Some thoughts on vessel engineering

Status
Not open for further replies.
Nov 22, 2008
3,562
Endeavour 32 Portland, Maine
I’m going to jump out of one of the threads below and offer some thoughts on engineering from the perspective of one who has spent his adult life designing the kind of things most of us here entrust our live to when we cast off the lines.

I’m not a trained engineer, having gotten sort of accidentally into naval architecture, but I’ve done a lot of engineering. Having to do it with simpler mathematical tools than most has forced me to often look deeper, or at least differently, at many problems. I did the research on sailing vessel stability that was a primary source for the establishment of regulations for sailing school vessels and developed the basic methodology of modifications to the existing passenger vessel rules. While meeting with USCG naval architects, it became apparent that graduates of some of the best marine engineering schools in the country did not actually have a clear idea what the numbers we were working with actually meant in terms of real wind striking real sails at sea. Their understanding ended with the number at the end of a complex calculation and how it compared to another number. It was illuminating and frustrating.

I’m sure some of you think I’ve been beating around the bush in my postings about one of the recent accidents for fear of being sued. There is the wonderful freedom of the layman in being able to come right out and say the obvious but that’s not what is at work here. It is a more nuanced perspective on what engineering and vessel design involves.

Someday, if there is a web page of my famous quotations, it will say right under, “Eggs don’t cause chickens.”, “Good engineers understand that numbers are only a guide to their judgment; poor engineers believe the numbers.”

Was the keel of that boat strong enough? Well, it sailed many miles and survived numerous (according to reports) groundings. Clearly, it was strong enough for something. It’s the “something” part of that equation that is the essential factor in designing things like boats but the emphasis is almost always on the other part.

Bridges are different. You can post a sign saying “No trucks over 10 tons.” If someone tries to sneak a 40 ton truck over and it breaks, you can blame them. With boats, you are designing bridges where a truck of almost any weight can come down the road. It becomes very statistical and “strong enough” becomes much less of a computational question than is commonly realized.

Often, the true nature of a structure isn’t revealed until it fails which is why testing to destruction is an essential part of may engineering development programs. If your vessel was rolled by a large storm sea, you wouldn’t necessarily expect to come up with the rig intact. You could well expect the mast column to buckle, stays to break, rigging pins to sheer. You might feel very differently however if the boat came up with all that stuff intact and big chunks of the deck torn out with the chainplates. It might be more critical than just your feelings since, in the former case, you would have something water tight and a base for erecting a jury rig. In the latter, you might well sink before you could stabilize the situation.

These relative relationships between the strengths of various components are often more critical to life and death than the absolute strength of a single component such as keel bolts.
 

RichH

.
Feb 14, 2005
4,773
Tayana 37 cutter; I20/M20 SCOWS Worton Creek, MD
The problem with (marine) material strength is thats its not a straightforward 'reading of the numbers' (ultimate strength, yield, deflection, etc.); plus, the maximum applied stresses are impossible to predict.
Therefore the engineer must consult the 'historical scantlings' (insurance records, etc. etc.) to see what worked and what failed ..... and then apply suitable 'factor of safety' (intentional 'overbuilding') to the structure. No one can accurately predict the loads, the impacts that result from the raging sea on a vessel. Marine designers and engineers however do have quite common and scantling-agreed values of 'factor of safety': 4X for the offshore boats, 3X for the 'coastal' boats and 2X for the strictly 'inshore' boats .... and yet still there are failures due to fatigue, design anomalies, etc. etc.

Just like aircraft, cranes, gantrys and other 'critical' stress designs, the ONLY way to come close to the 'absolute truth' is to build 'one', then cyclically and dynamically load it .... until it fails (ductile or brittle failure). Something that a small boat designer simply cant afford. Without such 'destructive testing' (which isnt perfect), designers and engineers MUST overbuild to 'historical scantlings'. Ductile failure is very easy to predict; however, brittle failure is extremely difficult to predict and that is what probably causes most failure in 'moving' or dynamic structures.

Once you gain an understanding of the progressive and extremely complex nature of 'brittle failure', then your design effort becomes a wee bit easier. My claim is that most rigging and rigging stress/connection points on especially sailboats is greatly undersized (even when overbuilt according to presently used 'factors of safety' ... if such were more conservatively designed, then one would hardly ever ever hear of a rigging failure.

The ancient Babylonians had a method to insure that a structure didnt fail .... if it failed and any occupants were killed or seriously harmed, the architect/engineer/designer was immediately executed.
 

