|
On Testing Steel
There was a life before I started to write about politics at Moon of Alabama. So when a news story comes up that relates back to my previous life as an industrial engineer I will certainly read it and at times even write about it. Here is one of these.
Metallurgist admits faking steel-test results for Navy subs
A metallurgist in Washington state pleaded guilty to fraud Monday after she spent decades faking the results of strength tests on steel that was being used to make U.S. Navy submarines.
Elaine Marie Thomas, 67, of Auburn, Washington, was the director of metallurgy at a foundry in Tacoma that supplied steel castings used by Navy contractors Electric Boat and Newport News Shipbuilding to make submarine hulls.
From 1985 through 2017, Thomas falsified the results of strength and toughness tests for at least 240 productions of steel — about half the steel the foundry produced for the Navy, according to her plea agreement, filed Monday in U.S. District Court in Tacoma. The tests were intended to show that the steel would not fail in a collision or in certain “wartime scenarios," the Justice Department said.
The strength of the special kinds of steel that allows submarines to go deep without imploding must be assured under all circumstance. (Especially when one wants to move undersea mountains by running into them, as the USS Connecticut recently tried.) Special castings on submarines are often used where things like the periscope or cooling water lines penetrate the hull. To have any potentially brittle material at those places could be catastrophic. Due to the falsified test results the navy might have to reduce the maximum allowed diving depth for some of its submarines.
But the reason given by Thomas for falsifying the test results is what I find really concerning:
When confronted with the doctored results, Thomas told investigators, “Yeah, that looks bad,” the Justice Department said. She suggested that in some cases she changed the tests to passing grades because she thought it was “stupid” that the Navy required the tests to be conducted at negative-100 degrees Fahrenheit (negative-73.3 degrees Celsius).
This is an alarmingly 'stupid' quote from someone who is supposed to be a metallurgist. These tests are not 'stupid' but necessary.
There is one standard test for impact strength of steel that is regularly done at subzero temperature. It is the Charpy V-notch toughness impact test (video).
bigger
For a Charpy impact test a part of a casting is cut off and machined into a well defined piece with a notch. Its edges are then put against an anvil. A swinging hammer comes down and destroys the test piece. The difference in heights of the hammer at the starting position and at the end of the swing is an expression of the energy that was needed to destroy the piece.
It is a simple, easy to do test and the results can tell a lot about the material characteristics of the test piece. The pictures below shows the test results for two kinds of steel. The upper piece fractured but did not break apart. The more brittle one below snapped.
bigger
Depending on its inner crystal structure the toughness of a metal can change with its temperature. More brittle material has a body centered cubic structure (BCC) with one atom sitting in the middle of a cube formed by eight other atoms. Tougher steel alloys have a face centered cubic crystal structure (FCC) where an extra atom sits at each face of the cube.
BCC structure – bigger
FCC structure – bigger
To find out which type of structure a piece of metal has one can cool it down and do a Charpy impact test at very low temperatures.
bigger
Below a certain temperature steel with a body centered (bcc) cubic crystal structure will suddenly become weak while steel with a face centered cubic structure (fcc) keeps it toughness. A Charpy impact test at low and normal temperature allows to differentiate between those.
The crystal structure of steel can be influenced in the foundry during the casting. The alloying, the temperature of the melt, the speed of cooling and/or an additional tempering will all effect the structure.
If the foundry has made a mistake during a cast it might have produced a steel with the wrong structure and characteristics. Material testing is the way to find out about possible mistakes. A Charpy test at different temperatures is a simple way to determine if mistakes were indeed made.
To skip the test or to falsify its values, as the person in question has done in this case, is a big no no.
But what really concerns is me that Thomas either did not know or ignored the above. She thought it was “stupid” that the Navy required the sub zero tests even while there are sound and rather simple reasons for these. One wants to differentiate 'good' steel from 'bad' one. It is not only the U.S. Navy which requires such tests. The classification societies and insurers for civil ships and the oil and gas industry have similar procedures.
