Question Regarding Titanic’s Rivets During Stress Testing

[Steven C]

On Steven C Said:

Then it didn’t help that the bulkheads were open at the top, allowing flooding to spread into undamaged compartments as the bow sank.

It should be noted the reason why Titanic’s bulkheads weren’t sealed at the top, was because it was and still is against maritime shipbuilding law to fit a passenger vessel with watertight bulkhead decks above passenger spaces with no means to escape. The only reason as to why modern cruise ships are fitted with watertight bulkhead decks is because there is no passenger accommodation below the waterline or anywhere near any of the subdivision on modern vessels. There are various other reasons why Titanic’s watertight subdivision was the way that it was. Titanic wasn’t the first nor last liner with open bulkhead “roofs” take a look at the S.S. United States, Queen Mary, and Queen Elizabeth.

Not sure I understand this, if watertight  bulkheads extend to deck level, but there is a stairway in each compartment, people can still escape. I thought the problem was that bulkheads stopped well below the deck to allow grand ballrooms etc. Connection below deck through bulkhead would be by watertight doors which could be closed by the crew on the event of an accident.

[Steven CParticipant @stevenc95127]

Not sure I understand this, if watertight  bulkheads extend to deck level, but there is a stairway in each compartment, people can still escape. I thought the problem was that bulkheads stopped well below the deck to allow grand ballrooms etc. Connection below deck through bulkhead would be by watertight doors which could be closed by the crew on the event of an accident.

It is entirely possible to extend the watertight bulkheads to the deck level but then you’d run into the issue of the difficulty for both crew and passengers being able to traverse the ship. Titanic was a 2 compartment ship meaning she could survive with any two compartments flooded, most modern cruise ships are also of the 2 compartment standard. In reality, Titanic was closer to a three-compartment ship and in a worst-case scenario could handle 4 compartment flooding in her foremost or aft-most compartments. After the disaster, her sister ships Olympic and Britannic were modified to raise their bulkheads to facilitate up to 3 compartment flooding anywhere in the ship. There’s also the science of flood curves, to which it’s a little harder to explain with just text.

[Nigel Graham 2Participant @nigelgraham2]

I think it instructive to examine what engineers of the time knew, and did, before criticising their work.

For a start, rivetted plate and beam construction is not inherently weak if it is designed correctly. How many 19C, let alone early-Edwardian, bridges etc. are there around the world still in full use?

Wrought-iron and mild-steel alone or in combination were still common in the early-20C; while material quality, strength and joint design were already well advanced. Iron rivets in steel plates were common but rivets made from a ductile grade of mild-steel were coming into use.

E.g, Hutton (1911) or Spooner (1913)

Also what becomes fairly clear from those texts is that when joints did fail, it was often by the plate tearing away from the rivets, not by the rivets shearing, though presumably both modes were known.

It is also important to know what caused failures in structures designed and built to best practices.

Corrosion was perhaps the most common and often due to neglect, poor structural settings or poor use. Hutton (op.cit.) illustrates from insurance records that most boiler explosions arose from combinations of those.

However, accidents could and did occur involving vastly greater forces and directions than the designers intended and could envisage in normal service with a sensible factor of constructional safety and proper use. Such accidents can and do still happen – but as with the Titanic are often traceable to human error, major in her case but more often chains of small errors.

.

Manufacturing practices had improved considerably too, for example with punching rivet holes already long known to be injurious.  Spooner calls that a “barbarous practice” superseded by modern drilling machines and high-speed steel tools, and hydraulic rivetting-machines were already common by then as well.

Hutton advises that higher-carbon steels should also be annealed before assembly to prevent embrittlement, but this appears to be in connection with boiler shells.

…

RMS Titanic sank because she crashed into an iceberg at speed, puncturing the plates and perhaps springing rivetted joints made to withstand normal service – including storms. Not because she was put together with rivets! Incidentally, I wonder what part of the ship had been the source of the tested samples?

