February 28, 2018

Propulsion Part 3

The main tasks of the naval engineer once oil and turbines had been adopted were making the existing plants more powerful and more efficient. The most obvious way to resolve the mismatch between the desire of turbines to turn quickly and the desire of the screws to turn slowly was to use gears.1 The problem is that transmitting tens of thousands of horsepower through gears is not easy.

HMS Hood, the first battleship designed with geared turbines

The geared turbine first went to sea in large ships in 1916 with the three ships of the Courageous class, although they weren't technically battleships. This was necessary to allow them to make 32 kts, by far the fastest sort-of capital ships of the day. They also introduced new small-tube boilers, which were about 30% lighter for the power than the large-tube boilers in use previously. The first proper battleship followed a year later, when the USS North Dakota was refitted from direct-drive to geared turbines. But since there are very few details on that plant available,2 we'll look at the first battleship designed with a geared turbine plant, HMS Hood.3 She had separate high-pressure and low-pressure turbines on each shaft, along with astern blades in the low-pressure turbine casing. The outer shafts also had cruising turbines which could be clutched in at low speeds. At full power, the gearing reduced the 1500 rpm of the HP turbine and the 1100 rpm of the LP turbine down to 210 rpm at the shaft. Total designed power on each shaft was 36,000 hp, making Hood the most powerful ship afloat, although in practice turbines could be safely overloaded, and she made 32 kts using 151,600 shp on trials at a displacement of 42,200 tons. 24 small-tube Yarrow boilers produced steam at 235 psi, and 1,200 tons of oil could propel Hood 6,400 nm at 12 kts.

USS New Mexico, the first turboelectric battleship

The USN took a different approach to solving the efficiency problem, turboelectric propulsion. In this, the turbines turned electrical generators, and the shafts were turned by electric motors. This was first used on the battleship New Mexico,4 followed by the units of the Tennessee and Colorado classes. It was planned for the 1920s South Dakota class, and used on the Lexington class battlecruisers, two of which became the US Navy's first big aircraft carriers under the Washington Treaty. New Mexico consumed about 20% less oil at 17 kts than her direct-drive sisters, but the major advantage of turboelectric propulsion was that it allowed greatly improved internal subdivision, so the Tennessee and Colorado classes had some of the best underwater protective schemes of all time. It also meant that the motors could be near where the shafts entered the ship, reducing the chances of shaft damage causing flooding. A less obvious advantage is that turboelectric power allows the props to be reversed instantly by switching the power around, and for full power to be applied in reverse5 and sustained indefinitely.

A motor winding from USS California

Turboelectric drive was not without its drawbacks. The most serious was that it was vulnerable to shock damage, which put several ships out of action during WWII, although power was usually restored fairly quickly.6 There were weight disadvantages, but they are often overstated: Lexington’s plant was of approximately the same size, weight and power as that of Hood, while West Virginia’s machinery was comparable to that of Royal Sovereign. However, turboelectric propulsion was not suitable for smaller ships,7 and the battleship holiday meant that development was focused on geared turbines. When new battleship designs were planned in the late 20s, turboelectric lost out on grounds of weight. It would have made the ship 20 ft longer and slightly wider, and cost a 21 kt design about 0.5" of deck armor. Later, during the studies leading to the North Carolina class, a turbo-electric plant was considered, but according to Dulin & Garzke, it was discarded as 180 tons heavier than the double-reduction geared plant ultimately chosen.8 The Germans, however, revived it with the intent of using it on the Bismarcks, although they switched to geared turbines before the ships were laid down.9

The control room aboard California

During the 30s, the temperatures and pressures used in power stations ashore climbed rapidly. In the US and Germany, naval turbines were built by the same firms that built them for use ashore, and steam plants gained rapidly in power and economy. The British bought their turbines from specialist naval engineering companies, and saw their traditional edge in that field reversed. The biggest innovation was the introduction of superheated steam. When water is first boiled, it is known as saturated or wet steam, because it remains at the same temperature as the water and retains some moisture. When steam is heated above the boiling point10 it is known as superheated or dry steam. Superheated steam carries more heat per unit mass, which in turn can be extracted by the turbines.

