Bigger is better, part 2

[Pelagic]

The kid in me just keeps substituting penis for chain in this thread. Keep it going...I am having a good giggle

[Andrew Troup]

JonJo

I guess my point is this:

The only time I want my anchor to start burying the chain is in a bottom so soft that it won't notice the size of the chain.

In better holding, to me it's just a pain if it keeps going for China. At some future time I will have to shag about getting the damn thing back.

And because my anchor is sized for strong winds and bad holding, the extra holding power in *good* holding, possibly an order of magnitude more than I need, is of academic interest.

I can see merit in your argument for someone who is always anchoring in good holding, and who wants to use the smallest, lightest gear possible.

In which case it would seem worth going to the logical conclusion and fitting a short wire leader between anchor and chain, and a really compact (poured-socket?) clevis, deep but thin, to closely embrace the anchor shank without the bulk of a shackle.

As for your other suggestion (or perhaps it was more like musing?):

I personally think G7 chain is already a bridge too far. I consider it too fragile in respect of longer term risks such as stress corrosion cracking, notch sensitivity (ie lack of toughness), intolerance of unfair loads or damage, regalvanising embrittlement risks et al.

So I guess it goes without saying I think the notion of using even higher tensile material for anchor rodes to be scary and misguided.

No doubt someone will push the envelope further, but I think there are good reasons why old salts leave that process to others, especially in respect of what is effectively the last line of defence any boat can have.

There is almost always a tradeoff, for a given class of material, between increased performance and reduced reliability, and/or tolerance of misadventure and accidental abuse.

[JonJo]

Andrew,

Early Rocna shanks were 800 MPa steel as are all Supreme, Boss and Excel shanks. A number of people have embraced G7 galvanised chain. Crosby, gal HT shackles are of a G80 specification (much promoted on this forum). I agree re-galvanising is an issue, but can you quote me any stress corrosion issues, failures due to notch cracking (this lack of toughness), intolerance of unfair loads - surely anchors are the prime location and the evil bugbear, hydrogen embrittlement.

I know imported construction bolts might suffer - but a G80 Crosby shackle, a G70 ACCO chain and an ASTM 514a Manson shank. I know the fears - where is the evidence?

Americans in particular seem to have adopted G43 chain without blinking an eyelid - its a Transport Chain (as is G70) - what is the connection, what is the similarity between a Transport Chain and an Anchor Chain, why is a lifting G80, G100 (might I mention G120?) chain not a better association?

We are potentially offered increased performance, you say at the risk of reduced reliability - where is the evidence that reliability has been sacrificed?

If a chain, say G70, (or G7), G100, is made by a reliable and reputable company with a record of good quality control and is correctly Proof Tested, 1m lengths off each end of the rode tested for ultimate break test - what is the issue (these 3 tests would cost $30 here in Oz). G70 chain has been used in the Transport industry for years - are there really many problems with notch cracking or stress corrosion?

Your views on anchors setting too deep are interesting - but not for this thread, though I bow to the drifters (of which I can easily join and participate)

I'm not a metallurgist - I'm not doubting your comments, I am simply requesting an education.

The deeper diving anchor, using a thinner rode, has lacked previous airing but one should not trivialise - but your comments - which relate directly to safety are more serious and meritorious comments.

Jonathan

[sailorboy1]

Originally Posted by Pelagic

The kid in me just keeps substituting penis for chain in this thread. Keep it going...I am having a good giggle

And how does that out for smaller/thinner for you

[DoubleWhisky]

Originally Posted by mpatter894

just thought I would mention it because there were some people debating the theory of getting a bigger anchor and using lighter chain, you of all people should know that's not a good idea or maybe you dont sailing around the coast of Poland around Gdansk where the wind can really howl

Hmmmm.... My own waters... And I really do know worse places, even here, in Poland...

[Andrew Troup]

Originally Posted by JonJo

Early Rocna shanks were 800 MPa steel as are all Supreme, Boss and Excel shanks.

Contrast the cross sectional area of an 8mm chain link with that of the shank of a 25kg anchor above and below the eye: consider further that the stress resulting from a given load on the former is always a combination of bending and tensile, half of which is taken through a butt weld. Lateral bending loads on an anchor shank are invariably "softened" by the flexural freedom: they do not present as shock loads, so they do not make demands on the toughness of the material.

A number of people have embraced G7 galvanised chain.
I cannot confidently predict there will be failures, but I cannot confidently predict there will not. It's too soon to judge this based solely on what has already happened, because the problematic issues I am concerned about appear over time.

The primary job of engineers is to predict and avert future failures, rather than sit on our hands waiting for them to happen so we can explain them.