When the metallurgist who was being trained to replace Thomas found out that she was falsifying test results he immediately recognized the gravity of the problem and informed the company. It was the right thing to do.
During a part for my engineering education I did an internship in the material testing lab of a large shipyard. We did the coooold Charpy tests only once a week because we needed liquid nitrogen to cool down the test pieces. Liquid nitrogen has a boiling point of −195.8 °C (-321 °F). It evaporates fast but is fun to play with (vid) which I, of course, did a lot.
Regarding the Titanic… it was indeed the virtually identical sister ship, the Olympic, that was sunk, deliberately. The collision had bent its keel irreparably. The naval engineers that had designed her concluded that she would probably do well enough for the coming summer season then would probably break apart in the high seas during the winter. The boat was towed to Belfast, where the Titanic was being readied for her maiden voyage and the names were changed. The original lettering was found on the boat when the wreck had been located and an explorer device sent down to take pictures.
As for the structure… She was designated unsinkable because of her watertight compartments, intended to prevent all pervasive flooding in the event of a hit by a torpedo, in anticipation of the coming war that the British had long been preparing for with the launch of the Dreadnoughts. However…
J.P. Morgan had bought the White Star Line with the idea of sticking it to Cunard, which had launched the Lusitania and the fabulous Mauritania and was planning yet third, the Aquatania, billed as “all the luxury a millionaire can handle” (its launch was delayed by the war). The Olympic was the first of White Star’s competing trio, the Titanic was the second, and the Gigantic was to be the third (it became the Majestic, in the end).
To save money, recycled steel was used. First-rate steel might well have resisted enough of the impact to make the tear across the side tolerable and let her limp the rest of the way to New York. Also, to save money, the watertight bulkheads were not built up to the top of the ship, which means that when the front compartment had filled to the top of the bulkhead, the water broke through into the next compartment, which was already filling up. When the second had filled, it also overflowed in the same way, and so on until the front half of the ship was so full of water that the ship went under. The break is thought to be due to the immense stress on the front part from the flooded watertight compartments that pulled her under while the stern, still mostly full of air, resisted.
In another profit-oriented move, a huge section of the boat deck from which the lifeboats were to be launched was devoted to private promenade decks outside the deluxe suites, which were going for about $5,000 and responsible for a major share of the income on the crossing. So, there was nowhere near enough lifeboats.
The captain was a White Star Line flunky who could easily be blamed afterward based on his previous record.
The Olympic/Titanic was sent out in the middle of a major coal strike when the cost of coal was colossal because it was being brought in from the Continent where the coal producers were taking advantage of the shortage of coal in the U.K., where coal fueled just about everything, with the U.K. ordinarily self-sufficient in coal. A second White Star Line passenger liner, temporarily out of service because of the lack of coal, was sent out at the same time with a full crew, no passengers but 3,000 sweaters and 3,000 blankets, to pick up the passengers. They missed the rendez-vous.
The day before the sailing the insurance on the boat was doubled. As soon as the sinking had been confirmed, White Star collected. There were two hearings, one in London, one in New York. In the London hearing, a ocean-liner officer with long experience on the North Atlantic run was called as an expert witness, and he testified under oath that in the dead of night, under the stars, an iceberg big enough to damage a boat the size of the Titanic (it was the biggest ship afloat, slightly bigger that the Olympic) could be discerned at least two miles away, giving the captain time to veer away. Also, there was a major iceberg warning out. The witness’s testimony was dismissed as irrelevant…
It was easily the biggest insurance fraud in history, £10 million, and those were 1912 pounds sterling, when the overwhelming majority or workers were paid in shillings and pence and were lucky to see a one-pound note a few times in their life but never a fiver.
Morgan was to preside over the sumptuous celebratory maiden voyage, but cancelled at the last minute, along with a significant number of other from the upper crust. Morgan’s excuse was that he was too sick to travel even in the luxury of the Titanic. Two days later her was seen sporting on the Riviera.
