Comment on Rebuilt US Battleships

Discussions about the history of the ship, technical details, etc.

Moderator: Bill Jurens

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Comment on Rebuilt US Battleships

Post by RobertsonN » Sat Apr 18, 2020 10:31 pm

This post is based on two of Friedman's books: (1) Battleship Design and Development 1905-1945 and (2) US Battleships.

On p. 94 of (1) there is a plan view of the hull subdivision of Idaho (1925). This had a mixed system of subdivision, mostly having main compartments divided into three by two longitudinal bulkheads. But the three boiler rooms were not so subdivided. On pp. 443-444 of (2) the stability characteristics of Tennessee are given as of 1921. For a beam of 97.5 ft, at normal displacement GM = 3.885 ft (1.19 m) giving GM/B = 0.04. As with many other ships of this period this ship had a low GM, which would have given it a long roll period and made it a very stable gun platform. However, this low GM would have made it prone to capsizing given serious flooding on one side, excepting at the boiler rooms, given the layout of subdivision employed.

The hull subdivision of the rebuilt New Mexico shown on p. 94 of (1) indicates the main compartments were all divided in three by two longitudinal bulkheads that ran the full length of the citadel. The stability characteristics of the rebuilt Tennessee in 1943 given on pp. 444-445 of (2) show major changes from the original ship. Now of 114 ft width at normal displacement GM = 11.45 ft (3.49 m) giving GM/B = 0.1. This ship now had a very high GM, which would have likely resulted in a relatively short roll period and made the ship a lively gun platform. But now its much higher initial stability would have allowed it to absorb much more off-centre flooding and the heel in turning would have been less so that the lower edge of the belt would have been less likely to have been exposed.

The original ship had stability characteristics that were very similar to those of the Royal Sovereign but the rebuilt ships had stability and subdivision characteristics that were very similar to those of the Bismarck (which is why this thread has been placed on the Bismarck General Discussion). It would have been very difficult to redesign the later fast US battleships to have had similar subdivision and stability characteristics to those of the rebuilt Tennessee without a significant reduction of their speed.

This is not to say that the US designers were directly influenced by the two ships mentioned. They may well have arrived at these characteristics from similar considerations to those of the foreign designers though,

Neil Robertson

Bill Jurens
Moderator
Posts: 739
Joined: Mon Oct 18, 2004 4:21 am
Location: USA

Re: Comment on Rebuilt US Battleships

Post by Bill Jurens » Sun Apr 19, 2020 5:58 pm

It is my general impression that the stability characteristics of Bismarck were largely defined by her rather large beam, which was in turn driven by a need to minimize draft. With length also limited by available building ways and draft fixed at a rather small value, the only ways to increase displacement would have been to increase the block coefficient (not ideal if speed was important, which it was) or to increase the beam. Although the situation varies a bit from design to design depending upon the precise shape of the hull and the center of gravity, as a general rule for most ships at a given displacement a very small change in the beam can cause a relatively large change in the metacentric height. At 100 foot beam with normal hull forms, the ratio is typically around 12:1, i.e. a 1% change in beam can result in a 12% increase in metacentric height. So, for an initial metacentric height of 12 feet, increasing the beam by one foot in one hundred, i.e. one percent, might be expected to increase the metacentric height twelve times that, i.e. to 1.12 x 12 = 13.4 feet.

That's a fairly big 'lever' to consider...

Bill Jurens

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Sun Apr 19, 2020 8:08 pm

The formula for displacement includes the product

CB x B

where CB is the block coefficient and B the beam. It follows that B can be traded off against CB. Friedman gives a CB of 0.55 for Bismarck, which gives about the correct figure for the normal displacement from the formula in metric units of

D = CB x L x B x T x 1.015

where 1.015 is the density of west Baltic seawater and L = 241.5, B = 36 and T = 9.33.

Bismarck had a high B but a low CB. Friedman gives CBs for Vanguard and New Jersey as 0.6 and 0.59, respectively. If Bismarck had had the New Jersey figure then B could have fallen to 33.55 m (110.1 ft) and for the Vanguard CB it would have fallen further to 33 m (108.25 ft). It has been said that width was determined by the sum of machinery width + torpedo defense width. But the Scharnhorst was only slightly less powerful and had a side protection system of the same nominal value, but that ship was only 30 m wide. The overall length of the machinery compartments in Scharnhorst was not more than that in Bismarck.

