Just a word about anchor tests: (or a short book, possibly :-o)
Given that NewGen anchors are "designed to the test", I think the implications of what I'm going to try and set out here might be quite broad-ranging.
These implications potentially affect a wider sector of interest than just those who pay attention to anchor test results.
There are a couple of wee problems, to my mind.
In order to come up with meaningful and repeatable figures, certain simplifying assumptions are made, or procedures adopted:
1) Anchors are tested in a homogeneous bottom (one which produces consistent holding regardless of location or direction of pull)
2) All anchors are set and loaded in identical fashion, at a constant angle (both scopewise, and in terms of
compass direction) and constant retrieval rate.
3) The tendency in presenting (or assessing) the data is to highlight the "winner" as being the anchor which developed the highest holding figure.
There are consequences to each of these.
1) As it happens, bottoms like this are also (in most cases) "good holding".
Given that an anchor which is big enough to hold a given boat in "bad holding" will not drag in "good holding", a wide-ranging sailor doesn't really need to consider how well an anchor will hold in good holding (which implies that tests are not much use to such a sailor)
Bottoms which offer bad holding include kelp/grass; rocky/bouldery; and coral heads interspersed with patches of shallow sand. Such bottoms are also notoriously inconsistent, hence completely useless for anchor testing.
Some anchors hold very well in good holding, and very poorly in bad holding. These do exceptionally well in anchor tests.
Other anchors (most notably, the fisherman) are virtually the opposite. It never holds particularly well, but it holds as well if not better in most types of bad holding as in good. The fisherman doesn't even do well enough in tests to be worth testing, but some sailors whose territory is largely bad holding still find room for one of these anchors of last resort. Generally, a big one.
2) A minor problem is that this procedure puts a premium on anchors which are not fussy about how they are set. It also arbitrarily favours anchors (even 'fussy' anchors) which happen to set well under the conditions chosen for that particular test. However some 'fussy' anchors may have unique attributes and virtues which may make them worthy of consideration, just as we might select a life-partner who requires careful handling, but repays it in spades, hearts and trumps.
More importantly: Two key things which (in practice) tend
work an anchor out of the bottom prematurely (unless moored to several anchors) are variations in angle of pull, both vertically and (especially, for some designs) horizontally.
Different anchors vary greatly in how well they handle this. The Bulwagga is a design which is inherently highly resistant to both, and the Danforth pattern to a lesser degree (at least in terms of horizontal variations: it has broad tolerance for vertical variations) while the
CQR could be expected to resist horizontal variations better than vertical (which you would expect would make it more fussy about catenary, which seems to be borne out in practice: it has problems in shallow bottoms, or with light chain)
The horizontal variations are imposed to varying degrees: some boats sail about incessantly at anchor, and some anchorages have a tendency to fire "bullets" of wind from various directions.
But I don't
recall seeing a meaningful attempt to address this in anchor tests.
Think about trying to extract a nail from a piece of timber, or a wisdom tooth from a strongly holding gum. The difference between the force it takes to "worry" it out, versus a brute-force, straight line heave, is considerable. And it's not (to my way of thinking) sensible to assume there will be a constant proportion between the 'worry' and the 'brute' force readings, for a given design of nail, or of tooth, across a range of substrates.
But that assumption, it seems to me, is implicit in anchor testing.
The third key thing is to do with variations in quantum of loading. It's rare for an anchor to see a constant loading in the real-world conditions which pull anchors out: It
can happen, if you're anchored just inside a breakwater or spit to windward, but in most situations high winds generate waves which modify the loading on the anchor. Anchor tests cannot affordably model cyclical, let alone irregular loading, realistically, so they do not model it at all.
And, like economists, anchor testers immediately forget the simplifying assumptions they have been forced to make, and draw conclusions for the real world which bear scant relationship to real-world cause and effect.
3) What interests me about an anchor is similar to what interests me about a
weather forecast, when I'm contemplating a challenging sail.
I'm not interested in how GOOD the weather might be, at best; I want to know how bad it might get at worst.
My favoured anchor will be the one whose holding is the least crappy, in the crappiest of crappy circumstances. Not a show-pony, but a hard-bitten mule.