Would any modern ship fare any better in a similar collision? Probably only in managing to stay afloat thanks to watertight bulkheads to main deck level, and high double hull walls. The objection that such bulkheads would impede movement of people in normal service is rather chimerical because the higher deck doors at least are open unless in emergency, and there are exit routes between bulkheads. There of course many bulkhead penetrations for pipes and cables, but they are not just holes through the wall. They are properly sealed.

‘

Noel asks about the propellor-shaft tunnels. I think we can exonerate them! A ship’s propellor-shaft passes through a pipe called the “Stern-tube” fitted with journals and stuffing-boxes, and those on Titanic could have played no part in the ship’s foundering. The stuffing-boxes do not need withstand much water-pressure, either, and any leaks would be small and probably into a fairly small compartment or tank. I don’t know the Titanic-class draught but a depth of 30 feet equates to only about 14psi (~ 1 atm). Hutton gives a diagram of the typical arrangement for the time, in a section devoted to the strength and design of such shafts. Incidentally the stern journal was often of Lignum Vitae wood, which works well as a bearing for steel or bronze in water.

.

Titanic was one of three similar liners. One was purloined as the hospital ship ‘Britannic‘ in WW1, and sunk by enemy action. The third, ‘Olympic‘, enjoyed a safe and uneventful 30 or 40 year career as an inter-continental ferry, the intended use for all three, ending in being scrapped.

.

As for the brittle “Victory” ships, what actually broke: the middle of the plate, the weld margins, or the welds?

”””

References:

Hutton, W.S. The Practical Engineer’s Handbook, Crosby, Lockwood & Son, London, 1911

Spooner, H.J. Machine Design, Construction & Drawing, Longmans. Green & Co, London, 1913

[Michael Gilligan]

On Michael Gilligan Said:

For your convenience … here is the URL extracted from that source:

https://www.nist.gov/publications/metallurgy-rms-titanic

… which then gives you the facility to have a Local Download:

https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=852863

 

Very interesting read – thank you.

[duncan webster 1Participant @duncanwebster1]

Lots of airplane wing skins are riveted to the underlying structure, I dare say under more controlled conditions than in a shipyard.

[Nigel Graham 2]

On Nigel Graham 2 Said:

…
As for the brittle “Victory” ships, what actually broke: the middle of the plate, the weld margins, or the welds?

…

None of the above.   The problem was a combination:

1. Welding allowed the designers and builders to square off doorways, hatches, and other junctions in a way not possible with a riveted construction where lots of overlap is needed to get the required strength.
2. Ships hog and sag in a seaway, stressing the whole structure,
3. Stress concentrates at sharp corners.
4. The steel embrittled and weakened at low temperatures.

Victory ships cracked at corners in cold weather, and once started the crack took the line of least resistance, which could be through a weld, along the margins or across the plates.

The immediate cure was to round off all the corners.  After that was done, upgrading the steel eliminated the problem.

I have a sneaking suspicion that the designers knew the cheap steel was weak, but didn’t care much because Victory ships were mass-produced to meet an emergency requirement and weren’t intended to last long.   Later, the problem may have been addressed when it was realised the Allies couldn’t afford to lose the crews as well.

It’s hard to know what engineers understood at the time.   The UK developed the first commercial jet airliner.   The Comet performed well initially, but soon began to mysteriously fall out of the sky, causing everyone to lose confidence and trashing a golden business opportunity.   The problem was eventually traced to sharp passenger window corners, where stress concentrated as the plane pressured and de-pressured with changing altitude.   The stress caused micro-cracks, which quickly grew out of sight until the body failed well below design strength.   Claimed to be a huge surprise to aeronautical designers in 1953, which is odd because ship-builders were already alert to stress-raisers.   I suppose it’s possible they didn’t speak,  or were too busy to keep up with their reading.    Similar  problem with the Tay Bridge Disaster, where Sir Thomas Bouch and the Astronomer Royal claimed to be unaware of the latest US and European work on wind pressure, leading them to underestimate the side force a storm would put on the bridge.

