Comments by "Keit Hammleter" (@keithammleter3824) on "Oceanliner Designs" channel.

  1.  @dovetonsturdee7033  Regarding the source you asked for in your 1st para, see https://www.nytimes.com/2008/04/15/science/15titanic.html. As to whether Titanic's rivets in the affected area, they were most certainly seriously defective, with a strength well below the norm for the time. As I said, this was determined in testing performed by an expert scientist at the US National Institute of Standards and Technology (NIST, formerly NBS). You can't get a much more authoritive source than that. Further, photos taken around the affected area by submersible showed plates still in position but with rivets not present in the holes meant for them - confirming that the impact shock travelling down the hull caused rivets to pop off beyond the impact zone. The NIST study plus the photo evidence showed that water entered six of the 16 water-tight compartments and this the ship could not cope with. Further, the impact area of the iceberg covered only 4 compartments as most, and the ship was designed to cope with 4 compartments flooded. Therefore, if Harland & Wolf had performed normal rivet testing and ensured that good rivets were used, the ship would almost certainly survived. Your comment on extra strips and replaced rivets etc is rendered unimportant by the NIST and photo evidence. Regarding the documentary "Titanic 100 : Mystery Solved" - I had not known of its existence. I will watch it, and post what I learn from it later. It may be a week or so. I suggest you don't respond to this post until I do.
    1
  2. 1
  3. 1
  4.  @dovetonsturdee7033  The conclusions, that the ship was not weak, spoken right at the end of the film "Titanic 100: Mystery Solved" are not supported by established facts, and not even supported by the facts shown within the film. The NIST showed very clearly that the ship had faulty rivets in the area of iceberg damage. The ship was indeed shoddily built and weak. The main trust of the film is about whether the ship broke in two on the surface due to a design or construction flaw, then does not actually answer that. It shows the hull splitting down from the top with the bottom plates hanging on last, proving that the rivets there were good. But, strangely, it doesn't address why the side walls gave way. If the side wall construction, including rivets, was good, the bottom should have failed in compression. Of course, the splitting of the hull has nothing to do with the damage caused by the iceberg, and the ship would have sunk regardless of whether the ship split in 2 or not. The film shows rivets missing from a part of the hull that was not bent, and rivets still in place in parts of the hull horribly bent. This confirms that the rivets used, which came from several suppliers, included good batches and bad batches. some rivets were strong and some were weak. The tests of rivets shown in this film are totally irrelevant because the rivets used were recently manufactured and did not have the slag inclusions that the rivets from the iceberg impact area had. Also the tests were done at room temperature and not at the sea temperature at which steel becomes a lot more brittle. Further, the test was done by slow build up of force until the first rivet failed. Doing this will always make one rivet fail first, since no two rivets will have precisely the same strength and receive precisely the same load. The actual impact with the berg produced a sudden impact over a wide area, causing many rivets to fail simultaneously. Then a shock wave propagated away from the impact area, popping more rivets outside the area of actual impact. The film completely ignores the fact that Harland and Wolf bought rivets from multiple sources - some were good and some were bad. The NIST scientists proved by testing actual rivets recovered from the impact area were bad - very bad. "Titanic 100: Mystery Solved" must be the most repetitive and boring film I've ever seen. It told us at least 30 times a storm is coming. Told us the RUV cable was snagged at least 20 times. Told us about 10 times that the steel they used to make test rivets was 100 years old - that's of no actual significance. Told us about 20 times they are making a map of the debris field. Its heavily padded with unimportant footage. Such as showing us the survey crew pulling up and old anchor weight left by a previous survey. It showed crew pulling up a rope and the spoken commentary implied it could be a Titanic artifact - not likely given it's a modern blue nylon rope. Not recommended.
