How to Test Anchors?

[boasun]

If you really want to test your anchor.
Pick a beach or area that the surface ground is similar to the areas that you normally anchor in.
Using a tracker or a large 4 wheel drive SUV.
Testing various type anchors one at a time, by placing it on the ground on its side and dragging it along until it rights itself and burys into the ground.
Continue pulling using a dynometer placed in the anchor rode and note what the high pull weight is when it is dragged out of the ground.
Another test is once the anchor has buried itself, reverse the pull and see if the anchor will hold or pull out and rebury itself again, or twist around while buried.
Do this all of the various types of anchors and you will soon find the best for that type of holding ground.
Don't recommend doing this on someone's unmowed lawn to see if it can defeat sea weeds.
But you will end up with a list of anchors that bent, broke, pulled out under light loads and so forth. It will be a vary confussing list if you vary the holding ground from sand to mud to rocks and soforth.

[Kettlewell]

I don't think pulling anchors across a beach is a realistic test as the saturated soils on the bottom, with shells, weed, etc. is very different holding ground. Plus, you have the scope issue to deal with--getting a realistic angle of pull is important.

[estarzinger]

Sorry I disagree with almost everything you wrote

Originally Posted by Kettlewell

OK, but doesn't it make more sense to have a main anchor that performs well 6 out of 7 times

My point is that ALL the main anchor designs will do just perfectly fine in a good hard mud or soft sand bottom, if properly sized.

I would venture to guess that mud and sand would easily account for more like 90% (maybe more) of the anchorages frequented by typical world cruisers.

No, just for instance, in the tropics, a very common 'bad bottom' is a thin sand or thin loose rubble over a hard rock or coral surface - that bottom is more than 10% by itself (in the tropics)

I would argue that it might be better using a special anchor in that situation.

No, that's impractical. First, as I said, we often don't actually know what the bottom is - and then following that strategy you would need to carry several different 'main anchor sized' specialized anchors - for differentspecialied bottoms. And no-one that I know plays with multiple specialized anchors when you get to the 50kg size I use. We mostly just drop the biggest anchor we have off the bow.

Shorter me, I would rather use two anchors, or maybe one specialized anchor for the oddball anchoring times, so that I can use a lighter, more efficient, general-purpose anchor 90% of the time.

Why do you want to use a lighter anchor, unless you don't have a windless? Two anchors, neither of which has set is a real nightmare waiting to happen.

On the short scope thing, it is far better to use two anchors to limit your swing.

Not if everyone else is on one short scoped anchor, which is typical around the world in the crowded anchorages. And in many of the anchorages you will spin around at night and end up with a tangled mess of rodes. And if the wind shifts you end up only sitting to one of the anchors anyway. One big anchor is much better.

......

[Jim Cate]

Originally Posted by estarzinger

Sorry I disagree with almost everything you wrote



......

100% agreement with Evans on all points from another cruiser who often anchors in crowded places with unknown bottoms and no local knowledge.

Cheers,

Jim

[Kettlewell]

I think we may be having an argument of degree rather than substance. For example, I use a 45 lb. Bulwagga as my primary anchor on a 38-foot heavy sailboat. That is really considered one-size oversized for my boat, yet it is still possible for me to pull in that anchor by hand with all-chain rode in 50 feet of water. I consider that reasonably light, but many anchor tables would consider it to be too heavy. It has reliably held me at 3 or 4 to one scope when I have needed to use short scope--in one case up to 56 knots when we were surrounded by reefs and couldn't put out more scope. However, I wouldn't want to consider using an anchor that required me to use say a 75-pound version due to its limited holding power in what I consider to be normal bottoms, sand and mud. What a waste hauling around all that extra weight for the occasional anchoring situation where the Bulwagga won't hold due to the bottom being too hard. In that scenario I would rather use two anchors, which would probably hold me much better than the single heavy one, and be much easier to handle, in my opinion. A good anchor should not require more than one pound per foot of boat length to hold up to gale force.