Much to my surprise I have been able to order copies of Kriegschiffsbau (1943) by Evers and the two volumes of the same title by Hadelar (1968), from which Friedman obtained his data on German ships,

Neil Robertson

hans zurbriggen
Member
Posts: 82
Joined: Sat Dec 21, 2019 8:15 am

Re: Comment on Rebuilt US Battleships

Post by hans zurbriggen » Mon Apr 20, 2020 2:08 pm

Hello,
to Mr.RobertsonN writing: "...It has been said that width was determined by the sum of machinery width + torpedo defense width. But the Scharnhorst was only slightly less powerful and had a side protection system of the same nominal value, but that ship was only 30 m wide. The overall length of the machinery compartments in Scharnhorst was not more than that in Bismarck."
Torpedo defense system in Scharnhorsts was max 4.7 meters wide at midship. In Bismarcks had a 5.5 meters width. As consequence, TDS was greaterly enhanced from the 'twins' to the 'beasts', thanks to extra-large beam.
Despite similar length, machinery layout was different and much less cramped on Bismarcks: 2 boilers were aligned transversally in Scharnhorsts at midship without longitudinal bulkhead (4 boilers per room, 3 rooms). 3 boilers were aligned in Bismarcks at midship separated by longitudinal bulkheads (2 boilers per room, 6 rooms).
IMHO, had not the German ports and Kanals restricted dimensions, Bismarck could have been built with beam (and length) even more than 36 meters, to ensure effective underwater protection also toward bow and stern of vitals (where width decreses).
High GM of Bismarck seems not to have affected her performance as gunnery platform, at least according to all sources/accounts.

hans

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Mon Apr 20, 2020 4:52 pm

Yes. I am aware that Raedar thought that for the first time in Bismarck the individual machinery compartments were of an adequate size. The boiler room subdivision of Scharnhorst was considered poor when this layout was one of a number considered for Bismarck in the design stage. That was certainly so for underbottom explosions. But I would have thought it was better for the normal type of torpedo hits on the side because in Scharnhorst the boiler rooms all had outboard wing compartments. The 14 in hit on Bismarck from POW that went under the belt would probably not have flooded a boiler room but just the wing compartment. The 1941 German report comparing Bismarck with Richelieu gave as one of the advantages of the French ship its considerably more compact machinery layout, which allowed a wider TDS albeit at the expense of inferior subdivision. In addition, with a 36 m hull the Scharnhorst layout would have allowed a TDS up to 2.2 m wider (up to 7.7 m wide) given the figure of 4.7 m TDS width in Scharnhorst you give. In Bismarck there would have been no need for the dome over the boiler rooms as in Scharnhorst and indeed there would have been slightly more room over the boilers. On the other hand, some of the unreliability problems experienced with Scharnhorst may have arisen from the higher steam conditions (temperature and pressure) compared with Bismarck. So the list of pros and cons was long.

As far as I am aware Moltke was the German ship with the relatively largest beam (B/D^0.333). Moltke had a beam of 29.4 m and a normal displacement of 22979 mt). Expanded to the Bismarck with a normal displacement of 45951 mt would have given a beam of 37 m. Moltke had an even finer hull form than Bismarck: Friedman gives CB = 0.46. So, as you say, a Bismarck with a greater beam and a lower CB was within the realm of possibility, ignoring lock limits. Bismarck would have needed a CB = 0.535 to have had the same draft had it had a beam of 37 m.

The old US battleships had an ideal subdivision by two longitudinal bulkheads throughout as in H39. However, their plans show many discontinuities in these bulkheads to evidently suit the space requirements of individual compartments. These discontinuities may have been a weakness from the structural point of view,

Neil Robertson
Last edited by RobertsonN on Mon Apr 20, 2020 5:23 pm, edited 1 time in total.

User avatar
marcelo_malara
Senior Member
Posts: 1240
Joined: Sun Oct 02, 2005 11:14 pm
Location: buenos aires

Re: Comment on Rebuilt US Battleships

Post by marcelo_malara » Mon Apr 20, 2020 5:08 pm

A block coefficient of less than 0.5 means two things:

-concave lines
-rise of floor

Is this possible in a warship?

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Mon Apr 20, 2020 5:44 pm

I don't know much about CBs. But the figures given by Friedman do not give the normal displacement of Moltke. Furthermore, he gives the draft of Moltke as being 9.2 m at normal displacement. The generally accepted figure was 8.2 m (given in Griessmer, for example). The value for CB required to give the correct normal displacememt of Moltke is 0.505, so that is just over 0.5. The floors of all the WW1 German battlecruisers did rise at the front in the region of the bow torpedo tube.