    1
  5.  @dovetonsturdee7033  The expert metallurgists employed by NIST clearly established the Titanic was built with at least one bad batch of rivets and that's what caused Titanic to sink. Go read the NIST report - an expert detail written report by expert metallurgists, illustrated by imaging, is far more authoritive than a dodgy experiment by a blacksmith who didn't even bother to replicate the correct temperature. You can easily access the NIST summary by searching [nist titanic rivets]. Titanic was built with rivets purchased from multiple suppliers. Some rivets were good and some were not. Neither NIST nor I have claimed that ALL rivets in Titanic were faulty - just the rivets in the area of impact, where it mattered. There MAY have been bad rivets used elsewhere - that we cannot know. Olympic was never put to the same test - it never hit an iceberg. In any case, Olympic started construction first and was launched 6 months ahead of Titanic, but the construction of the hulls overlapped. Since Harland & Wolf's normal rivet suppliers could not supply enough for both ships, H&W started buying rivets from other, smaller foundries - hence the probability of bad batches of rivets was higher during Titanic construction than it was for Olympic. The fact that empty rivet holes were found in unbent parts of Titanic's hull during expeditions to the bottom (and this was even shown in the film) is very telling that the rivets failed due to shockwaves in the hull and not directly due to local impact with the berg. That is pretty obvious. I said the film is bad, because it is bad. All that ridiculous repetition - we don't all have complete short term memory loss. Testing brand new rivets at room temperature, as they did in the film, is pointless.
    1
  6. 1
  7. 1
  8. 1
  9. The presenter said that the ship's rudder was adequate and the ship quite manoeverable, but this is not true. In the situation where an iceberg or object is detected ahead, Titanic was a lot LESS manoeverable that other large ships of the day. Notwithstanding what other ships could do because they had reversable engines, any ship that cannot turn or reverse to avoid an object in the time available after the object is sighted is NOT a safe and competent design. Witnesses stated that the first officer included a command for full reverse, as well as his steering command. This would have been a correct command if the ship was driven entirely by reciprocating engines - it would both slow the ship down, increasing the time to turn, and maximise the amount of turning by driving water over the rudder regardless of hull velocity. The first officer probably commanded reverse because he lacked experience on turbine ships, and trials on this new class of ship were inadequate and did not include a test of avoiding an object. At the time, most ships had reciprocating engines - turbines were new. Titanic had a turbine for cruising fuel economy, driving the 4-blade centre propellor. With the technology of the day, this turbine could not turn the propellor in reverse. Therefore, for manoevering in harbor, the ship had two smaller 3-blade wing propellors, each driven by reversable reciprocating engines (they also assisted the turbine in driving the ship forward during cruise). Thus, when the order to reverse was given, the turbine was simply just stopped, markedly reducing water flow over the rudder to only that imparted by the ship's slowing motion. If the ship had, instead, two rudders behind the wing propellors, the ship would have been as manoeverable as other reciprocating enegine ships and the iceberg easily bypassed. If a single much larger rudder had been installed, the forward motion of the ship could have meant sufficient water over the rudder to provide enough turning to avoid the iceberg. On that basis, the often stated view that the rudder was of inadequate size is in fact correct, and OceanlinerDesigns is wrong. It has been speculated that if the order to reverse engines had not been given, the dramatically better rudder effect would likely have saved the ship.