As to the two anchor thing, I know I will not get anywhere with everyone who insists that one anchor is the way to go no matter what, but I have to put in my two cents. I have spent hundreds of nights at anchor with two anchors out in harbors where most people are on one, and where most people consider the waters to be too crowded. Anyone here been to Cuttyhunk, Newport, Block Island, or Spa Creek in Annapolis? Did the same thing all over the San Blas Islands in Panama. Did it in Mexico where the bottom was too hard for one anchor to reliably hold. I routinely anchor in crowded places on two anchors, thereby limiting my swinging room and preventing incidents of swinging into others or others swinging into me. I often use two anchors to anchor in secluded and secure spots well away from everyone, but too tight to be able to swing on one anchor. Sure, it takes a big of technique, but I have done it for thirty plus years. I can't count the number of times the single-anchor folks have dragged away from me in unexpected blows or sudden wind shifts, no matter what the weight of their anchoring gear. Sure, tons of places don't require two anchors, but in many harbors you can be more secure on two anchors.

[estarzinger]

Let me suggest you look at this a different way - your objective is to do a test that helps people avoid dragging - that's your ultimate objective function right?

So lets talk about when and why people drag. I don't know any collected statistics on this, so we may disagree, but I think I have a pretty good sense of why people around me have dragged.

In order:
#1 reason they used (too) short scope (sometimes that's by choice but others its just poor skill - but in either case it would be nice if the anchor could compensate).
#2 reason is they did not set the anchor (again, sometimes by choice because the bottom just does not allow a solid set and others just lazy or poor skill)
#3 reason the bottom was bad - anchor did not stay set
#4 reason is wind shifted and anchor did not reset
#5 reason the anchor was set in a good bottom but did not hold.

So, I would suggest your proposed test is aimed at addressing the smallest source of dragging problems, and missing the primary sources.

3 out of 5 of the scenarios depend on the setting capability of the anchor, and the most single common scenario is scope

So, I don't know if you will agree with my scenarios or their rank order, but does this methodology (identify the actual priority causes of dragging and then test against the top causes) make sense to you?

---------------------------------------------------------------
As a small aside re your above post - my opinion is that a cruiser's main anchor should be sized for storm rather than gale conditions, and that's twice the force/loading (wind force goes with the square of velocity). Personally I think one lbs per foot is light for storm conditions - again it will in fact probably hold in a perfect bottom but not in a less than perfect bottom.

[Auspicious]

Originally Posted by Kettlewell

I think we may be having an argument of degree rather than substance. For example, I use a 45 lb. Bulwagga as my primary anchor on a 38-foot heavy sailboat. That is really considered one-size oversized for my boat, yet it is still possible for me to pull in that anchor by hand with all-chain rode in 50 feet of water.

Geez John. You're stronger than me. 50' of 5/16" HT plus your anchor is a bit over 100 lbs not including breakout and friction. Sure every foot you haul makes the rest lighter, but golly. Not something I want to do on a moving foredeck with wind and sea in my face. You are a better man than I.

I agree with Evans - I want an anchor I can count on the widest array of bottoms since I often don't know what the bottom type is and find that the chart data is often wrong.

[lancelot9898]

Many of these anchor tests seem to compare how well various designs hold when the anchor is set in general proximity to one another. I always wondered how well controlled the bottom is when that comparison is made assuming that the consistenancy of the mud etc could be different even within a few feet of the prior test. Sufficient number of tests would be needed to minimize that possible error.

As others have mentioned a straight line pull is useful info, but changing directions and dynamic loading is the failure mode of most anchoring situations. That to me is the most useful comparison.

One other test(destructive) that could be done is putting the anchors in a fixture allowing a force to be applied to see when bending(or breakage) is effident. Just how strong is the metal and the welds assuming that the anchor is holding firm on the seafloor. Doing hardness testing on the metal might also yield some useful information.

[Kettlewell]

In order:
#1 reason they used (too) short scope (sometimes that's by choice but others its just poor skill - but in either case it would be nice if the anchor could compensate).
#2 reason is they did not set the anchor (again, sometimes by choice because the bottom just does not allow a solid set and others just lazy or poor skill)
#3 reason the bottom was bad - anchor did not stay set
#4 reason is wind shifted and anchor did not reset
#5 reason the anchor was set in a good bottom but did not hold.