Rick D

.
Jun 14, 2008
7,204
Hunter Legend 40.5 Shoreline Marina Long Beach CA
Thanks for the Perspective, Roger

Your narrative illustrates that macro clarity in thinking is a critical component in the success of the design. I'd be assured leaving port in one of your designs.
 

Ross

.
Jun 15, 2004
14,693
Islander/Wayfairer 30 sail number 25 Perryville,Md.
When the people in the past needed a covered bridge they went to the barn builders. Barns must be strong enough to allow a team and a loaded wagon to enter and be unloaded. The barn builders knew from long experience what was needed for those loads.
The wooden boat builders also drew on long experience for method and scantlings and sometimes they still got it wrong. A brand new British ship of the line capsized and sank when she was untied from the dock on her maiden voyage. The early fiberglass builders didn't know how strong fiberglass and resin was so they built to wooden boat scantlings and then started to back away from that overly cautious approach. Today we have reached, in my opinion, the far edge of the curve for light weight construction. The extreme is in the construction of the Volvo race and the vendee-globe race. I believe that those boats are engineered for enduring one race before being completely inspected and refitted for the next. From what I read, the combined thickness of the inner and outer skins is only about one half the thickness of the glass on the deck of my boat. They push the limits of all of the components of those boats for the valid purpose of winning a race. But don't give one of those boats to a university sailing team with amateur crews sailing and maintaining it without the original builder's oversight.
 
Dec 4, 2008
264
Other people's boats - Milford, CT
overbuilt

<snip>

The ancient Babylonians had a method to insure that a structure didnt fail .... if it failed and any occupants were killed or seriously harmed, the architect/engineer/designer was immediately executed.
The only thing that would insure was that buildings are way overbuilt and builders would never learn from past mistakes.

On the general issue of safety factors. More critical failure points (keel falling off) should be more strongly built than less critical failure points (rig failure). Trying to overbuild everything so that nothing ever breaks would just make the boat so heavy to be impractical.

Todd Smith




Todd
 

Ross

.
Jun 15, 2004
14,693
Islander/Wayfairer 30 sail number 25 Perryville,Md.
The only thing that would insure was that buildings are way overbuilt and builders would never learn from past mistakes.

On the general issue of safety factors. More critical failure points (keel falling off) should be more strongly built than less critical failure points (rig failure). Trying to overbuild everything so that nothing ever breaks would just make the boat so heavy to be impractical.

Todd Smith

The story is sometimes told of Henry Ford asking if any components of the model "T" never had a reported failure. When he learned what they were he had them re-engineered. He felt that they were over built and therefore were wasting money.
You can over build and you can under build. But when keels start to show daylight at the joint it is time to strengthen that element of the construction.
One criterion that has been used for shrouds and stays is that any two should be stronger than the weight of the boat. For me that also means that the chain plates have to be at least as strong as the shrouds or stays and the point of attachment a bit stronger yet.
What is the risk of failure and what is the consequence of failure, those factors are at the center of engineering design. Cost follows safety. If you can't build a safe product at a competitive market price you have to reduce cost somewhere other than safety.
 
Jan 26, 2007
308
Norsea 27 Cleveland
"Good engineers understand that numbers are only a guide to their judgment; poor engineers believe the numbers."
I missed the earlier thread and don't have a great stake in this one. But this quote caught my eye. There are all kinds of numbers. If you are talking about physical properties of materials then there may be variability in composition or manufacture and I imagine you take that into consideration. But if you are talking about particular calculations based on those properties, along with design, then a mismatch between the numbers and your intuition is telling you something. If you can quantify the difference between calculation and measurement, you have a good start on understanding more basic properties of a phenomenon and therefore an opportunity to revise how the calculation is done in the future.
 
Dec 4, 2008
264
Other people's boats - Milford, CT
Safe

You can over build and you can under build. But when keels start to show daylight at the joint it is time to strengthen that element of the construction.
And make sure the repair is as strong as the original construction.

What is the risk of failure and what is the consequence of failure, those factors are at the center of engineering design. Cost follows safety. If you can't build a safe product at a competitive market price you have to reduce cost somewhere other than safety.
The usage of the product and the tolerance for risk of the user also have to be included. There is no simple criteria for "safe product". What would be safe in one context would not be safe in another.

Todd
 

RichH

.
Feb 14, 2005
4,773
Tayana 37 cutter; I20/M20 SCOWS Worton Creek, MD
On the general issue of safety factors. More critical failure points (keel falling off) should be more strongly built than less critical failure points (rig failure). Trying to overbuild everything so that nothing ever breaks would just make the boat so heavy to be impractical.
Todd Smith
Todd
Absolutely correct. A 'breakage' is a catastrophic failure and a boat that is too heavily built to sail well is a functional failure ... but both are still failures. Thus there is always the 'balance' between the two extremes.