    1
  10.  @DerpyPossum  : The purpose of the trials was to verify the ship met the speed and fuel consumption requirements and establish what the manoeverability was. These days computer simulation will tell you at negligible cost, but back then the only accurate way was to go to sea and try various manoevers out. Titanic's trails were limited to 12 hours and did not include a test for turning to avoid an object dead ahead. measured They performed an emergency stop ("normally all engines full astern - but in Titanic the turbine could not be reversed, as I said) and measured the time distance taken to achieve zero speed (3-1/4 minutes, 780 m) but did not perform an emergency turn test. Who says Olympic's handling was any different? The same test was not performed on Olympic until AFTER the Titanic sunk, in order that the Inquiry board could understand why it hit the iceberg. Changing the rudder would not have been a simple thing to do. It would have cost a lot of money when money was tight. Fitting two rudders would be a major design change. Fitting a bigger rudder would require hull strengthening and a much larger steering motor. Remember that it took over 3 years to build Titanic. Olympic wasn't launched until Titanic construction was already 2/3rds complete. Following Titanic's sinking, Olympic returned to the shipyard for major changes to make it safer, at significant cost. However, the changes were things like improving water-tight compartment integrity, so that the ship would be more likely survive a collision with an iceberg - improving turning was not considered practical. Some people considered the captain an idiot for going at full speed in a known iceberg area. Perhaps there is some truth in that, but if adequate manoevering trials had been done, he would have known an iceberg could not be avoided by turning or stopping and would have ordered a speed reduction. It is inconceivable the captain, and first officer who was actually in charge at the time, both experienced large ship men, would not have done so. It was normal with reciprocating engine ships to go at full speed, as their much better manoeverability und usually lower speed meant they could steer around icebergs. Some say Titanic's centre propellor was 3-bladed, some not. It doesn't really matter. It must have been appropriate in blade area and diameter to match the turbine output. In either case, ordering engine stop or ordering all engines full astern meant low water velocity over the rudder and thus a marked loss of rudder effect. A phenomena unique at the time to the Olympic class of ship and not appreciated by key people at the time. There was some controversy over just what orders the first officer gave, as different survivors said different things. The consensus was that he ordered immediate full astern, and all stop only after the collision. But, immediate full astern or immediate all stop - it doesn't really matter - both orders would result in marked loss of rudder effectiveness, though ordering full astern was clearly the worst thing to do. Modern calculations have shown that keeping the engines at ahead would have given enough rudder effectiveness to probably just clear the iceberg. The time to stop the Olympic class by going full astern was 3-1/4 minutes, as measured in a post-accident test. The time to put the engines full astern was considerably less than this - of the order of 30 seconds - the same time it took to fully turn the rudder.
    1
  11. 1
  12. 1
  13. 1
  14. 1
  15. 1
  16. 1
  17. 1
  18.  @Barkiro2010  : I'm no metallurgist or shipbuilder either, but as an engineer I do know some basic facts. Modern mild steel that we frequently encounter has a carbon content 0 to 0.3% and negligible other impurities. This makes it by metal alloy standards fairly soft and very ductile - hit it and it deforms. In contrast, Titanic's hull plates were nominally also mild steel, but with a higher carbon content and lots of metallic impurities - being the standard of the day. This means it was a bit tougher and more springy - hit it and it will bend at bit but also transmits the shock .Titanic's rivets were essentially (by manufacture and the method of installation) wrought iron. This gave them more strength than mild steel but also made them somewhat brittle. Experience of shipbuilders at that time showed that the brittleness was acceptable. However, Titanic's rivets that were recovered were shown by US National Insitute of Standards and Technology to be defective. They had a high sulfur content, a lot of slag from defective manufacture and were excessively brittle. NIST found that rivets had snapped off. I'm no metallurgist, but NIST have work class experts, and if their experts say the rivets were defective, then I certainly accept that. Their reputation is world's best. Tim Foecke was the lead scientist in the rivet analysis. It should be noted that NIST tested rivets from the affected ship's side. It is quite possible that most of Titanic's rivets were sound, and it was bad luck that the berg hit where the rivets were a bad batch. However it is part of NIST's findings that the rivets were defectively brittle because they were incorrectly installed - insufficient control over temperature (the type of rivets required installation while red hot). Titanic is currently disintegrating. It has lasted as long as it has due to great depth where temperature is low and oxygen also low. Plenty of old sunken ships have lasted just as long in less favourable depths. Plenty of large businesses have gone through very tight patches, almost gone broke, but survived many more decades.