I basically agree with your list, though I might swap #3 and #4 in the order and I would change #5 to wind strength overpowered anchor holding ability. Yes, #5 is a relatively rare occurrence, but it is also the one where your boat and possibly life might be put in danger due to the wind strength. I suppose it can then be argued that setting ability and resetting ability have to come before holding power in the list, except I would argue that if you don't have the holding power for the wind what good is it if the anchor sets easily? We could go round and round on this.

I see this problem in practice all summer long in New England. Crowded harbor, everyone on short scope, but all are settled for the evening comfortably with their anchors dug in for the prevailing conditions. Then a big thunderstorm comes through just after dark with winds of 45 knots and half the fleet, that was well anchored a moment earlier, is blowing away across the harbor. Setting ability may have gotten them to the point they thought they were well anchored but holding power is what will keep them well anchored.

By the way, I do have a manual windlass but I try to haul the anchor in hand over hand as much as possible as part of my exercise routine and I have only very rarely anchored in 50 feet--more often it is 10 feet or less.

[GordMay]

Originally Posted by Kettlewell

GMac: I'm sure you're aware of Professor John Knox's Anchorwatch system sold in the UK that measures loads at anchor and can act as an anchor alarm. He's done some interesting testing on loads at anchor onboard ordinary-sized sailing yachts and they have proven to be much lower than the ABYC tables. He wrote some articles for Practical Boat Owner back in 2002 that described some of his findings, and some of his formulas. He also did some interesting testing on things like how much chain you should use and what changes in scope did to holding power. Not sure what he is up to these days, but someone introduced a Knox anchor in 2010, though I'm not sure it has ever made it into retail production ...

Will my Anchor Hold ~ by John Knox
PBO July & August 2002

Part 1 ➥ Catalogue - Tests and comparisons of anchors part 1 - Sea Tech and Fun - Spade - (Version JPG) - 1

Part 2 ➥ Catalogue - Tests and comparisons of anchors part 2 - Sea Tech and Fun - Spade - (Version JPG) - 1

KNOX ANCHORS (Pictured below) ➥ Knox Anchors - Home

Professor John Knox ➥ Knox Anchors - About The Inventor

ANCHORWATCH
A Brief Outline of the Theory of Anchoring
➥ ANCHORWATCH- THE THEORY OF ANCHORING

[estarzinger]

Originally Posted by Kettlewell

I see this problem in practice all summer long in New England. Crowded harbor, everyone on short scope, . . .and half the fleet, that was well anchored a moment earlier, is blowing away across the harbor. Setting

Well then we can at least agree that short scope holding should be an important part of 'your test'. And this factor does vary quite a bit depending on anchor design geometry. You will remember that in my personal testing, I did a short scope test on soft mud, in addition to the 'bad bottom' setting test.

(as an aside on beach testing - I think its perfectly reasonable to test below the high tide line where the beach is still wet/water saturated. I do not know enough about soil mechanics to know how different the holding in a dry beach above the high tide line would be).

I have to admit that we not so infrequently anchor on only 2:1 scope. I don't recommend it to others, but we do manage to get away with it, never have dragged, I suppose because of the size of our anchor and because we do carefully set it in the slow fashion you mentioned in your initial post.

My personal sense is that the testing metric 'holding power per lbs of anchor, when set in a good bottom' is a hold-over from the thinking when CQR and danforth were introduced and trying to move cruisers away from the old fisherman designs. It made (marketing) sense then because the fishermen designs did need to be really big to have sufficient holding. It was never mentioned (in the marketing) that in return for the extra holding/lb you were giving up the universal bottom and almost instant setting capability of the fisherman.

I am not sure that it makes that much sense any longer as the primary metric to choose between anchors. The excellent news for cruisers is that all the current main designs have really really good holding when well set with long scope in a good bottom. We have benefited from goo design improvements and all the current main anchor designs fit the 'long scope/well set/good bottom holding' bill just fine and IMHO, the remaining differences between designs in "long scope/well set/good bottom" holding are much less critical than the differences between designs in setting and short scope and bad bottom holding.

[Adelie]

Originally Posted by Auspicious

I've been working on a test protocol for a couple of years, something that would allow results from different tests at different times to be reasonably compared. I'm a little stuck on bottom composition. I've been looking for a civil engineer with experience in saturated soils to help with material characterization.

