The problem of the Diesel-engined capital ship was considered in an article published in the July, 1929, number of the Naval Institute Proceedings. When this article was written the Diesel-engined cruiser seemed a remote possibility. Since that time much has happened, and it is the purpose of this study to make a survey of the subject, based on the situation existing at this time, January, 1930.
Characteristics of Existing Cruisers An important phase of the subject is the effect on type of fitting Diesel engines. This involves a general survey of the problem of cruiser design.
The conditions imposed by the Washington treaty provided a working capital of 10,000 tons without water and fuel. In the solutions arrived at by the treaty-power designers there is a general similarity of type. The battery, limited to guns of eight inches caliber, varies between eight and ten guns for the main battery. There is, however, a wide variation in the maximum power, which results in a wide variation in the weight available for protection.
The characteristics of late cruisers of the several treaty powers, as given by Brassey, are as follows:
In the table below the German solution permitted under the terms of the treaty of Versailles has been added. In spite of the heavy battery of 11-inch guns, this ship must be considered as falling in the cruiser class, coming at the top of the protection and armament scale, and at the bottom of the speed scale. The speeds given, in the last column below, have been adjusted to the speed curve on Fig. S. If the power of the Ersatz Preussen has been correctly stated it is evident that the speed will be nearer twenty-eight knots than the stated speed of twenty-six knots.
It is by no means certain that the heavy guns permitted to the Germans give a superiority in gun power.
The Problem of the Battery
The greater striking power of the heavier projectile must be balanced against the greater volume of fire with 8-inch guns.
If the same weight be allotted to the 8-inch battery, a materially greater weight of projectiles can be delivered in a given time with an 8-inch battery, than with an 11-inch battery. This follows from the law of similitude, first stated by Normand in 1896 with respect to reciprocating engines.
This law which holds for all types of machinery, with slight modifications imposed by practical considerations, may be stated as follows:
In similar machines the weight varies as the cube of the linear dimension, the output varies as the square of the linear dimension and, therefore, the weight per unit output varies directly as the linear dimension.
One of the conditions of similarity is that the rate of working remains the same.
In the reciprocating engine this rate is called piston speed. In the gun it may be taken as a factor which bears the same relation to shots per minute that piston speed bears to revolutions per minute.
The conclusion from these assumptions would be that the rate of fire would vary inversely as the caliber. As a matter of fact all evidence supports the conclusion that a materially higher “piston speed” is possible with an 8-inch gun than with an 11-inch gun.
The tables of British and foreign ordnance in Brassey indicate that the rate of fire of the 8-inch gun will be about two and one-half times that of the 11-inch gun.
With a conservative figure of two, the comparison between the main batteries of the most heavily armed Washington treaty cruiser and the Ersatz Preussen would be as follows:
In view of the impossibility of providing protection sufficient to resist the direct impact of an 8-inch projectile, without making a sacrifice in speed which would remove the ship entirely from the cruiser class, the decision made by the Washington Conference would seem to have been a sound one. In view also of the practical unanimity of the treaty powers as to the number of guns, and the importance of the space factor, the compromise of the U. S. designers on nine 8- inch guns would seem to represent a decision not open to revision.
The Speed Question
The weights of the hull are fixed; when reduced to the minimum necessary for strength and seaworthiness they cannot be changed. If the battery weights are assumed to be established by the adoption of a main battery of nine 8-inch guns, the problem is resolved into the determination of the distribution of the weight remaining between speed and protection.
This involves a determination of a weight schedule, a matter of some difficulty since there is so little authentic information available. Two estimates have been found which seem to be sufficiently accurate for the purposes of this discussion.
One of these made by a German writer, comparing a British cruiser with the Ersatz Preussen, was published in Werft Reederei Hafen of February 7, 1929. The other was given by Mr. James L. Bates of the Bureau of Construction and Repair in a discussion before the 1927 meeting of the Society of Naval Architects and Marine Engineers. These estimates are as follows:
There are certain inconsistencies in the above table, but those which are material are probably due to different methods of weight classification. In the British and German ships there are very material items of machinery carried under engineering which by the American practice are divided between Construction and Repair and Ordnance.