Crosby, gal HT shackles are of a G80 specification (much promoted on this forum). I agree re-galvanising is an issue, but can you quote me any stress corrosion issues, failures due to notch cracking (this lack of toughness), intolerance of unfair loads - surely anchors are the prime location and the evil bugbear, hydrogen embrittlement.

See above re anchors operating at lower stress levels and minimal shock load; and re time for these issues to appear. Shackles are not welded, hence no problems with the microstructure in the heat affected zone, and also statistics are in their favour. If one shackle (or anchor) in one thousand fails, that's not a big deal, but if one chain link in one thousand does, that's a huge deal.


...

Americans in particular seem to have adopted G43 chain without blinking an eyelid - its a Transport Chain (as is G70) - what is the connection, what is the similarity between a Transport Chain and an Anchor Chain, why is a lifting G80, G100 (might I mention G120?) chain not a better association?
...

Lifting chains are routinely inspected for nicks and condemned out of hand for the slightest indentation. In best practice, they are checked before every heavy lift.

Crucially, they are not permitted to operate in a corrosive medium or atmosphere: eg this from PWB: "Herc-Alloy chains and fittings should not be used in ... corrosive environments."

Moreover, anchoring loads are unquantifiable
Lifting chains are *never* used for unquantified loads, they must be condemned if they experience even a single shock load, and they are not allowed uncontrolled contact with jagged and rusty metal such as you can get in the seafloor environment. Especially in the higher grades, they are fragile flowers, like the components of a high performance racing engine.

Substitution of a lighter, "stronger" chain is double jeopardy: the shock loads will be further increased by the reduction in energy uptake from chain catenary and viscous drag, so that if a snubber fails, the chain is less likely to survive until the snubber can be reinstated.
Remember too that a 1mm deep indentation has a much greater effect on the strength of a 6mm diameter link than an 8 or 10mm link, regardless of the material.

Furthermore, any chain will sometimes have to be retrieved under snatching conditions, when the snubber is no longer available, the catenary is nonexistent, and the chain, being short, has little inherent linear elasticity. It takes only one shock to snap a chain.
...


I'm not a metallurgist - I'm not doubting your comments, I am simply requesting an education.

Thanks for your straightforward declaration. I am voicing reservations rather than being in a position to present a smoking gun case. And my reservations are about what I would personally be prepared to contemplate.

I have intimate and painful experience of the reductions in performance under shock loadings when a high tensile steel is substituted for a medium tensile steel. These reductions in ductility and hence toughness are the bane of an engineer's life, and arise from inevitable changes in the way discontinuities in the microstructure propagate and accumulate.

These reductions in performance are handled in the general case by more careful detail design and manufacturing (including allowing higher factors of safety), more careful quantifying and limitation of transient loads, more careful control of the operating environment (including handling precautions), more stringent and regular inspections ...

... but the only one of these which is applicable to the anchor rode instance, is careful manufacturing.
However, ALL reputable chain is carefully manufactured, so this hardly constitutes a mitigating plea.

Corrosive environments compound the problem by assisting the tips of the micro-discontinuities to propagate. Galvanising will assist here so long as it remains intact , so I guess if you do not range far, stay alert for nicks, and replace your chain often, you will probably be fine with G70, conservatively sized. I would never re-galv it, though.

The notion of pushing still further, into the realms of ultra-high tensile chain, fills me with the exact opposite of enthusiasm.

This is a much bigger topic than I am able to do justice to, but I hope something I have written will help someone to take a more measured view rather than an excessively hopeful or even wishful one.

I'm struck, looking at what I've written, by the asymmetry between the slim upside, and a potentially drastic downside, from a decision to use very high tensile chain in this application.

[Pelagic]

Originally Posted by Andrew Troup

I'm struck, looking at what I've written, by the asymmetry between the slim upside, and a potentially drastic downside, from a decision to use very high tensile chain in this application.

Back to being serious now.
Interesting stress analysis Andrew but one question?

When I think of the times ground tackle "chain" has been damaged it is often following a problem of retrieval.
After a major blow when a bight or knot of chain has entangled on a rock or cleft and you need to work it out @ short scope.

I think a larger.. more malleable chain would better survive those types of stresses without failure.. even though suffering damage rather than a smaller high tensile yet more brittle steel.

What is your opinion ?

[noelex 77]

Thanks Andrew it is good to get an engineers perscpective.

There has been some debate about the suitability of high tensile chain and this is a good topic to discuss on CF.

We are entering new ground (and I am one of the Guinea pigs ) with G7 chain and any failings are going to take while to show up.

Most people downsize chain size when going to G7, but some would argue 10mm G4 would be acceptable for my boat. Are there situations where 10mm G7 would break more easily than 10mm G4 ?

[Andrew Troup]

Originally Posted by Pelagic

Back to being serious now.
Interesting stress analysis Andrew but one question?

