Mantus vs Rocna

[thinwater]

In reference to the John Knox testing. I wasn't there, but I do have the original files of his notes. (massive tomes of paper). I note that most of the testing was done using steel cable, not elastic ropes. Presumably to take out another variable, and to be able to measure force that is applied to the anchor from the other end of the rode.
On scope, as most numerical testing was done at an angle equating to high scope, (>5), short scope effects were only included in the boat testing later.
On independence, at the time of the testing he hadn't invented his anchor. It was his opinion that the disappointing results required some design thought, and that led to the anchor development. There was a lot of subsequent testing on various iterations of the design, mostly using the same measurement rig used in earlier comparative tests.
Not all the testing was published, but a few still existing reference (web based) are listed on the website.

I know this is asking too much, but additional reports could be scanned and posted to DropBox or similar. I doubt the information is considered proprietary and the world would gain from it.

 

[NormanS]

In reference to the John Knox testing. I wasn't there, but I do have the original files of his notes. (massive tomes of paper). I note that most of the testing was done using steel cable, not elastic ropes. Presumably to take out another variable, and to be able to measure force that is applied to the anchor from the other end of the rode.
On scope, as most numerical testing was done at an angle equating to high scope, (>5), short scope effects were only included in the boat testing later.
On independence, at the time of the testing he hadn't invented his anchor. It was his opinion that the disappointing results required some design thought, and that led to the anchor development. There was a lot of subsequent testing on various iterations of the design, mostly using the same measurement rig used in earlier comparative tests.
Not all the testing was published, but a few still existing reference (web based) are listed on the website.

I seem to remember that the testing was done on a beach when the tide was out. Can you confirm if this is correct? If so, the results are not necessarily indicative of the behavior of anchors in real life.

 

[noelex]

I know this is asking too much, but additional reports could be scanned and posted to DropBox or similar. I doubt the information is considered proprietary and the world would gain from it.

+1

Here are a few of the articles written by Professor Knox. He had a refreshing scientific approach to anchor testing. His contributions will be missed. These are some of his published works for those that are interested:


http://www.spade-anchor.com/IMG/pdf/2011_-_Test_et_Comparatif_Ancres_-_PRACTICAL_BOAT_OWNER.


https://static1.squarespace.com/sta...47a583b3e/1509856497708/RHYC+article+2012.pdf


http://www.stfeurope.com/pdf/Practical-Boat-Owner-2002.PDF

There is more, but I do not have direct links.

If there is any unpublished work, It would be great to make these available.

 

[MM5AHO]

Yes, many of the tests were done on a beach (Longniddry) - you might just about see it from there? The second testing ground was west coast in various Mull places. (Boat was then moored Oban)

 

[MM5AHO]

I have it in mind to do that one day, or perhaps compite some of the interesting parts into an article.

 

[Neeves]

Yes, many of the tests were done on a beach (Longniddry) - you might just about see it from there? The second testing ground was west coast in various Mull places. (Boat was then moored Oban)

I vaguely recall one of the moorings was 3 retired, recycled, submarine anchors - either Danforths or danforth types.

Jonathan

 

[NormanS]

Yes, many of the tests were done on a beach (Longniddry) - you might just about see it from there? The second testing ground was west coast in various Mull places. (Boat was then moored Oban)

Yes, his mooring was close to our own one.

 

[Neeves]

Professor Knox tested his anchors at Longniddry at low tide - but in the remaining intertidal pools. I suspect this is still not exactly the same as the seabeds in which we actually anchor - but it is not simply testing in wet sand.

In the Knox article in PBO August 2002, I think it was titled 'Will my anchor hold' the rode is defined as:


"The cable was nylon multi plait or, for the heavier anchors, wire plus 1/4” chain.’

