Owner wins claim against UK builder

[Poecheng]

Hi JFM,

I am not exp in the CPR but just enough to know some of the consequences.

I was tempted to agree with you about that last paragraph so went back and read it. I think the reason is that the case is unappealable because there were no issues of law that were contentious and the appeal (I think) would have to be on a point of law rather than fact. So I expect J felt on safe ground to forget part two as there could effectively be no legal point appealed.

It got me reading the penultimate paragraph about the 40% and it's worth a re-read. I think I was too kind . Shimmel is the name of the expert and he suggests a working likelihood of 80% efficiency. Then he concedes that he has no expertise in the area at all. So his evidence is (effectively) dismissed and the judge goes straight to accepting the 40% in the manual. Well that is no more accurate than Shimmel and may be a whole lot less accurate - Shimmel may not have proven expertise but that doesn't mean he is wrong. He probably knows exactly what he is talking about but hasn't particularly studied it to describe himself as an expert in that particular area.
We know J actually went to 40% as I have done the calculations
Pump 1 and 2 = 1500 and 3700 g/ph
x 40% = 2080g/h
/60 =34.66g/m
convert to lt = 157.56 litre/min (and the J worked with 158 l/m)

Might be worth a squint at the manuals because I assume Rule are a US firm. If the g/h figures were US Gallons then it equates to 131 litre/min.
I assume they knew the difference (or the manual states it) between imperial and US Gallons? or maybe I am being too kind again.

It is odd though - but typical - find a reason to reject the person to then reject their evidence AND assume it cannot be correct. You may not accept it from them but it doesn't mean it cannot be correct. Yet you see the judge flip from 80% to 40% for no reason other than the man wasn't an expert in pumps. I think if my case depended on it I might have got some empirical data!

Just re-read it again:
Claimant makes witness statement saying 6" crack
In opening the case (Ie the opening speech to the judge at trial) counsel says that is not correct and the witness statement should not have said that. No one else says it as the diver's emails etc are not admitted. Therefore there is no evidence at all it was 6". We only know it was 30" later.
But they try to prove it was 6" or thereabouts by the known fact that the floorboards in the aft cabin were not flooded and therefore the pumps kept up with the leak. As you don't know the leak you look at the actual capacity of the pumps. But if that calculation is bollocks - being arbitrarily based on 40% - then the evidence about the size of the leak is suspect. Greater capacity pumps could keep up with a much greater breach of the hull. Too tired to work out who this helps but Humphreys (Claimant's expert) seems in §40 to be seeking to keep the size of the breach to 6" or less.
Gotta get on with work now :encouragement:

 

[JumbleDuck]

The 40% figure in the manual was a "might", written by a manufacturer who would want to cover off product liability etc.

The performance curves in the manual, from which I presume the 40% derives, are nothing to do with product liability. They are there to allow a designer to predict the performance of the pump in a particular installation. As I wrote, the larger pump is shown as working at 40% of maximum (zero head) capacity at ~13' discharge head and the smaller at 7'. That would be either to the highest point on the discharge side or, if the pipe was smallish, to the skin fitting if a syphon effect started up. Without a pipework diagram for the boat it's really not possible to say whether 40% was reasonable or unreasonable.

And if that doesn't convince, then you have the strange uncontested assumption of a 10m crack width for a 6 inch crack.

Agreed. Incidentally, it is extremely unlikely that a crack of that sort would open by the two edges moving apart in the plane of the material ... any opening is much more likely to come about from perpendicular movement - forming a scoop rather than a pair of lips, if you see what I mean. Technically it would mean Mode III rather than Mode I cracking at the ends. I suspect that GRP is much less tough in Mode III because there are not many fibres running in the normal direction.