Incidentally, I have since verified the CB for New Jersey of 0.593 given by Friedman in his US Battleships book as being consistent with its correct 'normal' displacement of over 55000 tons. Later US battleships had normal displacements not far off their full load figures and markedly higher than was the case in most other countries. That is also consistent with their practice of refuelling much more frequently than necessary (see Friedman book on Battleships and the book by Muir about the Iowa class) on the basis of their large fuel capacities in order to maintain the value of their side protection systems. It would have also helped against diving shells,

Neil Robertson

hans zurbriggen
Member
Posts: 82
Joined: Sat Dec 21, 2019 8:15 am

Re: Comment on Rebuilt US Battleships

Post by hans zurbriggen » Mon Apr 20, 2020 9:21 pm

Hello,
to RobertsonN writing: "The 14 in hit on Bismarck from POW that went under the belt would probably not have flooded a boiler room but just the wing compartment."
not sure of this: according to survivors, PoW hit exploded against the 45mm Ww torpedo bulkhead (5.5 meters TDS wide). On Scharnhorsts a similar hit would have exploded 0.8 meters inside the vitals (4.7 TDS width) after having penetrated intact the torpedo bulkhead. Splinters would have destroyed the thin (non armor grade) bulkhead between the wing compt. and the boiler room, immediately flooding a 4-boilers room and possibly destroying also other bulkheads (if unlucky, such hit could have flooded 2 boiler rooms (8 boilers) + 2 wing compt. at same time). The damage would have been much worse. It was the large beam of Bismarcks that prevented such damages, with only a 2-boilers room lost.

hans

edit: the 'wing compt.' on Scharnhorsts were generator rooms and auxiliary machinery rooms, not TDS void compt.

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Mon Apr 20, 2020 10:23 pm

As far as Scharnhorst was concerned what you say is true. But use of the Scharnhorst layout in Bismarck would have allowed for the distance from the hull side to the torpedo bulkhead to have been increased to up to 7.7 m (it could have been less than this had the width of the boiler rooms and/or the wing compartments been increased slightly). In this case, the shell might have detonated before it struck the torpedo bulkhead or it would have detonated on impact as in the actual case. In that case, assuming similar damage as in Bismarck, the wing compartment would have flooded but the boiler room would have been OK. Also in the more likely case of a torpedo hit, the presence of a wing compartment could have prevented a boiler room from flooding.

On the other hand, a shell or bomb detonating above the armor deck over a boiler room (as occurred in the Bismarck's final action) would have had more serious consequences in the Scharnhorst arrangement. A consequence of using the Scharnhorst layout in Bismarck with its further inboard torpedo bulkhead is that the slope of the armor deck would have been at a smaller angle to the horizontal, which would have made it even harder to penetrate.

The Scharnhorst arrangement with its off-centre quite narrow wing bulkheads would have given a smaller list if flooded, particularly in a ship with a lower value of GM/B (about 0.08 in Scharnhorst compared with 0.11 in Bismarck), compared with the actual Bismarck arrangement with its large off-centre compartments. The German battlecruisers of the First War period showed both the Bismarck arrangement in Moltke and Seydlitz (though the narrower forward boiler rooms were undivided) and the Scharnhorst arrangement in Derfflinger, though with a centreline bulkhead. As the danger of capsizing in old armoured cruisers with centreline bulkheads came to light, the Germans cut holes in these centreline bulkheads which were normally covered by slides (as related on the volume of Koop/Smolke on German dreadnoughts of the Nassau to Konig classes), which could be opened in an emergency. In the battlecruisers starting with Mackensen and the later Schnelle Grosskampfschiffe larger boilers were necessary and all the boiler rooms were unsubdivided as in later US ships. Even the finally selected ship of the line L20ealpha had such unsubdivided boiler rooms as it was significantly faster than earlier German dreadnoughts. Contrary to what some people believe German ships showed quite a design variety,

Neil Robertson

User avatar
marcelo_malara
Senior Member
Posts: 1240
Joined: Sun Oct 02, 2005 11:14 pm
Location: buenos aires

Re: Comment on Rebuilt US Battleships

Post by marcelo_malara » Mon Apr 20, 2020 11:11 pm

The calculation of the Cb must be properly done, the waterline length must be used along with the corresponding draft and displacement. Estimating it from overall length can lead to a lower than real value.