Whereas it is realistic to compare the anchors tested with the multi plait and to separately compare the anchors with the steel rode - it would be a brave man who could, hand on heart, compare those tested with nylon and those tested with steel.’ There is a very large difference in elasticity of the 2 rodes and in their diameter and the effect of that diameter aiding or detracting from the ability of the anchor to set (a thin wire will set much more easily than a beefy bit of nylon (especially when the thin steel is attached to a beefy anchor and the beefy cordage attached to a malnourished anchor).

edit

The effects of the different rodes would reduce, relatively, the hold of small anchors and enhance the bigger ones. As long as you did not compare large with small, and only compared small (and as long as in another test you used the same rode) then you could compare.

However I do recall a comparison was made in an attempt to extrapolate hold with size and the results achieved were simply contradictory to all other tests publicly available. Most anchors scale 'badly' double size and you do not double hold in fact there are no known anchors where if you double size (measured by weight) you do not double hold. Prof Knox did find doubling weight more than doubled hold - and this result has not be achieved by anyone else - so those scaling results need be view with some caution.

Oil rig anchors, where huge sums are invested, are closes to 100% efficient, they achieve 93%, Fortress is around 83% and Bruce around 70%.

It is a brave and possibly foolish man who thinks doubling weight of our anchors doubles hold - think around 70%/85% - so double size and increase hold by 70% and (optimistically )- maybe as much as 83%.

close edit.

The methodology of testing is well defined but the difference in use of nylon and steel is glossed over. The tests were conducted over a long period of time and given the difference in the rodes it is dangerous to compare tests made on one day with another (different) intertidal pools (the difference is possibly not relevant) and possibly different rodes.

Interestingly Prof Knox methodology is very similar to that used by Anchor Right in Australia. - and I am sure Anchor Right are pleased that Peter Smith approved of the technique

Jonathan

 

[NormanS]

My recollection of these tests was that some form of winch was used, giving a steady and prolonged pull, markedly different from the pull of a boat on its anchor. I also seem to remember that the anchors tested were of varying weights, and that some, although called by well-known names, were actually cheap copies. It may be that my recollections are wrong, but I do recall being unimpressed. I readily accept that it is very difficult to do realistic testing, but certain parameters are essential.

 

[MM5AHO]

I've seen the collection of anchors used. The CQR for example was a genuine one.
The fact that a steady pull by winch doesn't replicate a boat action on the rode doesn't make that specific series of tests of no value. There are many tests in engineering which are comparative rather than trying to emulate real life. In corrosion testing for example, the salt spray test for organic coatings, has no bearing on real life expectancy, its a relative test for comparison on one coating versus another. In the anchor tests, it is pretty much impossible to totally replicate a boats action for a series of tests between varying anchors, so a combination of science based tests for comparison alongside the experience of users is more meaningful that one with out the other.
One major point that many overlook is the difference between straight pull until an anchor slips (ploughs, drags), and a pull until the tension reduces as the anchor ploughs, then resting it, then resuming the pull until that loaded point plateaus in force. This John called static holding force, and then divided this by the weight of that anchor to get a ratio of this force to the anchor size. Most people measure the anchor by weight, though it's arguable that fluke area might be more relevant.
I wonder whether anyone will ever devise a test of this (strangely emotive) topic (anchors) that will satisfy all critics.

 

[Bouba]

I wonder whether anyone will ever devise a test of this (strangely emotive) topic (anchors) that will satisfy all critics.

Easily. As long as a Rocna comes out on top and justifies the €500 I spent, I will be satisfied

 

[NormanS]

As I have said, it is obviously difficult to replicate real life use of anchors in a test, and I accept that the method used gave some useful comparisons. However some care is needed. I have now found some details of the quoted tests, carried out on the beach when the tide was out, and would draw your attention to one particular paragraph:-

"Correction. Page 102 PBO August 2002.
Any reference to a Bruce anchor in the first article should read Claw anchor. Similarly, the anchor referred to as a CQR in the first article was, in fact a copy of the CQR pattern, and should be referred to as a Plough".