Or, to put it another way, if you have a sheet of GRP with a crack going in from one side, its easier to extend the crack by pulling it apart like a pair of scissors than by keeping it flat. Handy diagram from Wikipedia at http://en.wikipedia.org/wiki/File:Fracture_modes_v2.svg

For the claimant to win it has to be the case on B-of-P that (a) the crack wasn't initially 30 inches, AND (b) the crack was a 6 inch slit sprung open to 10mm wide not 7mm with clean Bernoulli flow AND (c) the bilge pumps were working only 40%. A failure to get over ANY ONE of those 3 hurdles means Pearl wins, yet Pearl didn't call a challenge at any of the hurdles.

If the impact force was enough to give a 6" crack it may well have been enough to give a 30" crack. That's how fast fracture works: once you reach the critical stress intensity factor the crack keeps on growing as long as the load is applied, and in fact can keep growing as the load reduces because the bulk stress increases as the amount of material decreases. In other words, trying to use the length of a crack to predict the force that caused it is not sensible. Other things, like tip radius, matter more.

All this is, of course, high speculative without seeing the actual engineering evidence produced.

 

[JumbleDuck]

We know J actually went to 40% as I have done the calculations
Pump 1 and 2 = 1500 and 3700 g/ph
x 40% = 2080g/h
/60 =34.66g/m
convert to lt = 157.56 litre/min (and the J worked with 158 l/m)

It would be most odd if the two pumps reached 40% simultaneously, since they require very different discharge heads to get that reduction. In fact if the 14A was at 40% the 02 would not be pumping at all, since its zero-flow head is 12'. Or, to put it another way, 40% for the two pump system would be reached, absent some very peculiar plumbing, for a significantly lower discharge head than for the 14A alone. Calculation of the head in question is left as an exercise for the reader.

 

[Poecheng]

It would be most odd if the two pumps reached 40% simultaneously, since they require very different discharge heads to get that reduction. In fact if the 14A was at 40% the 02 would not be pumping at all, since its zero-flow head is 12'. Or, to put it another way, 40% for the two pump system would be reached, absent some very peculiar plumbing, for a significantly lower discharge head than for the 14A alone. Calculation of the head in question is left as an exercise for the reader.

Does that mean one of them is a "lowloader":encouragement:

 

[jfm]

I think the reason is that the case is unappealable because there were no issues of law that were contentious and the appeal (I think) would have to be on a point of law rather than fact.

AFAIK this isn't one of the cases where statute demands appeal on law only (like employ tribunals and tax, for example). But I think the procedure rules say it needs to be on a point of law "generally". Possibly worth a shot then, on grounds of manifest and severe error of fact, plus possibly (though I have no idea if true) on grounds of materiality to financial health of loser and employment of its work force. All depends on the facts, which only Pearl know

It got me reading the penultimate paragraph about the 40% and it's worth a re-read. I think I was too kind . Shimmel is the name of the expert and he suggests a working likelihood of 80% efficiency. Then he concedes that he has no expertise in the area at all. So his evidence is (effectively) dismissed and the judge goes straight to accepting the 40% in the manual. Well that is no more accurate than Shimmel and may be a whole lot less accurate - Shimmel may not have proven expertise but that doesn't mean he is wrong. He probably knows exactly what he is talking about but hasn't particularly studied it to describe himself as an expert in that particular area.
We know J actually went to 40% as I have done the calculations
Pump 1 and 2 = 1500 and 3700 g/ph
x 40% = 2080g/h
/60 =34.66g/m
convert to lt = 157.56 litre/min (and the J worked with 158 l/m)

Might be worth a squint at the manuals because I assume Rule are a US firm. If the g/h figures were US Gallons then it equates to 131 litre/min.
I assume they knew the difference (or the manual states it) between imperial and US Gallons? or maybe I am being too kind again.

It is odd though - but typical - find a reason to reject the person to then reject their evidence AND assume it cannot be correct. You may not accept it from them but it doesn't mean it cannot be correct. Yet you see the judge flip from 80% to 40% for no reason other than the man wasn't an expert in pumps. I think if my case depended on it I might have got some empirical data!