    1
  19. 1
  20. 1
  21.  @DerpyPossum  : I did not dispute that H&W etc complied with LEGAL requirements. But they did not comply with MORAL requirements. They were the ones at fault by having poor ethics - they created a ship outside what the rules were written for, and thus should have gone beyond requirements. The concept of using lifeboats as a transfer means between two ships was a commercial excuse and not a Board of Trade idea. But the Board of Trade by 1870, under pressure of ship owners, accepted not having sufficient lifeboats for all passengers and crew in regard to high passenger density steam ferries working between Britain and France. (See Parliamentary Debates, London 1870, page 323.) In open sea in any ocean in the world that an ocean liner may go is quite another thing. It is often the case in the British system of governance that legal requirements usually don't keep up with advances in technology - in fact legal requirements get updated or created when deaths occur making the need very obvious, and such deaths often occurred, as with Titanic, when competitive pressure inhibits engineering thought. It is not reasonable to expect that those who make the laws and rules anticipate future technical developments. They are law makers and not engineers. Thus, while the lifeboat rules were eventually shown by the Titanic to be insufficient, the Board of Trade cannot be held to be at fault. Nor can the entire shipping industry be held at fault over Titanic - that's ridiculous, as most ships were nothing like the Titanic. The Board of Trade, when they set the rules, did not anticipate the construction of 50,000 tonne ships carrying 3,300 people across the Atlantic on a routine basis. When they set the rules, ships had much lower passenger density and nobody would have thought of such a thing. Modern history abounds with hazards, not anticipated by those who make laws and rules, being created by technical innovations. Mostly, though, the design engineers ask themselves "what can go wrong?" and do the right thing and self-implement what needs to be done to make it safe. Nobody but the design engineers have the specialist knowledge and can do this. The Titanic engineers did not ask themselves "what can go wrong?" They were unethical and at fault.
    1
  22. 1
  23. 1
  24. 1
  25. 1
  26. 1
  27. 1
  28. 1
  29. 1
  30. 1
  31. 1
  32. 1
  33. 1
  34. Ät 3:22: "the mast is wood as you don't want metal in contact with the aerial." Well, no you don't. But you really don't want wood in contact either - even seasoned wood has a moisture contact, is made of hydrocarbon, and is very lossy to radio frequency energy. In any case, during bad storms, when you really want the radio to work, masts get coated with conductive salt spray. With the high power (10 kw) transmitters used for world-wide communication, the radio energy can burn the wood, which is very undesirable. From the point of view of radio energy absorption, a metal mast is actually better, as it is a solid conductor and not a lossy thing. Further, for the lightning arrestor at the top of the mast to work, there must be a wire running from it down the mast to the hull, to conduct the lightning current safely away. Electrical insulation for the aerial wire(s) was and is provided by "egg" insulators inserted into the wire a few metres from the supporting mast. Egg insulators are made of glazed porcelain, which is an excellent electrical insulator and rapidly sheds water. Egg insulators are around 100 mm long, too small to show properly on a whole-ship drawing and too small to show in most photographs. But if you look carefully at 2:57 just under the corner of the flag, you can see two little dots, which are the egg insulators. Wood provides a degree of flexibility required in masts, with less weight than using steel. A steel mast would require regular inspection for corrosion, which in the top section is not so easy.