Panzerschiffe Deutschland

In the 30s, competition to the steam turbine began to emerge in the form of the diesel engine, although ultimately no battleships were built with them.11 Diesels offered improved fuel efficiency, particularly at low speeds, but they were too heavy and bulky to compete with steam plants in battleships. The Germans did use them in their Panzerschiffe of the Deutschland class, but these were essentially cruisers fitted for long-range raiding. The Japanese originally planned to use diesels on the Yamato class, but the test installation proved so unreliable that they abandoned them and switched to steam. The Germans actually developed diesels for their planned H-class, which would have had four engines on each of their three shafts. Each diesel would have produced 13,750 hp from 9 cylinders, a 65 cm bore and a 95 cm stroke.

SS Königin Luise, fitted with hydraulic transmission

One minor note of some interest is the hydraulic transmission that was planned for the German battlecruiser Fürst Bismarck, although work on her never progressed very far. I have no details on this system, but it deserves a mention.

Next time, I'll examine Iowa's plant in detail, revealing the operating details of probably the pinnacle of battleship engineering. They'd come a long way from the days of trunk engines on the Warrior.

1 Interestingly, gearing was also used on the very first screw steamships because the engines in question turned too slowly, and had to be geared up. This went away fairly quickly as steam engines got faster.

2 I've looked pretty hard, but details are incredibly scarce. I believe this plant wasn't particularly successful.

3 Hood was essentially a fast battleship, despite the British classifying her as a battlecruiser.

4 The prototype was the collier Jupiter, which later became the USN's first aircraft carrier, Langley.

5 Normally, ships were limited in reverse power by the size of the reversing turbine, which was much smaller than the ahead turbine.

6 An evaluation of these incidents can be found here.

7 There were serious economies of scale in the equipment, so it was about equal for battlecruisers, slightly worse for battleships, and much worse for cruisers and smaller vessels.

8 Something smells off about this number. 180 tons is not a huge number when compared against the 2,580 tons of propulsion plant in the final ship. Unfortunately, Friedman doesn't give numbers on this.

9 Interestingly, electric propulsion has made a comeback in recent years thanks to improvements in power electronics. The USN was a century ahead of its time!

10 Remember that boiling point rises with pressure.

11 Diesels had been planned for the center shaft of the German Prinzregent Luitpold of WWI, but they were never installed.


  1. February 28, 2018Andrew Hunter said...

    We discussed this somewhat abortively on SSC, but: suppose we were making a large surface combatant today and gas turbines weren't an option. We'd definitely go electric (possibly diesel, possibly steam turbine driven) right? Given modern expertise and efficiency of electric drivetrains?

    (If gas turbines were an option, I think you also said that given the time to work out the details you'd prefer LM-2500s driving turboelectric to what we have now...)

  2. February 28, 2018bean said...

    Modern (post-battleship) naval engineering is a scheduled topic, but probably not for a couple of months. (I want to get more of the battleship technical stuff done first.)

    That said, there have been large improvements relatively recently (last ~30 years) in high-power electric propulsion, which means it's what I'd select for future large-ship propulsion. Warship capability is defined more and more by available electricity, and it makes it a lot easier to lay out the ship. Put some turbines in the superstructure, away from underwater damage, and other safely buried. Motors are hardened and relatively small.

  3. February 28, 2018The Fatherly One said...

    What about diesel electric? This is a proven concept for the railroad industry, has anyone tried scaling this up for propulsion of a ship?

    Also fix this sentence: During the 30s, the temperatures and pressures used in power stations ashore climbed rapidly during the 30s.

  4. February 28, 2018bean said...

    It was used on US submarines before and during WWII (IIRC, pretty much everybody else ran theirs with separate generators and drive engines until after the war, when it became ubiquitous) and on some other ships. The Cannon class DEs are the most prominent surface ships, although I believe there were others. AFAIK it was never considered for battleships.

  5. February 28, 2018tocny said...

    Do you have any recommendations for books on naval engineering, as background? I'm not really looking for a textbook, but I guess I'd be open to that too.

  6. February 28, 2018RedRover said...

    Re electric allowing full reversing power versus the turbines: what is the practical importance of this? I can see it being useful for possibly preventing collisions like the Yorktown or some of the confrontations the Navy has had in the South China Sea, but otherwise I don't really see the practical value. Maybe full effort stops to try to evade a dumb torpedo?