All completely agreed. you are sensing my concern that Pearl haven't got the outcome here they might have got.

Just re-read it again:
Claimant makes witness statement saying 6" crack
In opening the case (Ie the opening speech to the judge at trial) counsel says that is not correct and the witness statement should not have said that. No one else says it as the diver's emails etc are not admitted. Therefore there is no evidence at all it was 6". We only know it was 30" later.
But they try to prove it was 6" or thereabouts by the known fact that the floorboards in the aft cabin were not flooded inserted by jfm: and the sink rate from the photos and therefore the pumps kept up with the leak. As you don't know the leak you look at the actual capacity of the pumps. But if that calculation is bollocks - being arbitrarily based on 40% - then the evidence about the size of the leak is suspect. Greater capacity pumps could keep up with a much greater breach of the hull. Too tired to work out who this helps but Humphreys (Claimant's expert) seems in §40 to be seeking to keep the size of the breach to 6" or less.
Gotta get on with work now :encouragement:

YES YES YES! That's what I've been saying all along!!! It helps Pearl. The claimant wanted to show the crack was only 6 inches and Pearl wanted to show it was bigger. If a 6 inch crack was more or less kept up with by 40% of the pumps, then if the pumps (by empirical evidence for example) were actually running at 60% then the crack was 9 inches ish, and that destroys the already flimsy claim that the crack size was 6 inches. The claimants had onus of proof to show the crack was small, and would have failed if the pumps were shown to be working materially better than 40%

 

[Tranona]

If the impact force was enough to give a 6" crack it may well have been enough to give a 30" crack. That's how fast fracture works: once you reach the critical stress intensity factor the crack keeps on growing as long as the load is applied, and in fact can keep growing as the load reduces because the bulk stress increases as the amount of material decreases. In other words, trying to use the length of a crack to predict the force that caused it is not sensible. Other things, like tip radius, matter more.

All this is, of course, high speculative without seeing the actual engineering evidence produced.

Perhaps even more complicated than that. If I understand the construction as described correctly, then the P bracket palms are set in a moulded recess in the bottom panel of the hull, and the 7mm thickness of laminate is at the edges of the recess (para 7 of the judgement). One can imagine a 6" crack in this weak point from the forces acting on the palm, but for the crack to be 30" it must have spread to the 20mm thick area of the panel. The length or width of the palm is unlikely to be more than 6" so the moulded recess is unlikely to be much greater than that either.

I was surprised that in 31 a) both parties agreed that the " final crack" was 30" in length but there was no explanation as to how and why it grew from a small crack (presumably in the thin area of the recess) at the point of grounding into a crack five times as long and extending into the main panel, which by their own agreed calculations requires a force more than 5 times as great.

One can see why they wanted to make the claim under contract as the breach was admitted. Much easier than trying to prove fitness for purpose on the basis of the data available. They would have to show that the boat's design and construction should be capable of surviving a grounding of such force as to crack the hull. As noted in para 8 there is nothing in the RCD that requires the laminate to be strong enough to withstand a grounding. Instead of the arguments being about the severity of the grounding (and therefore the forces acting on the laminate), they would be about whether it should be able to withstand any grounding.

 

[jfm]

The performance curves in the manual, from which I presume the 40% derives, are nothing to do with product liability. They are there to allow a designer to predict the performance of the pump in a particular installation.

100% agree. we are at crossed purposes on this point (forums eh) but no matter

As I wrote, the larger pump is shown as working at 40% of maximum (zero head) capacity at ~13' discharge head and the smaller at 7'. That would be either to the highest point on the discharge side or, if the pipe was smallish, to the skin fitting if a syphon effect started up. Without a pipework diagram for the boat it's really not possible to say whether 40% was reasonable or unreasonable.

Yup, I see all that. We are violently agreeing: everything you write says the judge and Shimell made a big error in using the 40% factor, to Pearl's disadvantage. I'm saying the same. A screw up, on a foundation point.