    1
  35. 1
  36. Mike omitted an important factor with regard to vibration: Whenever you have 2 or more propellors, you get propellor beating - pressure changes in the water as a blade of one propellor sweeps past/near a blade of the other propellor. There are three ways of addressing this: 1) separate the propellors laterally - there is limited scope for this on a big ship, especially if there are three or more propellors; 2) separate the propellors longitudinally (ie at different distances from the stern) - this reduces hull volume at the rear of the ship, which is not a good thing; 3) fit different numbers of blades, eg 3 on the outboard props and 4 on the inboard. I gather that Mike's conclusion that Titanic had a three-bladed centre prop is really solely based on a single digit in some engineer's personal notebook. This is not good research - it could be just an error by that engineer, we have no way of knowing. Mike said that the H&W engineers could have been doing a comparison experiment with the two sister ships but this does not make much commercial sense. An engineer would seek to come up with the best configuration he can, and that won't change until one of the ships completes sufficient voyages to show up a problem. An engineer would be risking his job if he was seen to intentionally make one ship inferior to the other, as must be the case if they are different. Sure, ships got their props changed back then (in pleasure boats today they still occasionally do) but in response to an encountered problem that needed addressing, not on speculation
    1
  37.  @journeyman_philosopher  I certainly did watch Mike's video. I did not say he's definitely wrong , I said that he drew conclusions on very thin ice. The only reference I could find that Titanic had a three-blade centre prop is the same one Mike quoted - that engineer's notebook. The only reference I could find that H&W were experimenting 4-blades vs 3 blades is a discussion on a Reddit forum about Titanic - hardly a reliable reference. I think you are misunderstanding the engineering of propellor specifications as it was back then. Sure, it was rule of thumb methods rather than finite element fluid dynamics, but they weren't stupid. Considerable experience with propellor applications had been built up. I've worked for a marine engine dealer - we used much the same methods to match props to hulls and engines - and got it pretty right nearly every time if the hull people got their bit (hull drag) right. (Props are like gears in a car - you must have the right prop blade angle etc for the ship's speed through the water, just as you need the right gear for the speed you are doing in a car.) But sometimes the hull guys got it wrong, and they sometimes did back in Titanic's day too. When that happened, propellors got changed. Incidentally, be aware of why professional engineers keep personal notebooks. These are the main reasons (then and now):- 1. Professional associations require the production of notebooks as proof of experience when granting corporate memberships;. 2. Source material for updating one's CV and job applications;. 3. In the event of a patent dispute or getting sued for infringement, a notebook can be presented in court as evidence of prior art; A high level of accuracy is not required; indeed, some young chaps don't bother with a notebook until they have to produce it, and then spend a couple of evenings writing one, trying to remember what they did in their earlier years.
    1
  38.  @gokulgopan4397  But back then there was only 2 kinds of test they could do: a) try a configuration out in a ship, which may need several or many voyages to get enough data on various sailing conditions; or b) tank testing. Tank testing has serious limitations, due to the need to scale the dimensions and velocities, coupled with the effects of Reynolds Number being non-linear. All this means that for any given ship, they could get things a bit wrong, realise it from either excessive fuel consumption, excessive or too low engine RPM for a given hull velocity, or excessive vibration, or any combination of these three. If so, they could decide to try a different propellor, making an informed guesstimate as to what difference(s) the new prop should have. If the new ship is very different to what the shipyard had built before (eg twice the displacement through increased length) there is obviously more scope to get it wrong. But to launch two identical ships into commercial service with different propellors in order see which one works best - I don't accept that. Its just not a way to get the company manager's respect. You do your calculations (rule of thumb) and the answer is the answer - the best you can do, uncertainty notwithstanding. If you launch the two ships with different propellors - you know that there is 100% certainty you are going to have to dry dock one of them early at huge expense and loss of revenue. If you launch 2 ships with your best idea of what the props should be, it will more than likely never be any need to early drydock one of them.