Regards

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Tue Apr 21, 2020 12:34 am

Yes. For the ships I have mentioned I have used the waterline length at the normal displacement. For some ships the waterline length is not given and so no accurate figure can be calculated. It is also the case that the block coefficient usually changes with draft and so I presume that the stated CBs are the values at the normal displacement. For later ships with vertical sides amidships, CB usually increased with increasing draft. But many earlier ships had some degree of tumblehome amidships, which meant that their BCs may have fallen with increasing draft. For ships with bulges the variation of CB will depend on the shape of the bulge and how far up the side of the hull it terminates. The fact that the Americans defined normal displacement as a higher percentage of the maximum displacement than most other nations means that their CBs may for later ships be a little overstated in relation to those of other nations. Often things are fairly simple at first sight but the more you think about them the more complicated they become,

Neil Robertson

hans zurbriggen
Member
Posts: 82
Joined: Sat Dec 21, 2019 8:15 am

Re: Comment on Rebuilt US Battleships

Post by hans zurbriggen » Tue Apr 21, 2020 7:54 am

Hello Mr.RobertsonN,
I agree both layouts had their pros and cons. However Scharnhorsts had no additional TDS layer compared to Bismarcks. The 'wing compts' were 100% vital compartments (boiler auxiliary machinery and electrical plants), not intended to provide additional defense to boiler rooms.
Boiler rooms in Scharnhorst were too large (with risk of important flooding), thus decision to have them placed more close to centerline and surrounded by smaller vital compts. In Bismarcks, with much smaller boiler rooms (6 instead of just 3) the compatimentation was much better and there was no such need.
With a width of 5.5 meters, the TDS of Bismarck withstood all (small) aerial torpedos and, according to survivors, also large ones from British cruisers during last battle. No need therefore to 'waste' space to have a larger TDS, possibly this need was only for TDS around magazines, where width is smaller (in this case Bismarck had 'wing compts' as well as Scharnhorsts (just passageways, not vitals, in this case).
In addition, propulsion plants of Scharnhorsts were extremely prone to problems, while Bismarcks' were not: I do not know if this was due to inefficient layout and/or to other reasons.
I agree however that Bismarcks machinery was wasting some space, but this was due to initial decision to install turboelectric propulsion. When traditional geared turbines were finally selected in 1936, then it was too late to efficiently redesign the use of all the space.

hans

RobertsonN
Member
Posts: 192
Joined: Wed Jan 05, 2011 9:47 am

Re: Comment on Rebuilt US Battleships

Post by RobertsonN » Tue Apr 21, 2020 11:48 am

I have just received the copy of Kriegschiffsbau by the naval architect H. Evers. Considering it is 77 years old it is in very good condition. The values for the block coefficients given in this book of WW1 British and German ships are often different from the ones Friedmann gives.

On the subject of TDS width not much in a ship became as out of date so quickly due to increasing torpedo and mine charges and more powerful explosives such as S1 and Torpex. For example, the Queen Elizabeth class reintroduced complete torpedo bulkheads in RN battleships, tested up to 280 lbs of wet gun-cotton. Nevertheless, it was found necessary to bulge these ships within a couple of years of their completion. The 1941 German comparison of Bismarck with Riehelieu said that the distance from the outer hull to the torpedo bulkhead should not be less than 6 m and that if was less than this that the torpedo bulkhead should be thicker than in then current practice,

Neil Robertson

hans zurbriggen
Member
Posts: 82
Joined: Sat Dec 21, 2019 8:15 am

Re: Comment on Rebuilt US Battleships

Post by hans zurbriggen » Tue Apr 21, 2020 12:59 pm

Hello Mr.RobertsonN,
I 100% agree: requirements for TDS were quickly evolving but Germans provided increasing TDS width from 4.7 (Scharnhorsts) to 5.5 meters (Bismarcks). Was it enough ? Difficult to say. Requirements for 6 meters width were recommendations: I remember the Scharhorst belt was initially recommended to be 350mm just to face the 330mm gun of Dunkerque (!), however, 320mm proved to be more than enough against 406mm shells while in 1939 H class design the belt would have been further reduced to 300mm.
Operation experience proved that Bismarck was not vitally impaired by the British torpedoes striking the TDS, while she was doomed by a single (small) torpedo hitting out of the TDS.

hans

Post Reply