    Docking is obviously done very gently, and I imagine usually with lots of tugs, and the hull shape seems like it would dictate against prolonged reversed operation.

    Anyways, I was wondering if this is a large tactical advantage, or more theoretical.

  7. February 28, 2018RedRover said...

    Addenda to the above: in modern commercial aircraft, and I think in air combat operations, changes in altitude are primarily used for generating separation in cases of possible converging paths. Thus TCAS generates altitude change orders, and most traditional fighter ops trade speed for altitude to maintain energy over the opponent. Because of this, speed changes are rarely used.

    Naval warfare, especially traditional line of battle, cross the T type things, obviously has a different set of tradeoffs, both for gaining tactical advantage over the enemy and to avoid torpedos. However, it still seems like there would be a dominant strategy? Did they mostly use heading changes or thrust changes for this sort of thing? Did WWI/II subs make use of different depths for tactical purposes?

  8. February 28, 2018bean said...

    Re electric allowing full reversing power versus the turbines: what is the practical importance of this?

    You've identified pretty much all of it, actually. It's great in theory, and it did come in handy when you needed to stop suddenly for one reason or another (avoiding collision or dodging torpedoes), but I'd personally classify it more as a nice side-effect than as necessary, and definitely not a huge tactical advantage. Do note, though, that the USN came very close to using turboelectric on the Yorktown-class carriers because they wanted them to be able to back at high speeds so they could land aircraft over the bow in case of damage.

    However, it still seems like there would be a dominant strategy? Did they mostly use heading changes or thrust changes for this sort of thing?

    AIUI, it was more heading changes. Ships are rather slow to respond to throttle (and I'm not really sure how TE did there relative to more conventional propulsion) and slowing down is also usually a good way to get hit.

    Did WWI/II subs make use of different depths for tactical purposes?

    They did, although (and I'm not an expert in this), they usually alternated between the surface, periscope depth (exactly what it sounds like) and pretty deep when they were trying to hide/evade.

  9. February 28, 2018bean said...


    Introduction to Naval Engineering is one of the very best books I’ve read in terms of coverage on any subject. Excellent detail on late steam plants (it’s written about 60s-era 1200 psi steam plants, but Iowa’s are very similar) and it helped a lot when I took thermo, too.

    Unfortunately, it’s not particularly strong on history, and what I know of the later parts of the steam has been pieced together. There’s a decent book up through the early 30s called A Short History of Naval and Marine Engineering although it’s stronger in the early years.

  10. February 28, 2018RedRover said...


    That's awesome, thanks for the answers. What's the fully developed yaw rate for a battleship?

    Also, your first link to "Intro to Naval Engineering" didn't parse correctly for me. Maybe it's a misformed link? I did extract the Amazon item number manually though, and it's on my wish list.

  11. February 28, 2018bean said...


    I’ve fixed the link. Not sure what went wrong there.

    What’s the fully developed yaw rate for a battleship?

    Iowa had a tactical diameter of 814 yds at 30 kts, which comes out to 2.38 deg/sec. The Iowas were considered pretty maneuverable for ships of their size, and that’s without using differential throttle to tighten the turn. On the other hand, maneuvering of any kind scrubs speed, and putting half the engines into reverse scrubs it really fast.

  12. March 01, 2018Andrew Hunter said...

    I can second the recommendation of Intro to Naval Engineering, a truly fascinating book.

    Bean: is there anything particularly interesting to be said about the difficulty of gearing/gear manufacturing (I wasn't aware gear-cutting equipment could be in short supply?) To a non MechE, it seems like two large gears in some fixed ratio is a fairly simple construction project.

  13. March 01, 2018bean said...

    To be honest, gear-cutting isn't something I know very much about. It was definitely difficult, but I can't describe exactly why. I will look into that more and get back to you. Maybe as a column, because gearing production had a huge impact on things like the WWII escort programs.

  14. March 01, 2018bean said...

    I've looked around a bit more, and come up with at least a partial answer. The basic problem is that cutting naval reduction gears takes some large, expensive, and specialized equipment. Early gears at least were cut by hobbing, and a hob big enough to cut large gears (Hood's main gear wheel was 12' across) is not something you buy off the shelf. Building one is going to be expensive and time-consuming. The USN presumably didn't have the capability available in the 1910s, and the US was always pretty paranoid about importing that kind of stuff.

