disc

On battleships, barbettes were largely above the waterline. Centerline turrets were typically some distance from the ship side, but obviously this was untrue for wing turrets. A torpedo was unlikely to hit barbettes directly on a battleship. Their main advantage might be blocking rare fragments, for which very heavy armor would be overkill. On cruisers, the barbette tubes may extend down to a wholly underwater magazine. Combined with a narrow beam, the lower reaches on barbettes could be close to the immediate blast zone. Of course, only some cruisers had this arrangement, others having the magazine box extend up to the waterline, so that barbette armor would be less relevant. So barbette armor might be somewhat useful on some cruisers against torpedoes, but less so on bigger ships, where only frag protection would really be needed. Of all things, strong deck armor should reduce torpedo damage. It tends to choke off damage and keeps it from affecting upper parts of the ship.

There were many night battles from the Russo-Japanese War through WWII. Port Arthur, Tsushima, Jutland, and Java Sea all had big night components. I think this would be important to add to the game, but I wonder how. Command and control was very difficult at night. It was hard to tell where one's own ships were, much less the enemy's. Giving direction to a fleet may well be impossible unless it remains in constant close contact. In conditions of radio silence, things were even worse. Friendly fire was a grave threat. To search for an enemy before radar, the primary method was lookouts with binoculars. During search, essentially the primary form of illumination was moonlight and starlight, as well as whatever lights were visible on shore or on enemy ships. Search planes could help; besides searching in the darkness, these could drop flares to illuminate areas. Despite the name, searchlights were not spectacularly effective at searching out an enemy. They had relatively narrow cones, so they would need to sweep over areas carefully (the same issue as with active sonars of the time, incidentally). Searchlights also provided a bright aiming point for any enemies in the vicinity, telling them where to shoot. Once contact was made with an unknown ship, it was necessary to confirm it was an enemy. If all allied ships were accounted for, then the conclusion might be obvious. Otherwise, if it was known that allied ships were in the area, but it was not known precisely where, it might be necessary to get a positive ID, which might require closing with the unknown ship or otherwise wasting precious time and the element of shock. If it was particularly unclear, one might try sending light or radio signals -- which would also give away any surprise and might be responded to by gunfire. Or one could open fire immediately, without any identification of the target, which could have unfortunate consequences. If the unknown ship was thought to be an enemy, searchlights could be switched on to allow gunners and torpedo launchers to get a good lock. This would immediately disclose to the enemy (or unfortunate friendly) that they were under attack, and at least the general direction of an attacker. Friendly ships might remain dark and use the illuminating beam to attack, maintaining their own stealth. If lights were only switched on periodically, it might be hard for the enemy to exactly track the lighting ship. Alternately, or conjointly, guns could fire starshells to illuminate the target for a short time. The simplest starshells were simply enormous flares, but more advanced types incorporated parachutes to keep them in the air longer. If the starshell burned close to or behind the target, relative to the shooter, then the target might be outlined and the shooter relatively hidden (being farther away from the light). Allied aircraft might also drop flares during this phase, to help with spotting. The flash of gunfire might be visible to the enemy, as might gunpowder launches of torpedoes. This could allow them to get a fix on the firer. The IJN in particular used reduced flash propellants on some smaller guns (I believe only 120mm and 140mm) to reduce this firing signature; compressed air torpedo launching had a similar bonus. If hits were obtained and the enemy ship burst into flames, then searchlights and starshells might no longer be necessary, as the burning ship became an easy target. On the other hand, if there was no fire, or if it were put out quickly, it might be difficult to tell if hits were made, or what damage had been inflicted. Such short, sharp, and close encounters would favor devastating short-range weapons that could be fired from relative stealth: Torpedoes. Destroyers and torpedo boats held an edge. On the other hand, a battleship might find it relatively hard to leverage its ranged gunpower against small attackers that suddenly appeared from the darkness at point-blank distances. Hits might not stop them from attacking. Even if one attacker were dispatched, there might be others drawn to the scene, and any searchlight used to help the battleship would become a target. The advent of radar (and, to a lesser extent, long-ranged sonar) changed this whole calculus. I think night battles are incredibly important, but I think they would be hard to implement in a meaningful way. I'd like to know what others' thoughts are here.