| Ersatz Preussen | Cumberland | U. S. Cruiser |
Hull | 3,600 tons | 4,300 tons | 5,100 tons |
Protection | 2,700 tons | 2,000 tons | 1,000 tons |
Engineering | 1,600 tons | 2,300 tons | 2,300 tons |
Ordnance | 1,700 tons | 1,000 tons | 900 tons |
Equipment | 400 tons | 400 tons | 700 tons |
Total | 10,000 tons | 10,000 tons | 10,000 tons |
Water | 40 tons | 380 tons | — |
Fuel | 3,500 tons | 3,460 tons | — |
It is known that in all of these ships a very considerable amount of the protection is in the form of special-treatment steel worked into the hull structure, but it is not known how this is classified in the weight schedule between hull and protection weights. Based on the above the following schedule has been adopted and put into a form better suited to the purpose:
Fixed weights
Hull 4,500 tons
Hull fittings 500 tons
Fixed engineering weights 400 tons
Equipment ... 400 tons
Total fixed weights 5,800 tons
Battery weights 1,300 tons
Balance 2,900 tons
In the above schedule the item of fixed engineering weights was arrived at by making an estimate of the weight which would remain in a steam-driven 10,000-ton cruiser, after all machinery for propulsion, and all auxiliaries and fittings necessary to its operation, are removed. This item includes all engineering auxiliaries for ship service, and a weight sufficient for all engine exhaust pipes in a Diesel installation. The item for hull weight is liberal, when compared with the Ersatz Preussen, and is sufficient to include aviation and other miscellaneous weights. The item for ordnance is that estimated for nine 8-inch guns, eight 5-inch guns, and the ammunition for both.
Based on the above schedule the curves shown on Fig. 1 have been plotted showing the relation between propulsive machinery weights and protection weights for various speeds. These curves are based on the power curve on Fig. 5, and a weight for steam propulsive machinery of 36.0 lbs. per S.H.P., which is very close to that of existing installations, after the fixed engineering weights are deducted.
The points on the protection-weight curve do not pretend to be the weights for the ships indicated, but the weights resulting from the above schedule at the designed powers of those ships.
The increasing penalty paid for speeds above twenty-eight knots is apparent.
The protection-weight curve vanishes at thirty-six knots. Between the two extremes, the Trento and the Ersatz Preussen, the protection-weight increases from 380 tons to 2,070 tons, a matter of 445 per cent, for an increase of 25 per cent in speed. A happy mean would seem to have been found in the Pensacola, but protection which does not protect is useless, and if, by establishing the speed at thirty-three knots, insufficient protection to withstand the guns of a destroyer has been obtained, the weight involved might better perhaps have been applied to speed.
The traditional American policy, first established in the frigates of the War of 1812, has been that regardless of other factors our ships should be superior in fighting strength to those of a possible enemy. This policy, as has been so clearly pointed out by Theodore Roosevelt in his Naval War of 1812, was a very large factor in the successes won in single-ship actions during that war. The advent of steam and armor did not change the situation, and with the design of our first battleships of the Indiana class, this policy was restated, and was adhered to up to the time of the laying down of the first 10,000-ton cruisers. If this policy is a sound one an inspection of Fig. 1 would seem to suggest that the speed should be somewhere near thirty knots. It is quite possible that when the effect of the Ersatz Preussen is evaluated the speed may be lower. For the purpose of this study a speed of thirty-one knots will be assumed. The power required will be, from the power curve on Fig. 5, 80,000 S.H.P.
The Problem of the Engine
The only engine designed or building today which it is possible to put into a highspeed cruiser is that of the Ersatz Preussen. It is quite necessary, therefore, before considering the possibilities of larger powers, to make a critical examination of the known facts. The official information, released in January, 1929, was meager, being confined to the statement that the total power was 50,000 and the weight per B.H.P., 8 kilos- 17.6 lbs. Since that time, however, information from various sources has reduced the matter to the status of an open secret. There is little room to doubt that this installation consists of eight 9-cylinder engines driving two shafts through gearing, runs at a piston speed of about 1,575 feet per minute, and, except for a general lightening of the frame and cylinder construction, that it is of the same type as the previous designs of highspeed engines built at Augsburg. The ratio of stroke to bore is probably about 1.25, although the poor economy indicated by the official statement suggests that it may be less. With this data it is possible to give the probable characteristics as follows:
Total power 50,000 S.H.P.
Number of cylinders 72
Power per cylinder 694
Piston speed 1,575 ft. min.
Brake M.E.P 75 lbs. per sq. in.
Mechanical efficiency 90%
M.I.P 83 lbs. per. sq. in.
Cyl. diam. 400 mm 16.00 in
Stroke 500 mm 20.00 in.
R.P.M 473
Wt. per S.H.P. main engines 17.6 lbs.
Scavenging wt. per S.H.P 4.0 lbs
Total engine wt. per B.H.P 21.6 lbs.
Wt. of gearing per S.H.P 4.0 lbs.
Wt. of engine and gears S.H.P 25.6 lbs.
Engine weights 570 tons
Wt. of shafting and propellers 100 tons
Wt. of ship service machinery 400 tons
Total engineering wts 1,070 tons
Wt. per S.H.P. total 47.9 lbs.