    A couple things to note:

    1. The problem was usually cutting gears of a certain size. For instance, the USN, when trying to build an escort for mass production during the Cold War, set a maximum gear diameter, which could be produced quickly in an emergency. (Of course, they blew right through the cap....)

    2. Buying more big gear-cutting machines in wartime is going to take capability from the specialist heavy machine tool builders. Their output is critical to lots of industrial expansion, and the skills involved are not easy to duplicate.

Comments from SlateStarCodex:

  • bean says:

    Naval Gazing: Propulsion Part 3. How turbines were made to work well with propellers.

    • gbdub says:

      In the post, you mention that turboelectric propulsion was not suitable for smaller ships. Why was this? I’d think scalability would rather be one of the strengths of that system.

      • Paul Zrimsek says:

        Curious about that too. Perhaps the size of the motors and generators doesn’t scale down much with decreasing horsepower?

        • Nornagest says:

          Doesn’t seem very likely. A 1 HP electric motor fits in the housing for largeish handheld power tools, a 250 HP motor fits between the rear wheels of a Tesla. And diesel submarines, where space is definitely at a premium, use electric motors too. It’d be strange if the scaling factor worked fine for all those applications but not for something cruiser-sized.

          All but submarines use just the motor, not the generator, but an electrical generator is basically just a motor in reverse.

          • bean says:

            Keep in mind that the Tesla has the benefit of a century of technical development over the TE system described here, and the hand tool isn’t specified for a duty cycle of 100%. Submarines were usually fairly slow.

      • bean says:

        As best I understand it (and this whole area is staggeringly poorly documented), the scaling factors on TE were quite strong. So the battleship plants were a little bit worse than contemporary turbines, while the battlecruiser plants were as good, and smaller plants would have been much worse. And small ships are generally more weight-sensitive than large ones, so the drawbacks would have been serious.

        • cassander says:

          There’s a fair comparison, the Buckley class, which were made with TE drives becasue of a shortage of gear cutting equipment, but I know nothing about how they worked out relative to contemporaries.

          • bean says:

            They were pretty successful, but they did have to be lengthened over the original design to incorporate the TE plant. This turned out to be a good thing, and all later DEs kept the stretch. But destroyer escorts were hardly pushing the limits of naval technology, either.

    • veeloxtrox says:

      Chiming in to say, I really enjoy your writing, please keep it up 🙂

      On a different note. Do you have plans to write an article explaining the difference between a battleship, a cruiser, and a destroyer (and maybe fast battleship, dreadnought ect.)? You mention them but I don’t have a solid understanding of what makes them different or what rolls they play in the fleet.

      • bean says:

        Try this. It should cover what you’re asking for. There’s a similar list for modern warships here.

        • veeloxtrox says:

          Overall that looks good. So to sketch my understanding of a hypothetical WWII battle between two equally sized navies. If possible, you use your destroyers you catch their battleships without their destroyers/cruisers and torpedo them. You use your destroyers and cruisers to keep your battleships from being torpedoed. You use mainly your battleships with help from cruisers to shoot their battleships and cruisers.

          That about right?

          • cassander says:

            Most naval battles aren’t two big fleets sailing out to shoot at each other for its own sake. Battles happen because one side is trying to accomplish a specific objective (enforce a blockade, send a convoy somewhere, invade a beach) and the other side is trying to stop them. Most navies have multiple things that they want to achieve, so they never really send their whole fleet out to do something, they send part of it. The essence of naval strategy/operations is really deciding which objectives are most important, which objectives the enemy thinks are most important, and trying to ensure local superiority in as many places (adjusted for importance) as possible. That is, you’re not so much trying to catch their battleships without their escorts but trying to ensure that the battles over your most important objectives are your 4 battleships against their 2 as often as possible.

            The specific tactics depend a great deal on what you’re trying to accomplish.

          • bean says:

            In the very roughest of terms, yes. Torpedoes are rather hit-or-miss, depending on how the situation goes, while big guns are more reliable. What you did with your destroyers was a tactical decision. The US tended to keep them close and use gunfire at night, which turned out to not work so well. Cruisers often substituted for/supplemented battleships.