To me, that somewhat diminishes the credibility of these tests. I think we all know that some copies of proprietary anchor designs behave in a much poorer way than the originals, and indeed have maligned the reputations of the originals.

 

[thinwater]

The problem isn't so much that we don't know how to test anchors. The problem is money. At consulting engineer rates, what would it cost to run each anchor through a full range of tests, varying scope, bottom, pull rate including soaking periods, yawing, and sustained veers, and then repeat for, say, 10 anchors? Way into 6 figures and more than the annual editorial budget of any sailing magazine is the answer. This is why test programs are comparative and abbreviated.

And who would understand the subtlety of the results? There would be arguments that the bottoms were not representative. There would be claims that some of the testers were biased, since surely some industry participation would be vital. In realistic (not consistent sand) bottoms, scatter in the data up to 50% is the rule (I've done a lot of testing), so on any given day, you can't be sure what the anchor will do. There would probably be no obvious winner, in all conditions. And rigging would make as much difference as the hook.

Good luck. I've done my share of testing, and I'm giving it a rest until I see another paradigm shift.

 

[Neeves]

It is an anchor thread - so I need to disagree.

It is quite feasible to test anchors using the same rode each time. There are many locations where it is possible to set in water using a winch and be able to keep the scope constant. One accepts the seabed varies so you test anchor A then B then C, 1m apart and then retest A, B and C and retest until you look to be generating statistically significant data.

You need to be slick as tide waits for no man. You do need committed support and it helps if it is a husband wife team - as you need someone to keep clean and record the data!

You now have a base load of data in one seabed. You can repeat this test in very different seabeds. I can think of locations I can use with clean sand, heavy weed, clay with some weed and very thin mud. To me the skill is reproducibility - if you cannot test the same anchor in the same seabed and not achieve similar results - there is an issue

As you build up the data you can vary the test - here is one on resetting

https://www.practical-sailor.com/issues/37_26/features/Anchor-Resetting-Tests_10981-1.html


The trouble being that who wants to be told the anchor they bought is not as good as they thought nor have been spruiking and is actually no better than any design of pre 1990. Mark Twain was right.

Without promoting any specific anchor or any specific design trait:

A 15kg old gen anchor will have a hold of around 1t and good NG around 2t in a decent sand seabed. But no yacht, or few, that would use a 15kg anchor are ever subject to their rode tension actually being 1t - yet anchors drag (in much more benign conditions). There are factors other than 'simple' hold that determine whether an anchor drags (care in setting being one).

But the fact that a carefully set Delta, which can have a hold of 1t, can still drag suggests there are other tests that need be conducted - IN ADDITION TO HOLD - but mechanics of 'failure' are not defined and agreed - so developing tests to measure the unknowns are few and far between. I made an attempt with my resetting tests and independently (and later) Morgan's Cloud came to the exact same conclusions (based on a different data set - real life) - but that is only part of the story.

So - my re-setting test was undoubtedly contrived (designed to be as close to real life as was possible) but its conclusions and those of Morgan' Cloud based on real life incidents are identical - maybe coincidence - but maybe contrived but carefully thought through tests do achieve the correct result - even if you do not like the answer. (Mark Twain - again)

Jonathan

 

[Bouba]

I can think of ways to test anchors. Only test two anchors at a time. This makes it easier to put them in near identical positions and easier to have a variety of sea beds and pulls from various angles and rode lengths. A days anchoring can provide a lot of data. Then decide which is the best, next day or whenever take the best anchor and a different one and find which is the best. And so on and so forth.
At the end of the trial we will have the best, nobody cares about second or third place. And when a new anchor is developed it only needs to be tested against the champion and it can be done in a day for the cost of an anchor.
If tides are viewed as a limiting factor then do the test in the Med. But with only two anchors both should experience the same tidal state.

 

[Neeves]

Bouba,

Valid - except that no one anchor can be perfect in every environment, or on every bow roller nor for every pocket. We then might want to throw in alloy.

Each anchor will have a weak point, or 2, and each will outperform other anchor in a specific bottom.

