Tuesday, December 20, 2011

Patent Trolls Again: High Temperature Steam, Boilers, and World War II.

If you follow the Admiralty's technological progress from Naval Estimate to Naval Estimate through the Nineteenth Century in the popular press, it's easy to end up making jokes about April Fool's Day and drunkard's walks. The Admiralty was always backwards, always conservative, always wrong. The forward looking inventor, like Galileo, can see what no-one else can see. And the Press, unlike the establishment, can see his point.

If you're a contrarian like me (and perhaps only like me), you end up itching to take the establishment's side. Because the innovators can be wrong. If you set up a process where the only way to escape the criticism is to indulge the critic's every whim until something goes transparently wrong, you get the capsizing of the HMS Captain. Who wants to pay a battleship and 500 lives to shut up one obstreperous inventor, much less sacrifice the eldest son of his main institutional patron to shut that man up? And yet, even after Captain went down, people continued to argue. No less a person than Baron Kelvin was called at the inquest, and proved Captain's recklessly dangerous design with numbers. And yet the matter continued to be controversial as long as anyone cared.

You think, you wish, that the gobs end up stopped. Two decades ago, the province of British Columbia dropped a godawful amount of money in a local shipyard for aluminum-hulled high speed catamaran  motor ferries. It was a transparent disaster in the making, complete with the provincial premier suing detractors.  (Wikipedia article.) By 2003, "the fast cats" were a bad memory. Yet a certain major national publisher,  bought a piece of high concept nonsense of the 1421 variety from one of the advisors that sold the premier on the fastcats and flogged it onto the market with great lashings of positive press. The press made a lot of money, the professionally-wrong author made a lot of money. A century from now, historians will have to listen to earnestly crazy people telling them that Francis Drake built a secret English colony on the inshore coast of Vancouver Island.

The professionally wrong can, and do, go on and on. 

I know. My outrage fails to  move you. These things happen. It's just that when Nathan Myhrvold raised his head in public again to babble about how we should respect innovators more, I can at least present a test case to refute the meta-argument.


This is about a guy that you've never heard of: Vice-Admiral Harold G. Bowen. Not famous enough for Wikipedia, his book not prominent enough for Google Books. If you've heard of him, it's through footnotes in books that are almost as iffy. And yet, from such sources, comes an enduring myth that was first propounded to me by a smugly-knowing naval wargamer in the early 1980s.

So. One of the reasons that Captain capsized is that it had sails. That was because the steam engines that it also had weren't particularly efficient. Efficiency is basically determined by how much of the energy released by burning coal reaches the water as thrust, and early steam engines lost a great deal of the heat energy along the way. It wasn't until the triple-expansion compound steam engine came along in the 1880s that warships and liners could do without sails.

In the tradition of motor engines that actually work, as opposed to ideal ones, triple expansion steam engines are a work of the baroque imagination, and they were done away with by Jackie Fisher for reasons of labour efficiency that I've discussed before. Sir Charles Parson's turbine, which replaced it, had from the first all the marks of a successful engine. Which is to say, it was obviously, intuitively, wrong. Expanding steam was blown through aerofoils, causing them to spin a shaft that in turn spins something attached to it: propeller, alternator, egg beater, I don't know. Since engine efficiency is theoretically determined by the difference between the temperature of the motor and the heat sink into which you're dumping your spent heat, you want to use very hot, hence very high pressure steam. This expands quickly, causing the shaft to spin quickly, causing the thingie to which it is attached to spin equally quickly. Heaven help you if you want to run in reverse or even vary the speed of the output.This is fine if you want very high-frequency AC electrical current or to boil water with your propeller. If you don't, you end up inserting intermediate gearing to step down the rotation speed for most purposes. Imagine, for a second, gears grinding when they're taking a hundred thousand horsepower. 

Fisher, who had a fine grasp of how the game is played, basically shifted the Navy over to turbines ahead of the development of that gearing. Crazy as Dreadnought's machinery was, it was a good prototype for what would come later, and actually work. That came in during World War I in a climate of crazed improvisation. As did another innovation; for, if you want your steam to be absolutely as hot as possible, you can take it off the boiler, put it through steam tubes, and "superheat" it further. HMS Nelson, the first battleship built after World War I, introduced superheat into the Navy, so that its mechanical guts worked at 400○F, 600 lb/sq in steam pressure. 

At this point, remembering that it was supposed to be conservative about these things, the Admiralty decided that the next experiment should use a destroyer instead of its new flagship.  HMS Acheron thus hit the water at 500lb/sq in and 750○F, with a mechanical efficiency at 0.608, compared with Nelson's .79i . Unfortunately, Acheron proved well-named. Superheated steam is crazy stuff to be playing around with if you're not sure what you're doing, and the Engineering Branch decided to back off at this point,as there were a great many other things to experiment with.

 This is why British destroyers and cruisers laid down in 1928 mostly operated at 300 lb/sq in, 625○F. Even by 1939 larger RN ships had not reached the "advanced" steam conditions tested in Acheron, operating at  400lb/sq in and 700○F.