I'm still working on mechanisms to realistically evaluate reset on veering.

I think (don't know - need to test!) that different anchor designs may scale differently and so three sizes of each design should probably be tested. Results should be shown by projected area in addition to by weight.

If I can ever get the paper finished I'd like to present at SNAME, perhaps their Small Boats Symposium.

Ahh, a man after my own heart, rigor in testing.

The big problem I see is not the testing protocol, it is finding or creating repeatable conditions. Even if you find a nice big expanse of seabed that all appears uniform, you really don't know if it is uniform underneath or only for the top 1" that you can see. Assuming you do have a nice uniform area, after several tests you run out of undisturbed areas and have to test in seabedyou have already dragged an anchor thru.

Here's my idea:

200'x100'x8' basin, concrete lined, waterproofed and set into the ground, so perimeter is at ground level. Bottom is lined at 6" spacing with pipes drilled with holes at a 6" spacing.

Passing over the basin is a truss, supported on rail-type cars at each end. This is the prep and observation car.

Running in a semicircle around the basin is a widegage railroad type track, possibly with a 3rd rail towards the basin from the 1st two but tilted away from the basin or with the rail horizontal. Riding on the track is a rail-type car counterweighted so it won't tip over due to a high side load towards the basin. To get the desired scope/angle of lead there may need another radiused track to carry an adjustable height frame. Mounted on the test car would be a hydraulic winch and load cell.

To prep the basin you would place 7' of the soil of your choice in it, then fill to the top with water. Now you have an underwater seabed. Air or water is pumped thru the pipes to thouroughly agitate the bottom. Now you have a uniform underwater seabed that can be recreated after each round of testing.

Next there are several options I see. One is to age the seabed, 1d, 1wk, 1yr, don't know. Aging gives particles in the soils time to settle into position becoming firmer. I would want to do a number of rounds of testing at various ages to see what changes. Probably different soils will react differently.

Another option is to vibrate the seabed in place using concrete vibrators (used for consolidating fresh concrete in forms) at a 6", 12" or 18" grid across the whole basin. This can be accomplished from the Prep Truss. Once again I expect different soils to react differently. Aging and vibrating may give different results, so you could need to test a number of different combinations.

Now we start testing. Drop an anchor and pull with a line and/or chain led to the test car.

Given that the test car is on a radiused track the pull can be maintained on an anchor as the load is swung thru an arc.

Probably the best thing to do is have a number of basins so that testing can be ongoing in one basin while the others are prepped and aged. Also you can have multiple soil types that are maintained for future testing as new anchor types are developed or new testing protocols and all the results are backward compatible with previous testing and anchor types.

But you are saying, this will cost a metric butt-load of money, for land and equipment, and labor, and facilities and construction, taxes, maintenance for years, etc. True, but good science costs and now you have repeatable conditions which is the beginning of good science.

But you argue this doesn't represent real world conditions. True again, but most science doesn't test true real world conditions, it tests one or several idealized conditions representing a range of real world conditions. Using that data it make predictions about real world results. Then it goes out and tests real world conditions to see how it compares to the idealized testing. Lots of unexpected significant variables emerge from such testing.

In the short run this setup would allow comparisons among a bunch of different anchor types in a variety of conditions. Even jerky cycling loads could be created with the right equipement.

[Kettlewell]

Evans, I think we're arguing about which comes first, the chicken or the egg, but in reality we can have both chickens and eggs at the same time. In the process of testing holding power we'll also find out how well the anchor sets. In practical terms, the current wave of anchor designers seems to be doing a good job on both of these fronts. From all that I read people are both exclaiming about the setting ability and the holding power of the "new generation" anchors.

