About the new "rivet" theory

I just wondered what the general consensus amongst our more knowledgable ship people might be concerning a book that came out pretty recently, ("What Really sank the Titanic" or something like that) think it came out in 07 or 08 written by two metallurgists, one of whom was named Jennifer Hooper. The assertion is that the ship sank because: 1) they used the softer iron rivets in the bow section where they couldn't fit the hydraulic riveting machine in, 2) These iron rivets were of inferior quality because they were having trouble getting enough of the good ones blah blah. Later, there was a Discovery Chanel documentary in which they use a machine to test two mock ups of riveted seems and compared how much force it took to pull them apart, and not surprisingly, the iron rivets failed sooner than the steel. I'm no scientist, but I couldn't help but wonder if the comparison was even relevant. Yes, the iron rivets might have been weaker but it still doesn't answer the question "would it have made any difference anyway?" Even if the steel rivets were stronger, against the force of an iceberg collision they might have still broken no matter what they were made of. Did anyone out there see this book or watch that show?
Rob H.
First, have you read the book?

I have.

Niether of the authors asserts that the Titanic sank because the rivets were softer. They do make the point that issues with the rivets may have been a factor in how quickly the ship foundered.

Getting rivets of good quality wasn't a problem for Harland And Wolff, however some of what was used came from stock known as Roll Three Best wrough iron because it was adaquate for the job and easier to use in the areas where they were driven.

You might want to get a copy of the book and read it for yourself. Forget the drek cobbled up by the news media which took bits and pieces out of context to contrive a pretext. While some of the issues they raise are not without controversey, and are not universally agreed with, they showcase the science which was done in the words of the people doing the science.
I believe I saw the same documentary you spoke of. They riveted two plates of steel together the same way it would have been done on the titanic, using rivets that were supposed to have similar properties to the ones in titanic's hull, and put it in a machine to see how much pressure it would take to pop the rivets. They calculated (if I recall correctly, going by memory here) That the force of Titanic brushing against the berg would have been about 14,000 psi, and the rivets in the test machine failed (I think) at around 9,000.

Personally, when i saw that, my first thought was "OK, and...?" It should be obvious that whatever force was exerted against the ships hull, it was greater than it was able to withstand, or it wouldn't be at the bottom of the Atlantic! I really would have liked to see them repeat the test with "better" rivets and maybe even plates joined by welding instead to see if they would have failed under the same circumstances. I have a hunch (an uneducated hunch, of course) that they would have. If you want your ship to float, its best not to run it into something at high speed, no matter what it's made of.

The documentary, if I recall correctly, was called Seconds to Disaster : Titanic or something along those lines. I believe you can find the entire documentary (split into 9-10 minute sections) on youtube.com

Apologies for resurrecting such an old thread, but I thought I could shed a little useful light on the discussion.

Doug Criner

I, too, have read the book. The book's title, "What Really Sank the Titanic," is very misleading. The ship foundered because it struck an iceberg, resulting in flooding. There will always be weakest links in any system's design - but to blame any supposed weak rivets on Titanic's sinking is completely silly. Too bad the authors allowed the publisher to use such a title.
Doug -- The sad truth is that publishers always get the last word on titles. They run roughshod over authors using the argument that the author's title won't "sell." It's about money, although the results aren't always a success.

Weak rivets undoubtedly played some role in the damage, else they would not have found such a high percentage of problems among those found on the bottom. However, this does not mean that weak rivets caused the ship to sink. The mass of the ship, the size of the iceberg, and the speed of impact all played greater roles in creating the damage.

Transportation accidents are almost never rooted in a single cause, but rather in a chain of causation.

-- David G. Brown
Colour me more than a tad skeptical about the rivets found on the bottom. Over 3,500,000 of them were used in the construction of the ship and the breakup zone notwithstanding, the vast majority of these rivets are still right where the builders left them in spite of a violent impact with the bottom which put a bend in the bow which is 120° from the vertical line.

My understanding of the rivets which were collected is that they amounted to a grand total of 37 and with little way of knowing with any degree of certainty where they came from. From a scientific standpoint, the usefulness of this sampling is problematic to say the least.

Keep in mind that I'm not presenting this as any sort of "indictment" of the work that the authors carried out. The samples they had were the samples they had and they simply had to make do. In the matter of reading too much into what they have to say, I suspect that they would be the first ones to urge caution.

The problem here is that media outlets and publishers have reading too much into things as their bread and butter. As David said: It's all about the money.
Correct me if I'm wrong, since I'm not a metallurgist or any other kind of expert on shipbuilding, but even if they ship had been built with a welded hull (which I'm sure was many years in the future in 1912), or even stronger rivets, wouldn't the steel have fractured anyway? Also, with a welded structure, can't a stress fracture propagate for a very long distance from the impact, instead of stopping at the next joint like it would in a riveted structure?
I can offer some half marks on the hull fracturing since the evidence of this actually happening is virtually non-existant. In fact, an examination of the wreck itseld shows metal which is buckled, bent, mangled, and torn, but not fractured.