When I think of the times ground tackle "chain" has been damaged it is often following a problem of retrieval.
After a major blow when a bight or knot of chain has entangled on a rock or cleft and you need to work it out @ short scope.

I think a larger.. more malleable chain would better survive those types of stresses without failure.. even though suffering damage rather than a smaller high tensile yet more brittle steel.

What is your opinion ?

I agree with your supposition, Pelagic. For a Grade L chain to be substantially weakened, the damage would be very evident, whereas a small nick could substantially weaken a high tensile chain.

For a vessel spending long periods in remote areas (particularly in places where damage from rocky bottoms is likely) a big advantage of using larger, lower tensile chain is that severely damaged links can be cut out and a hammerlock joiner

(or, in larger sizes, 16mm and up, a Kenter or similar joining link, which will pass over the chainwheel more readily)

can be used to rejoin the chain without sacrificing strength. This is because the aperture and envelope are both larger than in "same strength" high tensile chain.

A hammerlock would need to be kept under close inspection though, dried and oiled.

Back in civilisation, a better substitute could be arranged for a hammerlock, perhaps a custom joining link in 2205, carefully TIG welded in situ

[Andrew Troup]

Originally Posted by noelex 77

Thanks Andrew it is good to get an engineers perscpective.

There has been some debate about the suitability of high tensile chain and this is a good topic to discuss on CF.

We are entering new ground (and I am one of the Guinea pigs ) with G7 chain and any failings are going to take while to show up.

Most people downsize chain size when going to G7, but some would argue 10mm G4 would be acceptable for my boat. Are there situations where 10mm G7 would break more easily than 10mm G4 ?

That's a burning question, one I've thought quite a lot about, but I simply don't know.

It's certainly a possibility. Chain manufacturers are notoriously close-fisted about their exact material spec, so I'm forced to infer what they're using. Even if I guess correctly (which I probably can) there are so many compounding factors in the uncontrolled and aggressive marine environment that I don't have any confidence in a pat answer.

I realise I'm not being very helpful, so I'll go out on a limb and speculate wildly:

Based on what is frankly guesswork, I think the dilemma you pose is unlikely for G70 vs G43, but I think it's eminently possible for G80 vs G43.

G70 is essentially made from the same steel as G43, except that it's heat-treated (hardened and tempered) after forge-welding. Whereas G80 is a different, higher strength alloy.

And the dilemma you pose is, I think, a virtual certainty for G100 vs G43 in certain real-world anchoring situations (even neglecting a problem JonJo doesn't mention: you cannot galvanise G100)

The falloff in impact strength for a particular composition of alloy steel as the ultimate tensile strength pushes up into this region is remarkable, even in the absence of compounding factors like a mildly corrosive environment (which is the worst risk factor for stress corrosion cracking)

I attach a graph I prepared from the mechanical properties of Interlloy 4340, showing that in pushing the tensile strength upwards by a factor of two (essentially by adjusting the tempering process), the penalty in terms of impact strength is a factor between five and six.

Another interesting question would be this:

Are there circumstances which might leave G30 (Grade L) stronger in relation to suddenly applied loads than the same size G43?

It seems possible to me that the answer here, rather surprisingly, could be yes. But I would have to do lots of tests, and I have no plans to do that.

[JonJo]

Thanks Andrew,

Grade L, that Andrew references, is the same as, similar to G3 or G30.

If this is not a big ask, in laymens terms can you explain ductility.

In order to show my ignorance:

My understanding is that all steels have some elasticty, they stretch and when the load is released they return to the shape (or in terms of a chain) length that they originally had. At some point they yield and deformation becomes permanant - they stretch permanently until they fail.

Maybe you can correct and translate this into more acceptable language.

But you are suggesting there are major differences between this elastic stretch and permanent stretch when one looks at, say, G3 and G7 chain. How do these differences manifest themselves for chains of the same size, how much elastic stretch in a G3 vs a G7 and how much deformation stretch again between a G3 and G7.

And as its the only bit we want to work with, the elastic stretch - does its ability, either G3 or G7 have any meaningful 'use' or is it so small and insignificant vs whatever might be left of catenary (when under severe load).

Jonathan

edit, Andrew our posts crossed.

I intentionally did not mention galvanising of G100 - it is a bridge too far, today, but there are a lot of people looking at it.

A supplementary question - What might the difference in characteristics be for a Gal G8 and a Gal G8 chain?

[JonJo]

Andrew,

Again you might speculate:

But Americans have a G43, which is or was a transport chain (as is G7). Europeans have G4, which presumably is different to G43, or it would be called G43 and Australians have Grade P

What might be the differences between American, European and Australian G3 (or Grade L), American G43, European G4 and Australian Grade P.