Your point about problematic bottoms is harder to address. Your approach appears to be to get an anchor that is extra heavy and happens to have the right shape to dig into very hard and rocky bottoms, even if that anchor is relatively inefficient in more typical mud and sand bottoms. I would argue that weight is not a cure all though. I have been in areas where heavy anchors just would not set, but I was able to get a small, sharp Fortress to penetrate the bottom. Similarly, there are some weedy bottoms, and some very soft bottoms, where lightweight Danforth types provide the best holding. On the other side of the coin, I once rode a big blow in Key West with nothing but the point of a CQR anchor fouled on some underwater obstruction--as long as the wind blew strong I was well hooked, but after the blow the anchor came right up to the surface with nothing on it. Once when kedging off some rocks we used a big Luke folding yachtsmen style anchor which was able to wedge in between various boulders. In the Bahamas I have anchored by hand placing lightweight Fortress or Danforth anchors up on sandbars on either side of the scoured bottom where the deep water was. In one case, the anchors were high and dry up on land at low tide, but we were anchored safely in the middle where no one else could get a bite. So, I do think there is some argument to be made still for carrying a variety of anchors that can do different things in different bottoms.

[klem]

How about testing mooring anchors as well. There have been a few tests but the data is pretty poor and a lot of boats are lost to dragging moorings. I know this wasn't the intent of the original post but they are a form of anchor and an awful lot of people don't use them properly just like their ground tackle. I would love to see a real scientific comparison of a mushroom, a deadweight, a pyramid and a screw type anchor in several common bottoms.

[capngeo]

Originally Posted by Adelie

Ahh, a man after my own heart, rigor in testing.

The big problem I see is not the testing protocol, it is finding or creating repeatable conditions. Even if you find a nice big expanse of seabed that all appears uniform, you really don't know if it is uniform underneath or only for the top 1" that you can see. Assuming you do have a nice uniform area, after several tests you run out of undisturbed areas and have to test in seabedyou have already dragged an anchor thru.

Here's my idea:

200'x100'x8' basin, concrete lined, waterproofed and set into the ground, so perimeter is at ground level. Bottom is lined at 6" spacing with pipes drilled with holes at a 6" spacing.

Passing over the basin is a truss, supported on rail-type cars at each end. This is the prep and observation car.

Running in a semicircle around the basin is a widegage railroad type track, possibly with a 3rd rail towards the basin from the 1st two but tilted away from the basin or with the rail horizontal. Riding on the track is a rail-type car counterweighted so it won't tip over due to a high side load towards the basin. To get the desired scope/angle of lead there may need another radiused track to carry an adjustable height frame. Mounted on the test car would be a hydraulic winch and load cell.

To prep the basin you would place 7' of the soil of your choice in it, then fill to the top with water. Now you have an underwater seabed. Air or water is pumped thru the pipes to thouroughly agitate the bottom. Now you have a uniform underwater seabed that can be recreated after each round of testing.

Next there are several options I see. One is to age the seabed, 1d, 1wk, 1yr, don't know. Aging gives particles in the soils time to settle into position becoming firmer. I would want to do a number of rounds of testing at various ages to see what changes. Probably different soils will react differently.

Another option is to vibrate the seabed in place using concrete vibrators (used for consolidating fresh concrete in forms) at a 6", 12" or 18" grid across the whole basin. This can be accomplished from the Prep Truss. Once again I expect different soils to react differently. Aging and vibrating may give different results, so you could need to test a number of different combinations.

Now we start testing. Drop an anchor and pull with a line and/or chain led to the test car.

Given that the test car is on a radiused track the pull can be maintained on an anchor as the load is swung thru an arc.

Probably the best thing to do is have a number of basins so that testing can be ongoing in one basin while the others are prepped and aged. Also you can have multiple soil types that are maintained for future testing as new anchor types are developed or new testing protocols and all the results are backward compatible with previous testing and anchor types.

But you are saying, this will cost a metric butt-load of money, for land and equipment, and labor, and facilities and construction, taxes, maintenance for years, etc. True, but good science costs and now you have repeatable conditions which is the beginning of good science.

But you argue this doesn't represent real world conditions. True again, but most science doesn't test true real world conditions, it tests one or several idealized conditions representing a range of real world conditions. Using that data it make predictions about real world results. Then it goes out and tests real world conditions to see how it compares to the idealized testing. Lots of unexpected significant variables emerge from such testing.

In the short run this setup would allow comparisons among a bunch of different anchor types in a variety of conditions. Even jerky cycling loads could be created with the right equipement.

You wouldn't by chance work for the Federal Government would you?