That said, a welded structure might very well have been catastrophic for the very reasons you suggest. If there was a crack, it could have propagated itself around the hull with lightning speed. As it stands, with a riveted hull, and cracks would tend to start and stop with the individual plates.

Jim Currie

Has it ever occurred to any of these 'super scientists that ships very frequently collide with stone built or concrete quay walls .. even in 1912 and well before that.
The area of damage was described as being about 2 feet above the boiler room deck plating which makes it on the vertical face of the shell plating. How many ships actually were sunk during the berthing and un-berthing processes? During that time, it is very frequent that the ship's underwater parts bump and grind along the quay wall. Just after WW2, many ships were composite-built with rivets and welding. Usually the bows were rivetted. Very often rivet heads were sheared but this only resulted in a few holes here and there. The water tightness was effected by caulking the plate edges. They did the same thing in 1912. The most frequent breaches of the hull were caused by distortion of the plate edges when the ship's underwater shape was changed during impact. This often caused deformation of the rivets, loss of rivet heads and elongation of the rivet holes resulting in sudden loss of water tight integrity along the particular seam. What seems to have escaped the researchers is that the shell plates in question were attached at 3 feet intervals to vertical heavy shell frames which gave enormous stiffness to the structure. When impact took place, the shell plating would be heavily set-in between the frames. Thus, the plate edge in the way would be severely distorted. A bit like this dreadful sketch:

Without doubt,rivet failure would be part of the process but not, as far as I can see, the principal cause. However; nice bit of metallurgy but totally impractical given the rough handling ships got during the processes I mention. No bow or stern thrusters or any of that jazz.. just brute strength and ignorance. Any naval Architect will tell you that the ships of old were built like the proverbial brick khazi.

>>Without doubt,rivet failure would be part of the process but not, as far as I can see, the principal cause.<<

I agree. Let's face it: When you hit an iceberg at full speed, you will have rivet failure!!!!
Out of curiosity, do any of you think a modern vessel would hold up much better in such a collision? I thought maybe a warship would, but I have a hard time picturing a US Navy vessel getting into such a pickle in the first place (But now that I've said that, I'm sure someone more learned than I am will somehow prove me wrong) I just tend to think that this and the brittle steel theory that was popular a few years ago are sort of a red herring, every structure has a breaking point, after all!
>>Out of curiosity, do any of you think a modern vessel would hold up much better in such a collision?<<

With a third of the ship's length open to the sea? As Lightoller would have put it: "Not damned likely!"

Keep in mind that the Olympic class was designed to exceed the standards of the day, specifically that the ship should be able to remain afloat with up to two compartments breeched. The Olympics could do that with up to four!

The catch is that this standard still exists and the minimum is treated as the max.

Most warships could easily survive this sort of damage and worse, but there is little reason to design a merchent vessel to that sort of standard and a lot of good reasons not to.

Jim Currie

Good morning!

Michael K. you ask: "do any of you think a modern vessel would hold up much better in such a collision?"

It is possible. It depends entirely on a number of things including the design of the vessel.. merchant or military... her pumping capability and how much reserve buoyancy (water tight volume above the waterline) she has. In the past, I have been involved in pulling a very large vessel off the rocks and towing it to port with it's main deck no more than 9 inches clear of the water. As for shell plating strength:
Next time you see a modern ship, have a look at her side shell plating. Through it, you will see the ship's ribs.. as if the shell plating was made of tissue paper. You can actually see where the heat process of welding has caused ripples in the plates. In the old days, the emphasis was on amount and thickness of plating combined with double bottoms and sides in conjunction with longitudinal and thwartship wt bulkheads. The engineers had serious problems with power to weight ratios.
Nowadays, there is a much more scientific approach to ship construction metallurgy and scantlings ( dimensions of construction materials) as well as damage limitation capability.

As for rivets and their uses and tests; you might find this of interest:

Michael - That's pretty much what I had guessed. I'm sure designing a cruise ship with the extensive subdivision and thick plating of a warship would be prohibitively expensive and would probably have a negative effect on fuel efficiency, not to mention usable passenger space. Sort of like buying an M1 Abrams to drive to and from work every day =P

Jim - I would have liked to see that, I'm sure it was a hell of a job getting that ship back to port in one piece. Thanks for the link, by the way, it answered several questions that had been brewing in the back of my head.

I appreciate both of you taking time to answer my questions. I'm just starting to dig in deeper to the technical details of the titanic and shipbuilding in general so this is all new and very fascinating to me!