Dave

 

 

 

[noel shelleyParticipant @noelshelley55608]

My mention of the shaft tunnel was in the context of the vessel sinking by the bow and water entering from the machinery space and spreading aft, not water entering via the stern tube or stuffing boxes. Noel.

[Bantam BillParticipant @bantambill]

It’s a common misconception that the square windows of the comet caused the structural failure, they did not. The failure started at a small cut out in the fuselage for the ADF aerial and spread along the fuselage skin, this was made worse by poor manufacturing techniques (punched rivet holes) and the fact that the windows doublers were not bonded on as required,

[Michael GilliganParticipant @michaelgilligan61133]

Interesting page, here:

https://aerossurance.com/safety-management/comet-misconceptions/

including a nice little newsreel film about the test tank

MichaelG.

[Nigel Graham 2Participant @nigelgraham2]

More contemporary material relevant to two disasters…..

Failure to understand wind loadings were certainly a major part of the Tay Bridge Disaster, but one of several reasons. The others were:

– Serious design faults: possibly the use of cast-iron for the columns, but also the spans were not bolted to them. Instead they were kept on their seatings simply by their own weight. Cast-iron had been a common material for columns and plenty of Victorian iron is still holding up structures like railway-stations, but is a poor choice for a bridge subject to high lateral as well as vertical loads – even when the castings are sound and geometrically correct.

– Appallingly sloppy workmanship by the contractors. The investigation revealed:

…..the columns’ castings were of very low quality, and their hollow cores were far off-centre;

…..the rag-bolts holding the iron columns to the masonry and brickwork piers were about half the length Sir Thomas had specified;

…..the brick-laying’s very low standard;

…..missing bolts or rivets.

Local fishermen reported having heard parts of the bridge rattle when trains passed along it.

– Equally sloppy oversight by the Board of Trade of both the design and construction.

 

Tradition claims the fateful storm was unusually fierce even for the area, but it seems that was not correct. It was a powerful gale but not unusually so.

The Titanic and her sisters were designed and built to above the legal minimum standards, and to the high quality of materials available and expected, at the time.

 

[Ref:

The Engineer, 125-year Anniversary compilation, pub. Morgan-Grampian 1981; pp42-47.

This magazine is aimed primarily at professional engineers in senior managerial and directorial roles. The Anniversary edition covers its contemporary attempt to report on the disaster and the Inquiry, and describes how that was hindered by the Government refusing to allow it to photograph the bridge wreckage. The 19C edition outwitted that by discreetly obtaining photographs but printing etchings made from them – reproduced in the 1981 publication!]

””””

So how did The Engineer cover the Titanic‘s loss almost 33 years later? (Same anthology)

The 1981 anthology showed it had covered the Inquiry and the reporting as much as the technical aspects, but also made some unexpected comments.

It examined the question of speed. Had the ship been moving at half-speed, about 10 knots, the force of the collision would have been quartered, but also the estimated time from spotting the iceberg to hitting it would have doubled from one to two minutes, perhaps giving more chance to evade the ice. However, it also points out, the passengers expected a rapid crossing! It wrote:

“….. But that a modern liner should travel at … 10 or 11 knots for a whole night is inconceivable. Passengers would rather run the very distant risk of collision with an iceberg.”

I suspect the survivors may have have begged to differ, though re-thinking that, the author was probably right that the passengers would have thought such a collision very unlikely. They would have had full trust in the ship and her crew.

The magazine had previously examined the question of double-hull construction, but suggested the complicated joining of the two skins would not necessarily save the ship due to the damage to the outer being transmitted to the inner. It did though agree with a strong public demand for full lifeboat and raft provision for all on board.

Lest we think the magazine’s pre-WW1 staff were rather hard-hearted, they were not.