Many within the Engineering Branch were dissatisfied with this. On land, the electrical engineering industry was experimenting with power plants that pushed the temperature of steam up near the melting point of steel, and the big American names (GE and Westinghouse) jumped into marine plants, working with 850○F steam.ii

This is where Admiral Bowen comes in. As chief of the United States Navy's Experimental Boiler Facility in the late 1930s, he certified these plants for naval use and pushed through their use in the fleet.

 It was another of those extraordinary moments in the history of technology. As the reader will remember from tedious arguments with Truthers five years ago, while steel doesn't melt at 850 degrees, it begins to undergo a decline in mechanical strength. Chromium high-tensile steel, the standard product in use, shows particularly unfortunate symptoms. This wasn't a problem in land installations, because you can just use more steel,  more firebrick, and more space. at sea, you you can do none of this. In particular, you have to bend and fold the pipes in heroic ways. No problem, said Admiral Bowen, we'll just weld the pipes together. If it was pointed out to Admiral Bowen that welding didn't work very well on high-tensile alloy steels, the point was not taken up.

Fortunately, according to Admiral Bowen's 1955 autobiographical account of events, Ships, Machinery, and Mossbacks, there was no problem whatsoever. The whole thing was the most awesome success ever!

Except, well, no. Steel pieces exposed to high temperatures and stresses will eventually elongate in the direction of greatest stress. On a microscopic scale, the carbon in a piece flows in the direction of stress, congregating in areas of dislocation to maximise entropy. Or, in engineering terms,“graphitisation" occurs at alumina inclusions created in some killed steels. In December 1943, the annual meeting of the American Association of Mechanical Engineers was told of the failure of two recent Westinghouse/GE extreme steam condition power plants due to graphitisation at the weld points. (This is in the actual history of the ASME, if I recall correctly. So, big deal.) Although this result had already been detected and published in an experimental test (you'd think, considering how proud I was of tracking the citation down in the Journal of the Iron and Steel Institute, that I would have written the citation down somewhere), it was apparently only at this point that it was realised that any post-1937 United States Navy warship whose steam tubes were built of graphitisation-prone special service steels could be expected to have its steampipe welds unzip at some point in the near future.

That being said, hardly any USN ships with such plants had more than 18 months of operational service at VJ Day. There were no, or few, such events. The problem was basically restricted to some steels, perhaps (although I find this unlikely) even only those made in a few problem plants. Whatever, the FRAM refits, necessary on noise issues alone, nipped the problem in the bud. Yet, in retrospect, it really does look like Admiral Bowen walked into this with his eyes deliberately closed to the facts. Postwar investigation even poured salt on the wounds by demonstrating that high steam conditions failed to deliver on their promise. While American ships were, as Bowen pointed out, more fuel efficient than British, the advantage was provided by economisers. These had been left off the version generation of superheat-equipped ships for various technical reasons. The Admiralty had just cleared economisers for installation when the onset of the emergency building programme caused them to be set aside. The Americans, able to delay their building for a little while longer, adopted them.(iv)

So we've got Admiral Bowen kinda, sorta fibbing about a mistaken policy that he backed. Which is what you expect in memoirs. The facts are out there. But here's where the whole inventor narrative comes in. We listen to the inventor, and fail to check his facts. 

See, there was a charming young British officer in the engineering branch who sailed aboard Naiad and Duke of York when it was the flagship of the British Pacific Fleet, and a brutal disappointment to its sailors for its extremely short legs. (Not only lacking economisers, the King George Vs were designed at great expense in range to cruise at high speeds in order to better shadow German raiders.) He heard a great deal of lower decks grousing about very real problems with their machinery, and took it back to a postwar career in the engineering branch that kept rising to the penultimate ranks before being turned back. (The links are intended to imply that this might have something to do with the RN's move away from steam turbines to gas turbines, but I don't know what I'm talking about, so I'm leaving it as a speculation.) In 1967, Rear-Admiral LeBailly was laterally transferred to the intelligence branch, from which vantage point he became someone that even the Daily Telegraph's obituarist could describe as "iconoclastic." And in 1990, he wrote a memoir, The Man Around The Engine: A Life Below the Waterline that, when I tracked it down, struck me as being as undisciplined as any late-life memoir, full of ill-sourced and oddly-remembered anecdotes. 

Not that anyone would care, because it is a rare edition, and the only place that you're likely to find it, like Bowen's door-stopper, is in the footnotes to books like this. 

Oh, dear. That's a beautiful edition, isn't it? David K. Brown was a prominent warship designer and equally prominent naval author. Read Nelson to Vanguard, or any of his many other books on these subjects, as histories of naval designs, and you know at once that you're in the presence of an expert. And then you get to this.  A two-piece summary of Bowen's opinions about how awesome the destroyer machinery he created was. 