Wt. per S.H.P. propulsive mach 30.0 lbs.
Wt. per cu. in. cyl. vol 3.10 lbs.
The estimate of the total engineering weights based on an allowance of 400 tons, for the fixed auxiliaries under the Bureau of Engineering, is 530 tons short of that given by the German writer in Werft Reederei Hafen.
With the weights of the various items of machinery carried under Construction and Repair and Ordnance, including such weights as those of the steering gear, anchor hoists, boat hoists, ventilating motors, and turret and ammunition hoist machinery, the total might easily amount to as much as 570 tons.
A comparison of the weights of the Augsburg submarine engine with the high-speed two-cycle double-acting Berlin engine affords a check on the specific weight of 3.10 lbs. per cubic inch of cylinder volume. These two engines run at about the same piston speed, the same relative M.E.P., and represent about the same stage in the development of its type.
The weight of the Berlin engine is about 60 per cent greater than that of the submarine engine, per cubic inch cylinder volume.
Applying this ratio to later designs of four-cycle engines in which the specific weight has been reduced to 2.0 lbs. per cubic inch cylinder volume, the result for the two- cycle double-acting engine would be 3.2 lbs., which is very close to that estimated for the Ersatz Preussen.
There has been a disposition to explain the results of the Ersatz Preussen by assuming that they must be due to something new in the field of Diesel-engine design. The Engineer stated after the first information was released, “As the lastest M.A.N. engines of which we have cognizance have a weight of 55 lbs. per B.H.P., it is clear that the design of the new Admiralty type must possess some very novel features, the disclosure of which will be awaited with keen interest by the engineering world.”
As a matter of fact this decrease to about one-third of the weight of the previous type can be very simply accounted for by the combined effects of a reduction of about 25 per cent in the specific weight, an increase of about 25 per cent in the piston speed, and a reduction of about 43 per cent in the stroke.
The effect of this last factor, which follows from a consideration of the law of similitude, has not as yet found its way into any textbook on marine engines, so far as I know. The following editorial statement from Engineering of December 20, 1929, may therefore be of interest:
The laws of similarity indicate that in corresponding conditions, the engine weight per horse power should be directly proportional to the cylinder diameter, and at a meeting of the Institution of Mechanical Engineers, held in 1914, Sir Dugald Clerk exhibited a diagram in which the engine weights of internal combustion engines, plotted against the cylinder diameter showed a truly remarkable agreement with this theoretical deduction.
Direct or Geared Drive
This is a subject upon which there has been some confusion of thought, when applied to the problem of Diesel drive.
In slow and moderate-speed ships there are manifest advantages in using a number of small—and light weight—engines geared to slow propellers. As we go to higher speeds, limitations on propeller diameter force a progressive increase in the revolutions. When we reach the speeds of the 10,000-ton cruiser we are forced to use a speed of revolution for the propellers equal to that of a high-speed Diesel.
Since there can be no speed reduction there is no possible use for a reduction gear.
Somewhere between the extreme limits there is a point where any advantage due to the lower weight per B.H.P. of the small units disappears, and where the materially larger horizontal space required by the geared drive becomes prohibitive. This requirement of greater horizontal space is due, not to the gears alone, but to the fact that regardless of the size of the units the horizontal projection of the cylinders, in similar engines, is the same in all cases. It is obvious that a large number of small cylinders of the same total area, will with the necessary passages take up more space than a smaller number of large ones.
On Fig. 2 a comparison is shown between a two-shaft geared drive and a four-shaft direct drive, for an installation with the same total power as that of the Ersatz Preussen. The engine shown is the Berlin engine reduced to a stroke-bore ratio of 1.25.
The engines shown are exactly similar— that for the direct drive having been drawn by applying a factor of 1.25 to the linear dimensions of the smaller geared-drive engines.
The comparison is therefore an exact one. An estimate of the weights of the two installations is as follows:
The weight of the shafting and propellers will be materially less for the direct drive, since the total power is the same, and the propellers will have materially better conditions due to the lower tip speed for the same disc area.
It is evident that in the Ersatz Preussen there has been no saving in weight by the use of the geared drive, and that for the gain due to decreased height a very heavy price has been paid in the greater horizontal space required.
The Solution for 80,000 S.H.P.
It must be obvious that, for powers greater than 50,000 S.H.P., the direct drive with four shafts must be used. The maximum number of cylinders on one shaft, justified by precedent, is twelve. There is no valid reason why a greater number cannot be used, but in this case twelve are sufficient.
The power per cylinder will be 1,040, and with an engine similar to that of the Ersatz Preussen the cylinder diameter will be about 25 inches. Since no engine of this size, running at the high piston speed required, has been built as yet, the question might well be asked as to what extent the engine of the Ersatz Preussen can be used as a precedent
for the larger engine.