In fact after an elegy that reads as more for the ship than the people, a subsequent edition carried a memorial to the engine-room staff, noting they had probably all stayed at their posts to their end, and even more, printing a list of their names.

Later, it recorded the efforts by the Institute of Marine Engineers and “a special committee of Liverpool gentlemen” to raise a memorial to the Titanic’s engineers, and gave the addresses for contributions.

It would be interesting to know which if any of the public newspapers of the time were as thoughtful.

.

Finally, The Engineer pondered if a 19C invention may have avoided the disaster.

Its July 1864 edition had recorded an ice-warning system invented by A. Bryson, which had ‘received a most favourable report’ from the Royal Scottish Society of Arts.

[There seems to have been much less of the divisive Science / Arts / Crafts splits in the 19C, than we see nowadays. Perhaps engineering was much more obviously to non-engineers, the amalgam of scientific knowledge and creative crafts that it is.]

The instrument warned of an iceberg by using electrical signals generated by the temperature differential (in air or water is not stated) to sound an alarm on the bridge; and was estimated to cost ‘no more than £50 to produce’.

Yet by 1912, 48 years later, Bryson’s iceberg detector ‘… had sunk without trace.’

 

[Neil WyattModerator @neilwyatt]

Some interesting facts here, followed by speculation on the rivets.

https://titanichistoricalsociety.org/titanics-brittle-steel/

 

Neil

[Neil Wyatt]

On Neil Wyatt Said:

Some interesting facts here, followed by speculation on the rivets.

https://titanichistoricalsociety.org/titanics-brittle-steel/

 

Neil

Too many errors to list!   No such thing as ‘Battleship quality steel’, and even if there was Harland and Wolff didn’t build any warships until 1916.  The conclusion is unfortunate ‘I think these 3 million mild steel rivets might hold the secret.’ because it’s well documented that all the rivets were made of Wrought-Iron, not mild-steel.   And Science does not tell us that Siemens Marin Steel made in 1912 won’t develop low temperature brittleness above liquid nitrogen temperatures!

For it’s time the Titanic was a well built ship.  The problem was ships don’t tolerate have their sides scraped forceable along hard objects.  Titanic was unlucky, and although better rivets and steel might have saved her, leaks spanning 5 watertight compartments are likely to be uncontrollable.

Almost exactly a century later, the Costa Concordia, built of the best modern welded steel sank in warm water after side-swiping rocks at speed:

Her Captain was showing off and, ignoring the planned route, deliberately steered close to the shore, slicing open a long section of the hull.   Amazingly of over 3000 on board only 32 were killed.   If you must sink your ship, best to do it in shallow water close to shore in a flat calm, with help close at hand.

Dave

 

[Nigel Graham 2Participant @nigelgraham2]

The Titanic society did not invent the term “battleship quality”. Its paper clearly states that was the shipyard employees’ own name for it.

“quadruply riveted.” I assume this means 4 rows of rivets in a butt-joint with two cover-plates (or internal beam and external cover-plate) – normal practice for major plate-work structures.

The writer only “thinks” the rivets were heated – that was normal practice for pneumatic-hammer rivetting, though I am not sure about for hydraulic closing.

I wonder if this is one of those cases where modern historians look at something from a modern perspective and do not really find out, or work out, how things were actually done.

.

I have watched a video (sorry I don’t have the reference) describing the salvage of the Costa Concordia. This showed horribly clearly that the death-toll could have matched that of the Titanic’s loss. It was as low as it was because the ship was very close inshore in relatively warm water, and sank “only” to about half-beam.

She had come to rest on a submerged, sloping ledge. Had she staggered barely her own length further ahead, or slid sideways, she would probably have slipped into much deeper water. The bow was overhanging a steep slope and the salvors feared it would shear off as they righted the vessel, so had to add extra bouyancy tanks to support that.

As it was they had had to bolt steel frames to the rock both to stabilise the hull and as part of the righting arrangement.

 

[Author]

Posts