However, that a keel catastrophically fell off is most certainly not a functional failure. As Roger stated it would indeed be interesting to review the CG documentation of the findings of incident or at least be able to forensically examine the failure and approach to understand exactly what happened; but, .......... sadly the totally adversarial nature of the USA legal system would most probably prevent this seeking of the true facts, etc.
 

Ross

.
Jun 15, 2004
14,693
Islander/Wayfairer 30 sail number 25 Perryville,Md.
The safety requirements for a paying passinger carring vessel are quite different from the requirements of an extreme racing vessel. Paying passingers have a right to believe that their ride will be safe and that they will be protected from harm by the crew and by the quality of the vessel. The crews in the boats of the Volvo race know what manner of boats they are sailing and are willingly taking the rsks knowing full well the posible consequences.
 
Nov 6, 2006
10,216
Hunter 34 Mandeville Louisiana
Vessel Engineering

Yup, in making a design, some assumptions are made as to the loads that will be accommodated.. If the engineer/designer has a great basic understanding of what the design will be exposed to, then he can use judgment and experience (doesn’t have to be his personal experience) to provide a suitable design.. WHEW! In addition, in today’s world, there are “Risk Assessments” done on many designs to ferret out the points of failure and look at what is going to happen when that point fails.. Again that all assumes competent maintenance.
 
Oct 22, 2008
3,502
- Telstar 28 Buzzards Bay
One point I'd make is that often what seems to be a very minor change can have very large and serious unintended consequences.

I'd also point out that most serious accidents are rarely a single catastrophic event occurring. They're usually made up of a chain of smaller events occurring in sequence. Often, if the sequence were slightly different, or one of the events didn't occur, the disaster or serious consequences would be an order of magnitude less severe.
 
Jul 20, 2005
2,422
Whitby 55 Kemah, Tx
Someday, if there is a web page of my famous quotations, it will say right under, “Eggs don’t cause chickens.”, “Good engineers understand that numbers are only a guide to their judgment; poor engineers believe the numbers.”
One of the reasons I will trust a boat maker who had made tens of thousands of boats throughout 30 years of business to a boat maker who makes 10 boats a year. Hunter, who has been know to skimp on quality, has, in my mind, done a very good job of not skimping on safety (equipment that is cruital). I will admit they skimp on a lot of non-safety things to make it affordable like head hoses, cushions and so on.

They are always trying new things...take the open cockpit. It is a great safety feature. You can't keep an open cockpit full of water. Hunter was the first to come up with that (as far as I understand). They are not afraid to try new things and test them. Sometimes they make a mistake but I would have to say they have done more for the boat industry then any other maker in a long time because of their "lets try this" attitude.
 
Jul 20, 2005
2,422
Whitby 55 Kemah, Tx
The problem with (marine) material strength is thats its not a straightforward 'reading of the numbers' (ultimate strength, yield, deflection, etc.); plus, the maximum applied stresses are impossible to predict.
Therefore the engineer must consult the 'historical scantlings' (insurance records, etc. etc.) to see what worked and what failed ..... and then apply suitable 'factor of safety' (intentional 'overbuilding') to the structure. No one can accurately predict the loads, the impacts that result from the raging sea on a vessel. Marine designers and engineers however do have quite common and scantling-agreed values of 'factor of safety': 4X for the offshore boats, 3X for the 'coastal' boats and 2X for the strictly 'inshore' boats .... and yet still there are failures due to fatigue, design anomalies, etc. etc.

Just like aircraft, cranes, gantrys and other 'critical' stress designs, the ONLY way to come close to the 'absolute truth' is to build 'one', then cyclically and dynamically load it .... until it fails (ductile or brittle failure). Something that a small boat designer simply cant afford. Without such 'destructive testing' (which isnt perfect), designers and engineers MUST overbuild to 'historical scantlings'. Ductile failure is very easy to predict; however, brittle failure is extremely difficult to predict and that is what probably causes most failure in 'moving' or dynamic structures.

Once you gain an understanding of the progressive and extremely complex nature of 'brittle failure', then your design effort becomes a wee bit easier. My claim is that most rigging and rigging stress/connection points on especially sailboats is greatly undersized (even when overbuilt according to presently used 'factors of safety' ... if such were more conservatively designed, then one would hardly ever ever hear of a rigging failure.

The ancient Babylonians had a method to insure that a structure didnt fail .... if it failed and any occupants were killed or seriously harmed, the architect/engineer/designer was immediately executed.
very well put!!!!!
 
