'll respectfully disagree with Jay and the other naysayers. This is a common and accepted standard.

Good point. I'm not sure how I would provide evidence - I'm not aware of any surveying done on this question. Every test I've seen published about the strengths of the double overhand have shown that its plenty strong enough for anchor use, especially when you consider its only attached to one leg of a multi-part anchor.

So my anecdotal evidence: Amongst my friends here in the Pacific Northwest, I can't think of anyone who doesn't use an double overhand to tie their cordellette. I first learned about using it in 2000, became comfortable with using it within a year, and quickly saw it being adopted where I've been climbing over the past 10 years.

The AMGA teaches in its courses that an doubled overhand is an acceptable knot for a cordellette. In fact, that's how I learned about it in 2001. And I've seen it taught and used in every course and exam I've participated in since then (6 courses and 3 exams in total). I just spent a month in the Alps, and I recall seeing several IFMGA Mountain Guides using it there as well. This is very much a non-issue in the guiding community. Both knots are acceptable.

RCer "experts" know way more than any stinking AMGAers or IFMGAers

Heill mein AMGA Führer!

It is true that there isn't much evidence out there, so we have to make do with what there is. The only relevant tests I know about are in http://efclimbers.net/...nd-cord-strength.pdf. What Blue Water found is that a flat overhand (EDK) in a loop of 7mm cord had a breaking strength of nearly 13 kN, with breakage at the knot.

The breaking strength of 6mm cord is 2000 lbs, or 9 kn. 7mm breaking strength is 2800 lbs, or 13 kn. While these are adequate for static loads like rappelling, no one would consider that adequate for a shock load.

My imagination of the OP scenario looked like this: I want to use this 20' tall tree as my anchor, so I untie my cordellette, do two wraps around the tree, retie the cord, and then tie a master point. Voila.

Maybe, but if you were going to wrap the cordelette twice around the tree, there would be no reason to untie it, since you could just wrap the tied loop around once.

It seems to me (as a very occasional cordelette user) that there is no reason to untie a cordelette ...

The above method will not equalize the pieces well as most of the load will be transferred to the piece with the double strand. The single strands will stretch more because there is less material on those legs.

You could just NOT tie the cordelette into a loop.... by pulling the loose ends through the one main knot.

Pretty easy to do.
-Grab both ends together (creating an imaginary loop).
-Clip your pieces and pull down the loops to your loose ends. You can adjust the length of the loops by pulling slack on loose ends(great feature).
-Tie your knot (8 or overhand) with the loops and loose ends.
-Tighten and dress....even the loose ends

You will end up with double strands to all pieces (with no knots), a big knot, 2 loop masterpoint (for a 3 piece anchor) plus the loose ends exiting from the bottom of the knot.

Credit to knudenoggin for this concept.

Best post ever! (besides my own, natch )

Though, seriously, if one is using HMPE/Technora-cored cord,
DO tie off the tails, snug to the powerpoint knot, with an
offset water knot (EDK) --to guard against slippage through
the powerpoint knot, which in being so bulky multi-stranded
will have less "nip"/grip on individual strands,
and the slippery, low(no)-stretch cores have been seen to
pull free of overloaded-&-broken sheaths.

*kN*

Have tested this method and often he loose ends becomes so short that it is hard to tie an EDK without taking of the mittens. Therefore I keep the cordelette tied in a loop.

This doesn't add up. You said the cord is rated to 19KN, but you also said it was 6mm and nylon. There is no nylon cord of a diameter suitable for a cordlette rated anywhere close to 19KN single strand.

You'd be talking about 9mm for that kind of rating.

Fine, but :
(1) with non-hi-mod cord, you shouldn't need them tied;
(2) or have the knot pre-tied, but arrange for it
to be positioned *through* the powerpoint knot so that
you have the same result --end-2-end knot is unloaded.

AND, given the usual advice to "leave long tails" for the
offset water knot (EDK), well, that makes one pause about
how extensive one's cordelette loop is in first place, no?!
(One does NOT need long tails for this knot positioned as
suggested above, and esp. if tied snug to the powerpoint knot,
where rolling isn't a real possibility. Similarly, the grapevine
bend consumes a fair amount of rope, too.)
I.e., encountering the tails-are-too-short-to-tie situation
implies that you have made a cordelette structure that is
longer than would've been possible to make if it had some
pre-tied knots-that-consume-much-cord (EDK & long tails,
grapevine) --at least the difference musn't be great between
these).

A pre-tied EDK w/non-long tails should work pretty well,
with then the need being to simply snug this knot up
both in its own setting, and putting it near the powerpoint
--the overhand already made, and only the tightening left
to be done, to finish : mitten doable?!

*kN*

Do you think the snug-to-the-powerpoint EDK could make untying the powerpoint more difficult? (this is one of the exact reasons I try to keep the pre-tied fisherman away from the powerpoint)