@Steeltrap I often wonder the same thing. Were hydrophones effective in detecting and tracking torpedoes? I'd love to know of any good references on the subject. A central question I have is whether hydrophones were capable of independent search. Would they find something if not pre-alerted? I think their effectiveness at night -- ie when it is unclear if enemies are even in the area -- may be the best measure of this. A vast number of ships equipped with hydrophones were sunk by torpedoes, of course, so they were no panacea. But how many ships managed to avoid torpedoes by using their sound gear? For American hydrophones, I don't know. Supposedly the 1945 WFA set could detect torpedoes, but I don't know about earlier sets. No information given in US Naval Weapons. Best reference I can find is that MV hydrophones mounted in surface ships could detect other surface ships while in motion. An important difference from the game is that passive sonar would not generate reliable ranges, only bearings. It might be clear that a contact was approaching or receding, though, so over time, in combination with bearing rates and loudness, rough estimates could be prepared. Active sonar would be too directional to be of particular use in search, I think, but they might conceivably be useful in tracking already spotted torpedoes.

What I mean to say is that I've never heard of a battleship in our presumed time period (1890+) getting additional main guns from a refit. I don't know of any that managed to cram in an extra gun into a turret, upgrading, say, from a twin to a triple. This would be very hard to do. For the same gun caliber, an extra gun in a turret almost always requires a bigger barbette, which would cost an absurd amount of money to refit. I have certainly never heard of a nation adding a whole extra main turret in a battleship refit. There were many ships that had their main guns increased in caliber, but this got rarer and rarer as ship size increased. Some weight compensation was often needed -- bigger guns add topweight, so usually the total number of guns decreased. Large barbettes would be extremely hard to expand, which limits the size of replacement turrets, limiting the size of any potential guns. Additionally, the Washington Naval Treaty of 1921 forbade increasing gun size on already-existing battleships, except for France (which never made use of this clause) and Italy. Thus, say, we never saw the US replace their three-gun 14in turrets with two-gun 16in ones, even though it was probably possible. I think changing gun caliber was not uncommon with destroyers. Several of the US Clemson class got 5in guns, and some got twin 4in instead of single 4in. I think a whole host of early Royal Navy destroyers got extra 76mm guns to replace 57mm. As so many of these ships used pedestal mounts for their guns, replacing them was very simple. Things get more complicated with bigger destroyers in the interwar era with bigger or dual-purpose guns. With light cruisers, there are a good number of examples pre-WWI, and during WWI a large number of German cruisers traded 105mm guns for 150mm. Post WWI, a number of light cruisers, such as the C-class, Isuzu, and Delhi, entirely replaced their 140mm or 152mm main battery with smaller dual-purpose guns (102-127mm). With the big Treaty cruisers, the only examples I know of are Mogami and Tone, which traded 155mm triples for 203mm twins. The 203mm twins were actually slightly smaller than the 155mm triples, which is why this was possible. For battleships, the Italian Andrea Doria and Cavour had their guns bored out from 305mm to 320mm; these ships underwent a positively enormous refit and lost their amidships triple turret in the process. I believe the external dimensions of the bored-out guns was relatively unchanged, so no barbette expansion was needed. Post-treaty, the German Scharnhorst class was scheduled to replace their 283mm triples with 380mm twins, but this never happened. I have never heard of a refit project to reduce the caliber of a battleship's guns during our timeframe, unless the main guns were totally removed, but undoubtedly this was technically feasible.

I don't have a good answer for how the game should handle it. Increased speed, torpedo bulges, longer stern or bow, extra armor, new guns, new torpedo launchers, new or better engines, extra fuel, new masts, new bridges, new fire control, airplanes and catapults -- all were done. Only things I have never heard of are additional main guns on battleships (occasionally on cruisers and smaller) and movement, addition, or widening of armored main gun barbettes (they were sometimes removed). I also don't think armored citadels were ever lengthened appreciably, although I guess they were made taller on American missile cruiser conversions.

Yeah, the idea is that a modern torpedo protection system will stop flash and fragments from getting to the magazines, even if flooding is not stopped entirely. Older and smaller designs were vulnerable, so I think a deflagration chance should still be there for lesser anti-torpedo belts -- think HMS Royal Oak. A weakness with the current system is that repeat hits to the same spot do no extra damage. Torpedo belts are ablative, so more hits would have a compounding effect. There's little implementation of flooding instability, either, a major (perhaps the most important) cause of loss.