    1
  39. 1
  40. 1
  41. 1
  42. 1
  43. 1
  44. 1
  45. 1
  46. 1
  47. A turbine input of 9 Lb/in^2 absolute cannot be correct. It is not the absolute bit that’s wrong, it’s the amount. It is likely a reporter’s error in the reference Mike cited. I show why below. An engineer would get an accurate picture by consulting standard steam tables and a lot of math too complex to show in a YouTube post. However there are some online steam plant calculators based on accurate math that we can use. The best and easiest to use is the US Department of Energy’s Steam System Modular Tool Steam Turbine Calculator. To use this calculator we need to enter:- # the steam mass flow rate, # turbine input steam pressure and temperature, and # the outlet pressure. # the turbine thermodynamic (isentropic ie assume no heat or mass loss) efficiency. The DoE calculator calculates the power output and the temperature of the steam at the turbine outlet. The DoE calculator also electrical generator efficiency as in input, because these days the main application of steam turbines is power stations. Since we only want shaft power, set this to 100%. Harland & Wolf never disclosed the mass flow. However www.titanicology.com shows how it can be calculated from the published reciprocating input pressure, HP cylinder volume and RPM – 6200 Lb/min, i.e., 372 kLb/hr. [Note, Titanic’s boiler capacity, 260 Lb/min per boiler, exceeded this by about 17% – due to the need to clean boilers while underway and to feed electricity generators and auxiliaries. So Titanicology’s estimate is reasonable. A reasonable value for a large low pressure turbine thermodynamic efficiency is 80%. The outlet pressure is no problem, it is 1 Lb/in^2 absolute i.e., -13.7 PSIG , an entirely typical condenser operating point at that time. We are told by The Shipbuilder special edition that the input pressure is 9 PSIA i.e., -5.7 PSIG but it gave no temperature. This doesn't matter, we can just try progressively higher temperatures until no condensation occurs in the turbine. Steam condensing within a turbine would cause serious problems – blade erosion, loss of efficiency, vibration due to unsteady flow conditions. It cannot be allowed. The DoE calculator shows that 385 F just avoids condensation with 9 Lb/in^2 absolute input, producing a shaft output of 14 megawatts i.e., 18,800 HP. It can’t of course be as high as 385 F as that is close to boiler temperature. We need to change another parameter. There is little scope for changing the output pressure, increasing it to 2 Lb/in^2 absolute only drops the required input temperature to 319 F – still way too high. Let’s try an input of +9 PSIG, with all other values unchanged. This time we get a warning that there must be water in the inlet, for all temperatures up to 238 F. We can’t accept that, as it would hydraulic lock the reciprocating engines and wreck them. So +9 PSIG cannot be right. And 238 F is still too high to allow the reciprocating engines to work at proper reduction. Trying a turbine input of -2.8 PSIG (11.9 PSIA) in the DoE calculator, with 202 F, we don’t get condensation in the turbine or reciprocating engines. The calculator then gives us 12.0 MW i.e., 16,100 HP. Say 16,000 HP allowing for friction etc. Conclusion: - # A turbine input of 9 Lb/in^2 absolute is possible but not in Titanic as it requires a steam temperature almost as high as the boiler output. # Neither can 9 Lb/in^2 gauge be right, as that would mean heavy condensation in the reciprocating engines, causing rapid catastrophic damage. # A turbine input of 11.9 Lb/in^2 at 202 F is compatible with Titanic’s plant, does not give condensation in either the turbine or the reciprocating engines, and produces a turbine shaft output of 16,000 HP, which is correct. In practice, we would want a safety margin against condensation, operate the turbine with an input of 204 F. Perhaps the reporter for The Shipbuilder magazine wrote down 9 when he should have wrote 11.9.
    1
  48. 1
  49. I feel that our friend Mike missed a trick on this. The Enquiry looked into such things as the correctness of orders given by Murdoch once the iceberg was sighted, and the construction of the ship, but politicians are not competent to do that. We know that Murdoch's order was to turn to left and most likely was to put all engines to reverse, to go around the berg as there was insufficient time to stop. The order to turn was correct and but the order to reverse was not, and Murdoch probably gave it because he was new to Titanic and, having no time to think, gave an order appropriate on his previous non-turbine equipped ships. (Some people believe he ordered Stop All Engines, but that makes no sense at all - no experienced officer would give such an order.) If the enquiry had appropriate marine experts asking the questions, instead of ignorant politicians, Murdoch's error might have come to light. If the enquiry had appropriate marine experts asking the questions, they might have brought to light shortcomings in the construction, such as Harland & Wolf using rivets of unknown quality, but this is unlikely. Note that while nobody in the industry would have believed that newspaper nonsense about Titanic being unsinkable, it was entirely reasonable to believe the ship could survive reasonably likely iceberg collisions. It didn't (it was known that it struck the berg in such a glancing blow that there was no hull buckling and its speed was unaffected), so it was reasonable, even back then, that a competent enquiry would seek answers as to why it didn't survive, and dig down until they found the answers.
    1
  50. 1