Jun 7, 2007
875
Pearson- 323- Mobile,Al
What would be really great IMHO would be for mass boat manufacturers like HUnter,Catalina Beneteau to make "blue water" boats. Not fundamentally different than their stock coastal boats. You know things like thicker fiberglass heavier rigging etc to make their coastal boats sstrong enough for offshore work. I would think that a few thousand dollars of reinforcement would make a substantially stronger boat using existing molds. That way us less fortunate sailers could afford better boats. Of course some designs are not suitable for offshore work no matter how strongly built.
 
Jun 6, 2006
6,990
currently boatless wishing Harrington Harbor North, MD
Civil engineering phlosophy

I was taught in college that if you can't "fully" define the problem or the loads etc, you should design the structure in such a way that it fails "safely". This will give the occupants a clear warning to get out or take some other action to protect their lives and property.
The case of a de-masting is a good example. Tear the dang mast off but make sure the attachment points don't go with it. Another example is the shear pin on a bulldozer blade. It is the pivot point that allows the blade to go up and down and transmits the pushing load from the blade to the dozer frame. If you push too hard the shear pin ($15) shears off and saves the blade ($30,000) from breaking.
So I'm not so sure that you actually need to be able to calculate the load that will actually occur. That load will surly "break the boat" to pieces. I guess my engineering assumption would be that the sea can break my boat no matter how strong I make it. So I'd design to make my boat have those properties that make it strong where it needs to be to fail safe or that enhance it's safety. This could lead to designs like make it lighter so you can sail faster and escape the storm more readily.
 

Ross

.
Jun 15, 2004
14,693
Islander/Wayfairer 30 sail number 25 Perryville,Md.
Storms in the Atlantic travel 10-35 mph. It will require very good weather information to avoid such a storm. No sailboat can escape that.
 
Sep 24, 1999
1,511
Hunter H46LE Sausalito
"thicker fiberglass"

You know things like thicker fiberglass heavier rigging
MoonSailer, you seem to want to take yacht design backwards. These days production companies such as Catalina and Hunter are reinforcing critical stresspoints in hulls and decks with Kevlar. Works a lot better than the "thicker fiberglass" approach of the '70s.
 
Oct 22, 2008
3,502
- Telstar 28 Buzzards Bay
Kevlar is a lousy material in many ways for reinforcing high-stress points on decks. It does a lousy job of adding stiffness IMHO, since you can't push on a rope. :) Carbon fiber makes a lot more sense for highly loaded areas where the deck stiffness and strength need to be increase.

Kevlar's best use in modern laminates, IMHO, is under the waterline. In many cases, a kevlar-reinforced laminate can take an impact and not tear open. Kevlar has great impact resistance and doesn't tear readily. In a cored-hull, having kevlar reinforcement along the interior laminate often means that in an impact, the outer laminate and core may be damaged, but the boat will often retain its water-tight integrity, due to the resilience of the interior skin and the kevlar reinforcement.

Using a core material throughout most of the deck and hull allows the manufacturer to create a lighter, stiffer hull and gives them the ability to add thicker laminate, with or without exotic material reinforcement, at all high-stress points, without increasing the weight of the boat significantly, due to the great weight savings allowed by using a cored laminate throughtout most of the rest of the vessel. Solid fiberglass is very heavy for the strength and stiffness compared to a properly engineered cored-laminate.
MoonSailer, you seem to want to take yacht design backwards. These days production companies such as Catalina and Hunter are reinforcing critical stresspoints in hulls and decks with Kevlar. Works a lot better than the "thicker fiberglass" approach of the '70s.
 
Nov 22, 2008
3,562
Endeavour 32 Portland, Maine
Kevlar is a lousy material in many ways for reinforcing high-stress points on decks. It does a lousy job of adding stiffness IMHO, since you can't push on a rope. :) Carbon fiber makes a lot more sense for highly loaded areas where the deck stiffness and strength need to be increase.
There was a whole rash of fiberglass rudder stock failures on high tech boats a few years (maybe decades, it's all kind of blurring together) ago. The stocks were too flexible so they added a layer of carbon fiber to stiffen them up. However, the carbon fiber layer wasn't strong enough to fully take the load. It was stiff enough, however, that it would become fully stressed before the fiberglass stock and bent enough to take any real load. As a result, the carbon fiber would fracture. The result was like that little scratch you put on a glass tube or sheet to snap it cleanly in two. Rudders that might have just flexed snapped off and drifted away.
 
Status
Not open for further replies.