I think it is important that if you choose to move at max speed, you start running out of fuel quick. Fuel consumption goes up more-or-less exponentially with speed. @Shaftoe brings up the important point of cruising speed. There are other issues. For example, reciprocating steam engines don't deal well if pushed to max power 24/7. Turbines are inefficient at lower powers, so we see dedicated cruising sets. There was impetus for extra fuel stowage in US submarines, for example, because extra fuel allows a faster cruise speed and proportionally more time at the patrol destination -- even though the "extra range" was not necessary itself. High-powered Diesels were desirable partially because they had to have their bugs ironed out, so they would probably be more reliable at cruising speed.

You're right, only a few others with >127mm -- Brooklyn / St. Louis, Worcester, Duca degli Abruzzi, and Mogami (prior to rebuild). I'd still like the choice. It would be nice to try heavily armored small ships. Instead of a hard stop, I think there should be escalating costs for heavier armor, though.

I think we're going about this the wrong direction, personally. I think "towers" should be subdivided into separate parts. Right now it's two big blocks that limit turret and funnel placement. We're beholden to these monoliths, and thus on the superstructures the devs have introduced and are planning on. Makes designs look the same. I think it should have a Lego-like approach. We could have different blocks: deckhouses, bridges, conning towers, masts, fire control tops, secondary directors, and searchlight supports. You slot them together and get a custom result. Want an heavier fire control top? Use a stronger mast or place it lower, where its field of vision is worse. Or, have a light one atop the mast and a heavier one on the conning tower. Want a better bridge? Build up a bunch of platforms on the mast and make your own pagoda, or start from scratch with a big block like the King George V class. Crucially, blocks could overlap to some extent. No more "wrong-size" barbettes -- just slot in a different one instead. On those hulls with cut down quarterdecks, the barbette could overlap with the hull structure, so no awkward squeezes there either. Issue is that this would take a lot of work.

I think the desire for new destroyer hulls is long-standing, so hopefully a couple are in the works. I agree that IJN cruiser hulls would be good. I think DE hulls make sense if submarines are also implemented -- if only in abstract would be fine. I think your last point would be best in a pure sandbox mode, where there are no restrictions whatsoever. Custom battles have been a half step in that direction.

I think hydrophones of the style installed on Yamato and the heavy cruisers would be omnidirectional, with phase-cancelling used to find a bearing. So that would be nice against torpedoes. I may be mistaken on that account, but that's my interpretation of E-10. Active sonars would be searchlight types with a narrow sound beam. Bad for search (ironically) but good for tracking contacts. Possibly they could be operated in a passive manner without echo-ranging, depending on receiver directionality. I would question their utility against a torpedo spread. The US evaluators found Japanese hydrophones and sonars to be pretty modern and effective. Seems the big downfall was translating effectiveness into a coherent submarine tracking and attack system. Do you know of instances where Japanese ships heard torpedoes during an attack? I am by no means denying it happened -- I just don't personally know of examples. @Shaftoe I am sorry if I said something foolish or offensive. As E-10 says, Japanese hydrophones were stationary and the sonars had nonretractable domes. So the devices theoretically could be operated at any speed. But increasing noise at higher speeds likely would make this less feasible.

I do not know how the Type 0 or the similar Type 93 would usually perform, or if they were particularly useful at picking up incoming torpedoes, but I think they didn't help as much as might be hoped. It does appear they were regularly operated at sea, but I don't know if they were reliable, if they were operable at high speed, or if the operators were competent. Many Japanese cruisers used the Type 93, but many of these were torpedoed and sunk. Obviously hydrophones are of little avail against aerial torpedoes or point-blank submarine shots, but at least Haguro was sunk by surface torpedoes. Maya was sunk by multiple submarine torpedo hits at about a range of ~1.5 nautical miles (~2.8km), after having witnessed Takao and Atago get hit ~24 minutes prior -- ie obvious forewarning of submarines was there.