Jonathan

Why have Australians gone out on a limb with a completely different nomenclature to the rest of the world?

[Andrew Troup]

Jonjo:

Elasticity is a property of all materials, not just steel.

What's happening is that the individual atoms are changing their spacing temporarily and without any jostling or rearrangement. The resistance to elastic stretch (or bending) reflects the strength of those bonds, and is highly predictable. It's virtually the same for ALL steels.

Ductility is not a property of all materials. Some materials (most metals) are ductile under the right conditions. Glass and toffee are classic non-ductile materials.

In a material behaving in a ductile way, when the elastic stretch rises above a level where the attractive forces between atoms are equal and opposite to the applied load, the atoms rearrange themselves in a way which redistributes the load better. This involves the shape of the item changing permanently, even when the load is removed.

In a material behaving in a brittle way, this rearrangement cannot happen, the interatomic bonds are ruptured and the material breaks.

And most ductile materials have a transition temperature, above which they will behave in a ductile way, and below which they will be subject to brittle failure. It was recently discovered that the hull plating of the Titanic would have been below this temperature when the iceberg scraped alongside, because the metallurgy of steel at that time was not well understood or controlled. Samples brought up from the wreck showed that the transition temperature was unacceptably high (by modern standards), and certainly higher than the ambient temperature on the night of the sinking.

Incidentally, performance at low temperature starts to be a problem for G100 chain ...
the problems have not so much been solved by modern metallurgy, as 'kicked upstairs'; it's a problem which these days largely bedevils high- to ultra-high tensile steel alloys.

Ductility is one way of achieving toughness (the ability to absorb suddenly applied loads without rupture) Think of a lead-faced hammer.

Other ways to achieve toughness involve trapping the dislocations arising from the rearrangement mentioned above before they can become long enough and organised enough to give rise to cracking. Animal bones, fibre reinforced plastics, and wood, are three different approaches which are outstandingly successful in doing this. However introducing toughness-enhancing mechanisms like this into metal microstructure remains an elusive and currently unattained dream.

I said at the top that all steels exhibit the same resistance to increasing load, measured by how much they lengthen for a given load, during the elastic phase.

What is different for stronger alloys is that they keep resisting for longer. This means the ultimate tensile strength is greater. However when the tensile stress reaches this value, the stronger the alloy, typically the less ductility they exhibit, and so they are unable to absorb further energy by internal rearrangement.

It is perfectly possible to heat-treat such alloys to the point where they behave like glass: very high tensile strength, but very little toughness.

And this is the extreme illustration of a tradeoff which engineers spend so much of their effort trying to manage, but are never able (as of this writing) to avert altogether.

[Andrew Troup]

Jonjo

regarding chain designations:

I have no knowledge of how the current situation came about. I was under the impression that (for example) G7 is an abbreviation of G70 (which in turn is an abbreviation of G700, for 700Mpa nominal) because it's more legible when stamped on a small chain link.

I had assumed that "Grade L" and suchlike were PWB's proprietary designations, which I think were around before the likes of G30, but I could be entirely off beam.

[JonJo]

Andrew,

Thanks again,

My interpretation of one part of what you posted suggests that:

The ratio between the elastic phase and ductile phase varies dependent on the tensile strength of the steel. High tensile steels have less ductility but greater elasticity (as a proportion of total deformation).

So a G3 chain would have greater ductility than a G7 (or G8 chain).

And/or A G7 or G8 chain would have greater elasticity (than a G3 chain - which one might have thought as being advantageous). So for the same strength of chain (so we would have a smaller G7 chain) the G7 will have 'better' or higher elasticty - it will stretch more and have an ability to return to its original size - but when permanent deformation occurs it will fail soon thereafter. The suggestion is that G7 could sustain a higher Proof Load test, higher than the current 2 x WLL, than G3 and it should be possible better testt the welds of the G7?

But earlier you drew a distinct difference between a G7 chain, which is a Q&T G43 chain and a G8 chain which is an alloyed chain (but also Q&T). What might the differences be between a G7 and a G8 chain (which will be even bigger for a G7 vs a G100 chain) - in terms of elasticty and deformation. Will the G100 be even more elastic,

Crudely quantifying

If G3 fails at 20% stretch, then 10% might be elastic and 10% (so 20% total) might be permenant deformation, G7 might be 15% and 5% (and G100 18% and 2%). I'm assuming same strength, but smaller and smaller chain.

If this uneducated 'summary' is correct then why is not this high elasticity good on the further assumption that WLLs are similar and are not exceeded. I have missed something, somewhere.

Basically Noelex 10mm G7 should be able to absorb more stress than the same sized G4, or G43 - its got more inherent elasticty.

Jonathan