Posted by: RJPJR | Nov 11 2021 2:42 utc | 121
Robert Macaire @132
Sorry for my belated reply. Yes, like all ‘interatomic chemical bonding’, metallic cohesion is electrical in nature. However, the brittle to ductile transition in bcc metals at low temperatures is not due to a slowing down or a reduction of the kinetic energy of the valence / bonding electrons.
As light elementary particles with very low mass, electrons have some peculiar ‘quantum mechanical’ properties. One of these is that the ‘orbital’ velocities of the electrons around an atom are quite high, of the order of 1000 to 2000 km per second. The velocities of the valence electrons in condensed phases (solid or liquid) remain similar to the velocities in isolated atoms. These velocities are not affected much by temperature at all, at least for temperatures of order of 10, 000 deg C or more, above which the atoms start to ionise.
The origin for the brittle – ductile transition in bcc metals lies in some characteristic differences in the 3-dimensional spatial arrangement of the atoms (more precisely, actually the ionic cores of the atoms) in the bcc lattice structures, as compared with fcc lattice. The fcc structure has fully close-packed planes, lying perpendicular to the four body diagonals of the fcc unit cube shown in b’s diagrams. Every atom in these planes is ‘touching’ six neighbours in the same plane. Consecutive close-packed planes in the fcc structure therefor have maximal inter-planar spacing compared to other sets of parallel planes, since they contain the maximum possible number of atoms.
This large separation minimises the extent to which the bonds have to be disrupted as the planes try to slide or ‘slip’ over each other under the action of an applied shearing stress. That disruption is further minimises by the close-packing of the atom within the planes, which minimises the size of the ‘slip steps’ or ‘slip transitions’ that are required. Slip steps are the elementary processes by which ‘inelastic’ (‘plastic) deformations associated with metal ductility and toughness occur, in response to mechanical stresses that are sufficiently large.
By contrast, in the bcc structure, there are no close packed planes (no two –dimensional close packing). There is only a one-dimensional close-packing direction, along the body diagonals of the bcc unit cube. Each atom is only able to ‘touch’ two nearest negbours in these atomic rows, rather than six, as in the fully close-packed planes present in the fcc structure.
As a result, the body-diagonals are the favoured directions of easiest slip in bcc metals. The possible ‘easiest slip’ planes that contain the body diagonal slip directions in the bcc structure are planes that contain paqirs of parallel face-diagonals and the diagonally opposite and parallel edges of the bcc unit cube. These slip planes are not close packed, and are therefore more closely spaced (smaller inter-planar separation of neighbouring sliip planes) than in the fcc structure, making slip in bcc more difficult, especially at low temperatures.
Event in the fcc structure, the amount of energy that would be required to make an entire plane slip by one atomic slip step over one of its its two neighbouring planes is very large. In all practical situations, the slip is able to proceed by a series of steps, one row of atoms at a time, due to the presence in real crystals of large numebrs of inherent and spatially extended line defects, or ‘fault lines’, known as dislocations.
In fcc metals, there are three easy slip directions in each slip plane, parallel to face diagonals of the fcc unit cube. In each of the four fcc slip planes, dislocations can move (propagate) quite freely without impediment in those three directions, at all temperatures. The fc structure is therefore said to have 4 x 3 = 12 physically equivalent and active ‘slip systems’ (slip system = slip plane + slip direction) at all temperatures above absolute zero. For a metal to be fully ductile for all possible directions of applied stress, its crystal structure needs to have a minimum of five active slip systems.
By contrast, the bcc structure has 6 slip planes, each with two easy slip directions (along the two body diagonals of the unit cube contained in each slip plane). The number of slip systems in the bcc structure is thus 6 x 2 = 12, as for fcc. Below a certain temperature however, all twelve of the physically equivalent bcc slip systems ‘freeze’ and become inactive, because they need some extra ‘thermal activation’ or ‘atomic vibrational energy input’ in order for slip of dislocations lying in these slip planes to occur.
There are a number of informative websites that discuss the relationship between crystal structure and metal ductility. These include the ones linked below:
Influence of the lattice structure on the ductility
What is the diference between fcc and bcc crystal structure?
Posted by: johnF | Nov 22 2021 6:23 utc | 143
|