@HusariuS was very faithful to the source material. From Skulski, Anatomy of the Ship Yamato: "Yamato was equipped with Type 0 sonar arrays which could detect a submarine when the ship was dead in the water or proceeding at low speed." The diagrams Skulski provides show the apparatus was placed in the forefoot, in the ship's bulbous bow. That's it, unfortunately. No other information that I know of on Yamato's particular setup or its usefulness during various actions. Lengerer and Ahlberg's book on Yamato does not mention the sonar -- though I reckon this is due to its more narrow scope. We do have more information on the Type 0 installation. https://pacificwararchive.wordpress.com/2018/04/21/reports-of-the-u-s-naval-technical-mission-to-japan-1945-1946/ See E-10, Japanese Sonar and ASDIC. The Type 0 was a passive (ie hydrophone) 30-element double-ellipse (4m across the long axis) array using moving-coil receivers (I have not seen a picture of the hydrophone array, but my impression is it might look like two nested ellipses of individual receivers set around the bottom). It had an accuracy of 3 degrees and a sensitivity of 35db. Detection ranges are not specified, but the broadly similar Type 4 could detect a slow moving submerged submarine at 1000 meters, or a torpedo at 6000 meters, the scanning ship sailing at 12 knots in either case. The Type 93 was a very similar device to the Type 0, more common, with 16 elements in an ellipse array using moving-coil receivers, with an accuracy of 5 degrees and a sensitivity of 35db.

No, I never have. I think many of the issues people post about trace back to the primitive damage model in place. I've been wondering how to rework it into something more realistic and accessible.

@Steeltrap Is this the damage model you are referring to? What did you find valuable about it? Any glaring deficiencies? I agree with the spirit of @baltic1284 's post. Should be able to give light cruisers additional armor, if so desired. US 6in-gun light cruisers from Brooklyn to Worcester had 6.5in turret faces -- not possible in the game. I would go the "opposite" way for battleships. Several pre-Dreadnought classes had only 6in belts, as their thin Harveyized and Krupp-style armors were so much better than thicker compound belts. We should be allowed to use minimal armor ourselves.

The question is whether a 17% increase in weight is justification for the damage scaling. I think we need to see damage numbers to know.

This page specifically has details on powders by nation. http://www.navweaps.com/index_tech/tech-100.php I posted the passage from Japanese Cruisers of the Pacific War here.

From Lacroix and Wells, Japanese Cruisers of the Pacific War, Appendix H. Covers the major powders used by the IJN from inception to 1945. Page 767: Page 768: Page 769:

I like the idea, and I've been working on a system myself. Might add it once I've come up with a mutually approved arrangement. I think a key element is individualizing compartments per ship.

Japanese Cruisers of the Pacific War by Lacroix and Wells is superb. Appendix H gives a fine history of Japanese propellants, from black powder to 1945. Campbell's Naval Weapons of World War Two has great information on propellants of the seven major navies: UK, US, Japan, Germany, France, Italy, and Russia.

It certainly did occur, though I do not think it was particularly common. It seems that big vessels and those with good torpedo protection were fairly safe. I am aware of the following carriers, battleships, and cruisers which very likely experienced a magazine or fuel detonation or deflagration immediately after being torpedoed or mined. All sank, almost universally with heavy loss of life, except for USS New Orleans. I have not included any destroyers or other types of ships. Note the relative lack of heavy fleet carriers, Dreadnoughts, and super-Dreadnoughts. Carriers: Unryu, fuel tanks and suicide plane warheads detonated. Shinyo, massive fuel deflagration. Taiyo, fuel deflagration. USS Wasp, fuel deflagration with possible magazine deflagration. USS Liscome Bay, titanic magazine detonation. Battleships: Hatsuse, pre-Dreadnought, mine detonated forward magazine, sank in 90 seconds. Petropavlovsk, pre-Dreadnought, mine detonated magazines and probably forward torpedo tube, sank within two minutes. Suffren, pre-Dreadnought, sank in seconds with no survivors. HMS Royal Oak, super-Dreadnought, cordite deflagrated. Cruisers: Pallada, detonated immediately with no survivors. Zhemchug Takasago, mined SMS Prinz Adalbert, sank immediately with three survivors Armando Diaz Java, rear magazine detonated HMS Pathfinder USS Juneau USS New Orleans, survived. Apparently the black powder magazine detonated -- probably the same mechanism behind USS Arizona's loss. The following ships may have experienced magazine detonation (most likely several did), but it is unclear. Akitsu Maru, amphibious assault carrier, huge loss of life but unclear if detonation occurred. Fuso, Dreadnought battleship, not clear if magazine was detonated or if bow separated during sinking. HMS Majestic, battleship, sunk by "huge" explosion with torpedo hit -- but with few killed. Tama, cruiser, split in half with no survivors Cesare Rossarol, cruiser, split in half by mine. Amiral Charner, cruiser, sank in two minutes with one survivor. Takachiho, cruiser, three survivors. SMS Bremen, cruiser Nachi, cruiser, forward magazines detonated, but exact cause (ie bomb, torpedo, or fire) is unclear due to big number of hits. Naka, cruiser, broke in two, either a bomb or torpedo. Here's a painting of HMS Pathfinder blowing up after torpedo hit.