The antidote is here, by the way. Do I have to spell out the point that it turns out to be an inside job by the Gibbs of New York, as a way around the Parsons license? Although some people I greatly respect should  avoid reading Sumrall's discussion, which drags in a Former Naval Person and Gibbs relative. Bowen's blather, it turns out, is suspiciously in aid of a bunch of patent trolls. Sigh.  

That's not the whole of it, though, because I'm not done with Brown. In another place, Brown actually publishes an abstract drawing of what he calls an "Admiralty boiler," and compares it to one of its competitors. The "Admiralty boiler" drawing has less doodads on it. Clearly, the Admiralty didn't know what it was doing! What? What the hell? (Possibly somewhere in  here.)

The fact is that once he moves away from naval architecture, Brown is often just plain unserious. And it's so hard to pick and choose from Brown's work, as I want to do, to pronounce that this bit is weak and this part is fundamental and solid. What if it's actually confirmation bias speaking? That is, I love the part where Brown demolishes the idea that British shipyards were less productive than American because I agree with it a priori, and take issue with the bit about boilers because it conflicts with my anti-declinist stance?

Never mind that bit of soul-searching, because I think that history of technology has a real problem here. I've found an author who needs to be treated more carefully than perhaps some have, and revealed (or, rather, pointed out that Robert Sumrall has revealed) some patent trolls who've been pulling the "look over there" game. Not bad results for another negative research project. (In other words, Corelli Barnett made me do it.)

 Still, I'm left with some troubling thoughts. In the process of the investigation, I found scattered discussions of major technological changes that actually did come in during the war. They include significant progress in boiler control, fuel vapourisers, and reheat. Reheat's best known, although that's not saying much. 

So let me double down on the first two. These aren't just technologies of some interest to naval warship  geeks. We're talking about the places where 90% of the world's electrical power are generated. How these places changed is clearly hugely important to our society. Moreover, I'm talking about developments that are key to the progress of boiler automation. In other words, that go to the origins of modern industrial automation. Knowing more about them will probably tell us a great deal about how we backed into the computer age. Someone should research this stuff.

At the moment, no-one has. These are things that happened that history knows nothing about. And that's the kind of thing that will continue to be closed to us as long as we keep chasing inventors, as opposed to innovations.

That's why it's a good idea to do some research, and read the technical papers, and evaluate the claims on the merits and... Hey? Where's everyone going? No, don't listen to that slick salesman guy! Heck, half of his amusing anecdotes are recycled! Listen to my mumble-mouthed, muddled explanations of metallurgy and thermodynamics instead. And, by the way, don't be too alarmed if my explanations have technical errors because I'm stretching myself so thin. Trust me. My heart's in the right place. Not like that other guy's!

Damn. I'm seeing a problem here.

Totally Messed Up Endnotes That Illustrate That I Did a Lot of Work That I May Never Get To Publish Anywhere Else At the Rate Things Are Going.

iT. W. F. Brown, A Marine Engineering Review: Past, Present and Future.” Transactions of the Royal Institution of Naval Architects 102 (1960): 391—425; Tables.
iiSir John Kingcome, “Marine Engineering in the Royal Navy: A Review of Progress during the last Twenty-Five Years,” in Proc. Inst. Mech. Eng. 160 (1949): 174; Robert F. Sumrall, Sumner-Gearing-Class Destroyers: Their Design, Weapons, and Equipment (London: Conway Maritime Press, 1995); for overweight problems in American destroyer plants, see M. J. Whitley, Destroyers of the World War II: An International Encyclopedia (London and Sydney: Arms and Armour, 1988): 263, 265; for the panicked reduction in steam conditions in subsequent classes see Ibid, 290; and M. J. Whitley, Cruisers of World War II: An International Encyclopedia (London and Sydney: Arms and Armour, 1995): 270; for the culmination of experiments with superheat, see the inglorious story of the Mitschers in Conway’s All the World’s Fighting Ships, 1947-1995 (London: Conway Maritime, 1997, 2nd. Rev. Ed.); on Acheron see Rear Admiral (E)John E. Cooke, wartime Engineer-in-charge of Portsmouth Dockyard, “The Changing Pattern of Maintenance and Repair of the Machinery of the Fleet,” in Proc. Inst. Mech. Eng. 169 (1955): 935; and Admiral I. G. Maclean in discussion of Cooke’s paper, 955.
iv. F. W. Harbord and J. W. Hill, The Metallurgy of Steel (London: Charles Griffin and Company, 1918): 383–394; Edgar P. Trask, “the Use of High Elastic Steel in Ship Construction,” Trans. Soc. Nav. Arch. Mar. Eng. 43 (1942): 145; Ireland, M. L. H. W. Semar, and N. L. Mochel, “Higher Steam Conditions for Ship’s Machinery: Problems in the Selection and Application of Cycle Components and High Temperature Materials,” in International Conference of Naval Architects and Marine Engineers, 58–100 in Transactions of the Royal Institute of Naval Architects 93, 1951; Preece, Sir George. “Naval Machinery: Some Factors Influencing Its Design.” Proc. Inst. Mech. Eng. 151 (1944): 62-9..

No comments:

Post a Comment