How to pick the champion. - when you provide that choice - it will be the foundation for another debate - and a lot of manufacturer will be unhappy.

But I agree with the fundamental - limit the examination to a few, build up a foundation of reliable data and then take that data (and one key anchor, to compare another limited batch.

The Med - very attractive - but uneconomic from here

Jonathan

 

[noelex]

The problem isn't so much that we don't know how to test anchors. The problem is money. At consulting engineer rates, what would it cost to run each anchor through a full range of tests, varying scope, bottom, pull rate including soaking periods, yawing, and sustained veers, and then repeat for, say, 10 anchors? Way into 6 figures and more than the annual editorial budget of any sailing magazine is the answer. This is why test programs are comparative and abbreviated.

This is the problem. Sailing magazine budgets are not what they used to be and sadly I don’t think we will see, at least in the near future, a repeat of the comprehensive, independent tests that were conducted around 2006-2009. This is a pity, many modern anchors have no independent holding tests.

Unfortunately, anchor tests conducted by or paid for by anchor manufacturers always result in the manufacturer’s anchor winning so can be largely ignored, or at least they should be read with a great deal of skepticism.

To stay with the OP’s question, there have been two tests that have included both the Mantus and the Rocna.

The Fortress Chesapeake Bay test:
It was obviously paid for by Fortress who and also did the analysis. The Mantus recorded an average holding ability of almost double the Rocna’s result.
I don’t think either of the results are representitive of these model’s capabilities. The standard deviation was way too high to draw meaningful conclusions about these two anchors, but this test is one of only two pull tests that has included both the models the OP is interested in.

The Kippari test:
This is a Finnish sailing magazine that conducted an independent test in a disused quarry of a number of anchors including the Mantus and Rocna.

The Mantus and Rocna were the only two anchors out of the 10 models that were tested to be awarded 5 stars, with the Mantus declared the Testivoittaja or test winner. The holding ability of the Mantus was about 20% higher than the Rocna when tested in the more difficult firm substrate, but both anchors did very well and the results were very close overall, especially as there was no analysis to determine if the Mantus win over the Rocna was statistically significant.

In comparison, the Delta (which I think is useful benchmark anchor) recorded under half the holding ability of the Mantus in softer substrate and only about 1/5 of the holding ability in the firmer substrate. The only anchor that came close to Mantus and the Rocna was the Fortress which as expected did exceptionally well in the softer substrate, but poorly in the firmer substrate and was only awarded (I thought rather harshly) two stars overall.

I will see if I can find a link, like any test you are better to read all the results rather than rely on a summary.

For me, far more convincing than these limited and somewhat flawed tests is observing the performance underwater of a Rocna and identically sized Mantus. I have deployed both anchors for over 1000 nights each. On many of these occasions I have dived to observe exactly what the anchor is doing on the seabed. My conclusion is that they both great anchors and you cannot go wrong with either, but the Mantus has the edge, especially in some of the more difficult substrates.

I sold the Rocna with my old boat and kept the Mantus.

 

[Neeves]

A Norman has mentioned, indirectly, the Kippari tests was conducted in a quarry and the seabed was the quarry floor - it might have been damp - but no water in sight. A Norman quetioned the location of Prof Knox work I have to question the relevance of work on anchor performance on a quarry floor (though it might have been on a stockpile of quarry 'fines'? - but certainly no water in sight)

If you want to reply on a dry substrate, or a recommendation from someone who think a dry substrate is applicable, to choose an anchor - be my guest.

Noelex you make some really questionable comments, like a big anchor is safe at short scope, like a Spade fluke plan view profile is not very similar to that of a Rocna, like a Rocna shank profile is not similar to a Spade, like tests in a dry quarry floor have credibility - it would be very advantageous if you could justify these extravagant claims. It would also be useful if you can explain how your Mantus set with the fluke at 16 degree in the seabed, performs (in contradiction to work conducted by Vryhof, the US Navy and HM Navy and a number of Universities) as well as a Rocna, of the same size/weight whose fluke/seabed angle is 30 degrees.