            Of course, this ignores cassander’s point about naval operations being for a purpose, not just to have a fight, and the critical issue of aircraft, and the somewhat lesser problem of submarines.

          • John Schilling says:

            Naval operations are for a purpose, but it is very common for admirals to notice that all of their purposes would be a lot easier to accomplish if they could just win one big fight to wreck most of the enemy’s ships and leave the terrified survivors cowering in port. If you’ve got admirals on both sides thinking that way, or if admirals on one side think that way and can arrange a massive attack on something the other side has to defend, you can get yourself a very nice Trafalgar or Jutland or War Plan Orange or Midway, where there’s only a fig leaf of any sort of objective beyond “today our fleet will smash the enemy fleet on the high seas”.

          • johan_larson says:

            Japanese naval strategy was particularly focused on the idea of a single decisive battle on the high seas.


          • dndnrsn says:


            I’ve wondered if the impulse that led to the “one single battle; win by one decisive stroke” is culturally linked to the way that Japanese martial arts (or, at least, karate and judo) often have as the “best” way to win in competition scoring one full point (eg, ippon in judo). (An unfriendly person might say that Japanese naval strategy in WWII resembles the overly-complicated, requires-your-opponent-to-play-along techniques in some forms of karate, aikido, etc).

            EDIT: And now I recall that one author claims that the “one-point-win” in judo was due to the influence of the military on sporting competition in the pre-war period; I’ve been reorganizing my books and I’ll see if I can find where I read that.

          • John Schilling says:

            I’d be reluctant to attribute Kantai Kessen to kendo, judo, bushido, or any other aspect of Japanese culture or character. Because it was pretty much the mirror image of War Plan Orange, conceived for the same purpose by a bunch of Dead White American Males.

          • dndnrsn says:

            Was War Plan Orange as precise in expecting response xyz from the enemy as some Japanese plans were? If so, drat. My attempt to draw a martial arts parallel seems much less cool.

          • John Schilling says:

            Not so much “response”, because War Plan Orange was drafted by Certified Good Guys who knew that it would be the Inscrutable Orientals who started the war and took the early initiative. But it had some fairly specific expectations for what the Japanese would be doing while the USN went about setting up the decisive battle.

            And, yes, assumed that when Japanese patrols spotted the US Pacific Fleet sailing towards their Combined Fleet, the latter would steam eagerly towards the great and decisive battle to come. The bit where the US Pacific Fleet couldn’t sail towards the Combined Fleet because it was resting on the bottom of Pearl Harbor on the first afternoon of the war, we didn’t really have a plan for that.

          • dndnrsn says:

            Huh. Thanks! My knowledge of the overall strategy in the Pacific is less than in Europe. I thought it was a bit less decisive-battle focused than that.

            I take heart in the fact that while my thesis is incorrect, it’s interesting enough for the humanities. (I was going to go on to compare some imaginary “western way of war” to boxing and wrestling, and then throw a comparison of the USSR in WWII to “wall-on-wall” fistfighting because even though it’s implausible, it gives it tripartite form. Gotta have that intro-thing 1-thing 2-thing 3-conclusion form)

          • cassander says:


            The japanese also lost their war. That they were fond of the idea of settling it all at once doesn’t mean it was a good idea. It’s not a bad idea if both sides are relatively evenly matched and on similar power trajectories, but if that’s not the case, then things will get tricky for you, just like it did for the japanese around guadalcanal

          • bean says:

            Because it was pretty much the mirror image of War Plan Orange, conceived for the same purpose by a bunch of Dead White American Males.

            I’m not an expert on US strategy pre-Pearl, but I don’t think it was quite that simplistic. At some point (can’t remember exactly when), US strategy changed from “Go directly to Manila” to realizing that the Philippines couldn’t be held, and planning for a measured advance across the Pacific. Yes, even pre-Pearl. There was some planning for a big battleship showdown, but that was pretty common everywhere, and I certainly don’t think it distorted US thinking anything like the “decisive battle” doctrine did the Japanese.

  • Leave a comment

    All comments are reviewed before being displayed.

    Name (required):

    E-mail (required, will not be published):


    You can use Markdown in comments!

    Enter value: Captcha