Agreed. Incidentally, it is extremely unlikely that a crack of that sort would open by the two edges moving apart in the plane of the material ... any opening is much more likely to come about from perpendicular movement - forming a scoop rather than a pair of lips, if you see what I mean. Technically it would mean Mode III rather than Mode I cracking at the ends. I suspect that GRP is much less tough in Mode III because there are not many fibres running in the normal direction.

Yep, I get all that and sounds very correct. I guess we have to assume this was factored in because it was undisputed that a 70kN force would crack the hull @7mm and 400kN@20mm, with no information as to what mode of fracture occurred at those values other than that the eminent High Modulus did the numbers. Any argument, if sustainable, that "the kN figures were wrong because they assume a Mode I crack instead of Mode III" is of course v helpful to Pearl so this is yet another thing about which one wonders was it done correctly (but it prob was if High Mod did it, and er were asked the right question!).

If the impact force was enough to give a 6" crack it may well have been enough to give a 30" crack. That's how fast fracture works: once you reach the critical stress intensity factor the crack keeps on growing as long as the load is applied, and in fact can keep growing as the load reduces because the bulk stress increases as the amount of material decreases. In other words, trying to use the length of a crack to predict the force that caused it is not sensible. Other things, like tip radius, matter more.
.

Yup, I get all that. Thanks, good stuff, I hadn't thought that through but I can see it is very correct (alas I'm becoming more lawyerish and less engineerish as get older :/). You get the impression this wasn't in the expert evidence but I don't know for sure. This point helps Pearl too: the upshot is that you cannot deduce force from crack length (you are more likely to deduce time for which force was applied, which is wholly irrelevant to this case). That being the case, as the totality of the evidence fro force applied was crack length, there is no good evidence as to the force applied, and therefore claimant has failed to show that the grounding would not have sunk the boat if built to 20mm, ergo claimant loses.

As each new point of discussion emerges in this, and your piece immediately above is a fine example, we see another reason why Pearl shouldn't have lost. Thanks for sharing your insights

 

[jfm]

One can imagine a 6" crack in this weak point from the forces acting on the palm, but for the crack to be 30" it must have spread to the 20mm thick area of the panel. .

I agree all your post Tranona except the above. I think the correct spec of hull panel is generally 7mm, reinforced to 20mm in the area of the palm recess plus a bit beyond, say something like 400x400mm or whatever. The supposed spread from 6 inches to 30inches was therefore a spread along an entirely 7mm piece in the case of this boat that mistakenly didn't have the extra 13mm of laminate build up applied. I think!

 

[Tranona]

I agree all your post Tranona except the above. I think the correct spec of hull panel is generally 7mm, reinforced to 20mm in the area of the palm recess plus a bit beyond, say something like 400x400mm or whatever. The supposed spread from 6 inches to 30inches was therefore a spread along an entirely 7mm piece in the case of this boat that mistakenly didn't have the extra 13mm of laminate build up applied. I think!

Not according to the description in para 7 - which is why I was surprised this was not pursued in greater depth. It is described as being 20mm all over and the 20mm should have been maintained in the edges of the recess. There is no explanation as to how or why it ended up at 7mm at the edges (which presumably is where the crack occurred).

 

[JumbleDuck]

The claimant wanted to show the crack was only 6 inches and Pearl wanted to show it was bigger.

It's quite an intriguing situation really: the company who were defending themselves against a claim of shoddy construction wanting to claim that there was a huge amount of damage and the owner's insurers wanting to claim that it wasn't that bad.

 

[JumbleDuck]

Perhaps even more complicated than that. If I understand the construction as described correctly, then the P bracket palms are set in a moulded recess in the bottom panel of the hull, and the 7mm thickness of laminate is at the edges of the recess (para 7 of the judgement). One can imagine a 6" crack in this weak point from the forces acting on the palm, but for the crack to be 30" it must have spread to the 20mm thick area of the panel. The length or width of the palm is unlikely to be more than 6" so the moulded recess is unlikely to be much greater than that either.