However I'm not expecting any justification from you - you have been making these extravagant (and dangerous) claims for months and/or years and when questioned - you never, ever, support your assertions.

https://www.practical-sailor.com/issues/37_74/features/An-Inquiry-into-Anchor-Angles_12153-1.html

A new article with extra hold data is with the editor at Practical Sailor now, incorporating comparative hold data - which simply re-affirms the 2017 article.

The Fortress Chesapeake tests on Mantus are an almost exact prediction of how an unweighted fluke anchor would perform with a fluke/seabed angle of 16 degrees. Subsequent tests with load cell show the same result - Mantus has the same hold as a Delta in sand and about 50% of the hold of a similar sized Spade.

Guessing performance is not part of anchor assessment. sadly making dangerous asserations on the internet without any support or justification - is the new 'Normal'. Though how Noelex can deny the evidence of his own photographs is a complete mystery. Credit where it is due - without that excellent and compressive collection of images collected by Noelex it would have been more difficult to question the performance of Mantus - interestingly the thread, which had been acclaimed and made a sticky was allowed to sink and be archived (along with the videos on the Mantus website) after the 2017 PS article was published. Nothing is lost on the internet.

Mantus will perform well in a really hard seabed. Mantus sets well, as does Rocna, Spade, Excel - but if you are interested in an anchor that performs well in a very hard seabed and optt for a Mantus - please ensure you buy one twice the size of the recommended Rocna, Spade or Excel - as you will then have the same, roughly, hold in a more common and seabed.

Jonathan

Edit

It is easy to fool people it is much more difficult to convince them they have been fooled.

close edit

 

[Neeves]

For me, far more convincing than these limited and somewhat flawed tests is observing the performance underwater of a Rocna and identically sized Mantus. I have deployed both anchors for over 1000 nights each. On many of these occasions I have dived to observe exactly what the anchor is doing on the seabed. My conclusion is that they both great anchors and you cannot go wrong with either, but the Mantus has the edge, especially in some of the more difficult substrates.

I sold the Rocna with my old boat and kept the Mantus.

Either you are collecting multiple Mantus anchors, or this is a new one - or maybe you painted it for the occasion - but this is not an anchor used for over 1,000 times

http://www.cruisersforum.com/forums/f47/bestevaer-49st-147499-113.html

Post 1686.

Interesting - though you have used both a Rocna and Mantus for over 1,000 times I do not recall once seeing an image of your Rocna (in the seabed) - yet you featured other anchors set by other people. You suggest Mantus is
'better' but you never thought it interesting to compare them even though you dived on both.

You also mention the Mantus is better in 'more difficult' seabeds - so where were these 'difficult' seabeds, how did you know the Mantus was better (did you try the Rocna in the same seabed) - and importantly did you take a record, image, underlining why the Mantus is better.

You are being critical of Rocna - I think it would be polite and fair (to CMP/Peter Smith) to see some evidence and educational for owners (of Rocna)

Jonathan

 

[noelex]

As promised below is the link to the Kippari test which tested the holding power of the Mantus and the Rocna anchor. I would encourage reading the report rather than relying on my summary. You can use Google/Bing translate, although I would suggest the references to the penis in some translated versions may not accurately reflect the authors intent

There are also some nice side photos of the fluke/shank angle of different anchors on page 40.

Unfortunately, these modern tests do not have the budget that was available for the earlier large scale studies so they do have some limitations. Nevertheless, this is an independent study that looks at the holding power of both the Mantus and Rocna anchor so I think it worth reading, together with Fortress test if you are interested in the performance of these two anchors (although I would further caution that the Fortress test has been paid for, conducted and analysed by an anchor manufacturer).

http://www.uchimata.fr/wp-content/uploads/2015/03/Essais-comparatif-Mantus-03-2015.pdf