For more context, claimed firing cycle for the Japanese 16in gun on Nagato was 21.5 seconds, and that was a conventional screw-breech bag gun.

Full speed crash stops are typically performed on trial runs. I do not know about potential damage to the drivetrain, but I suspect it would be unlikely unless many repeated stops were made. Cavitation on the props would be the big concern, I think. For the trials of the 224,950 tonne steam-turbine tanker Esso Malaysia: Traveling at an initial speed of 16 knots, a crash stop took 17.6 minutes with an ahead reach (distance traveled forward) of 4389 meters. A turn at 16 knots would have an ahead reach of only about 900-1000 meters. For the 26700 deadweight-ton single-screw steam-turbine tanker Esso Lima, a crash stop from full speed (presumably about 10-18 knots) required 8 minutes with an ahead reach of 1463 meters. The prop was stopped within 32 seconds of crash-stop initiation, and it was in full astern RPM by 1 minute 37 seconds. There are issues with crash stops. Turning is typically faster, easier, and requires less ahead-distance. Stopping can take time to enact: With steam turbines closing and opening the throttles may take ~20 seconds. Stopping will markedly reduce directional stability. For the Esso Malaysia, the ship ended up 503 meters to starboard and perpendicular to the original direction of travel. Not ideal. Additionally, the ship will take a long time to reaccelerate to its original speed, unless it has extremely powerful engines. Crash stops can be useful, however, when there is insufficient room to maneuver, or when the ship is already traveling slowly. Principles of Naval Architecture gives a figure of 6 knots as a cutoff for single-screw ships. I suspect it would be higher for multi-screw ships with powerful astern power.

Most ships can "crash stop." This might be defined as reversing the prop revolution in an attempt to stop the ship from speed. On a steam engine, changing the valve timing will reverse the engine, which will have a powerful braking effect on the prop shaft. For a steam turbine ship, the great majority have reverse turbines. Steam is cut off to the main turbines and introduced to the reverse turbines. On many (most?) ships there is no reverse clutch, so it does not need to be clutched in -- always attached to the prop. Electric transmissions can rapidly switch from full reverse to full ahead; usually the limiting factor is heat dissipation. Diesels use air-injection to brake, similar to reversing steam engines. I do not know if this was possible on the vessels of our era, but I suspect it was given that many used air-blast fuel injects. For a controllable-pitch prop, changing pitch will work. Likewise, a turnable prop pod can spin. Rudder cycling will slow the ship, though I am not sure it is particularly useful. Water brakes were trialed on US battleships -- and probably a bunch more commercial ones -- but they didn't work very well. Here's a picture of one on USS Indiana.

To clarify, the game calculates the chance that at least one shell in a salvo will hit. The formula is = 1 - [ (1 - x) ^ y ] , where x is the chance for an individual shell to hit and y is the number of shells per salvo. For example, a battleship firing 9 guns at 5% accuracy has a 1 - [ (1 - 0.05)^9 ] = 37% chance of hitting with at least one shot in a salvo. Convesely, there is a 63% chance that no hits are obtained in that salvo. The cumulative model used by @supershanks is not displayed in the game. It calculates the average number of hits given a certain number of shots. For a 9 gun salvo at 5% accuracy, the cumulative probability is just 9×0.05=0.45. This number gives the expected quantity of hits for nine independent shots. Only an average of 0.45 hits are expected, ie less than one. If we assume the gun fires 200 times, then the value is 200×0.05=10. Thus, an average of ten hits.