Yup, good point. Simplistic models of stress, strain and toughness just don't cut it here.

 

[jfm]

It's quite an intriguing situation really: the company who were defending themselves against a claim of shoddy construction wanting to claim that there was a huge amount of damage and the owner's insurers wanting to claim that it wasn't that bad.

Yes there is plenty of harse about face generally with this one

 

[JumbleDuck]

As each new point of discussion emerges in this, and your piece immediately above is a fine example, we see another reason why Pearl shouldn't have lost. Thanks for sharing your insights

How central do you think all this crack length business was? I'm not clear about whether it was part of the claim or part of the defence. Incidentally (it's an unrelated point but I don't want to make too many posts here) the High Modulus results are quoted as being for bending failure. It would e interesting to know how much work they did on fracture and crack propagation through the composite, which is a very different thing.

 

[Tranona]

Crack length is central as the claimant is trying to prove that it was a gentle nudge - sufficient to crack a 7mm thick laminate, but not a 20mm. Because there is no concrete evidence of the size of crack caused by the impact such as reliable eye witness or photos the charade of calculating the water inflow (less what the pumps removed) in the period immediately following the grounding is the proxy for measuring the size!

All (on here anyway) agreed that this analysis is suspect, but it was not seriously challenged by the defence and accepted by the judge. Claimant +£400k, defendent minus same amount plus costs and a reputation for building expensive fragile boats, even if the company stays afloat (couldn't resist that one)

 

[jfm]

How central do you think all this crack length business was? I'm not clear about whether it was part of the claim or part of the defence.

oh I'm crystal clear on that. It is of critical importance. The claimant needed to show this specific grounding was of a hardness that would rupture the 7 mm hull but not a 20 mm hull. It therefore was critical to the claimant that the grounding was slight not heavy. Judge decided the grounding was slight based substantially on crack length, which as we're agreeing here is fundamentally flawed. Onus of proof is on claimant, so if crack length is an engineering red herring as to the hardness of the grounding then claimant hasn't proved his case and pearl should have won

 

[jfm]

Not according to the description in para 7 - which is why I was surprised this was not pursued in greater depth. It is described as being 20mm all over and the 20mm should have been maintained in the edges of the recess. There is no explanation as to how or why it ended up at 7mm at the edges (which presumably is where the crack occurred).

I don't read what you read in para 7, especially if you allow for the judge's clear struggle to understand and explain engineering.

In addition a 20 mm general hull thickness on a 60 footer is implausible

No worries!

 

[jfm]

the High Modulus results are quoted as being for bending failure. It would e interesting to know how much work they did on fracture and crack propagation through the composite, which is a very different thing.

that's why I wrote of HM above " if they were asked the right question"!!

 

[Aardee]

Can I just say a huge thank-you to jfm, Poecheng, Tranona and everyone who's contributed to the discussion - My legal knowledge is limited and I've learned a huge amount from this thread. This sort of stuff is why the forums are such a valuable resource!!

 

[Jamesuk]

Best news is that no one died! Very difficult for Pearl, and bad timing as they are pushing their boats further upmarket.

I thought it was
Concluded that it would have sunk regardless of original design or not?

What's the golden rule : don't bash your boat into a rock!!

 

[JumbleDuck]

oh I'm crystal clear on that. It is of critical importance. The claimant needed to show this specific grounding was of a hardness that would rupture the 7 mm hull but not a 20 mm hull. It therefore was critical to the claimant that the grounding was slight not heavy. Judge decided the grounding was slight based substantially on crack length, which as we're agreeing here is fundamentally flawed. Onus of proof is on claimant, so if crack length is an engineering red herring as to the hardness of the grounding then claimant hasn't proved his case and pearl should have won

Danke schön. I was having some difficulty working out whether it was the claimant or the defendant who introduced the crack size business.