As far as the dynamic stresses are concerned there is no question but that it can be done. If similarity is adhered to the unit stresses will be precisely the same in the large engine as in the small engine.
The most troublesome factor in Diesel- engine design does not, however, follow the law of similitude. The heat stress depending upon the mean cylinder temperature, which is a function of the mean indicated pressure, varies directly as the thickness of the cylinder walls. These stresses increase directly as the cylinder diameter.
They do not increase with the piston speed, but rather tend to decrease; and since there are many engines in service with higher values of mean indicated pressure, and with cylinder diameters up to thirty-three inches, there is not the slightest room for doubt on this score.
The two sketches, Figs. 3 and 4, show a type of cruiser made possible by the use of Diesel drive. Based on the results attained in the Ersatz Preussen, the following weight schedule should be possible:
(All weights in tons)
Fixed weights
Hull 4,000
Hull fittings 500
Ship service mach. 400
Equipment 400
Aviation 150
Total fixed weights 5,450
Battery weights 1,300
Engineering weights
Engines 929
Scavenging wts 107
Shafting and propellers 160
Total propulsive mach. wts 1,196
Protection weights 2,054
Total 10,000
Fuel weight 3,500
Water weight 50
Total displacement 13,550
The general characteristics of the ship shown in these sketches are as follows: Dimensions
Length, L.W.L 600 ft.
Beam 60 ft.
Depth 40 ft.
Battery: nine 8-inch guns, eight 5-inch guns, and one quadruple torpedo tube
Machinery: four 12-cylinder, 2-cycle, double-acting engines
Cylinder diameter 24 inches
Stroke 30 inches
R.P.M 320
Brake M.E.P 80 lbs. per sq. in.
Aviation, twelve airplanes, six of which are stowed in hangar.
Boats: two 40-foot motor launches, two 35-foot motor boats, and two 20-foot motor lifeboats.
The engines shown have practically the same over-all dimensions as those of the Augsburg engine of the same stroke and bore. A slightly higher piston speed and a materially higher mean indicated pressure have been used. This is justified by the use of the loop scavenging system of Tartrais, which, with the same length of effective stroke, permits a port area about 70 per cent greater than that of the Augsburg engine.
In the engines of the Ersatz Preussen the ratio of the effective stroke to full stroke will be about 0.70, with a mean indicated pressure of about 83 lbs. per square inch. With the system used by Tartrais, under the same scavenging conditions, an effective stroke ratio of 0.78 should be possible at 1,600 feet piston speed.
Under these conditions, since the mean indicated pressure possible is a function of the effective stroke, the possible value of the brake mean effective pressure should be about 83 lbs. per square inch.
Performance
In the official statement in regard to the Ersatz Preussen, the radius at twenty knots was given as 10,000 miles.
Compared with known results of steamers under service conditions this was impressive. The Engineer stated: “Were not these figures vouched for by the German Admiralty, they would be almost incredible. We need hardly say that no vessel now afloat, whether man-of-war or liner, is able to steam anything like that distance at the speed stated.”
As a matter of fact, analysis of these figures indicates a very poor performance, and if correct, suggests that this engine may have an abnormally low ratio of effective stroke to bore, due to the low effective stroke ratio possible, combined with a low stroke-bore ratio.
The power curve on Fig. 5 shows a power of 16,000 S.H.P. for twenty knots. With a fuel capacity of 3,500 tons results would be
Endurance at 20 knots 20.83 days
10,000/480
Fuel per day 168 tons
3,500/20.83
Deduct for ship service 3 tons
Propulsive fuel per day 165 tons
Propulsive fuel per hour 15,400 lbs.
S.H.P 16,000
Fuel S.H.P.-hour 0.96 lbs.
The fuel consumption curve of Fig. 5 is based on the shop tests of an ex-German submarine engine, as given by Mr. Philip Malozzi and Mr. John F. Fox, of the New York Navy Yard, in an article published in the Journal of the American Society of Naval Engineers for February, 1929.
These results, from an engine designed about 1914, show a minimum consumption, for conditions approximating those which might be expected in service, of about 0.48 lbs. per B.H.P. per hour. Since in recently designed two-cycle, double-acting, solid-injection engines fuel consumptions as low as 0.36 lbs. per B.H.P.-hour have been reported, the results shown on Fig. 5 must be accepted as conservative, even for an engine with the low stroke-bore ratio of 1.0.
The brake mean effective pressure for the Ersatz Preussen at twenty knots will be about 33.6 lbs. per square inch. This plotted against fuel per B.H.P.-hour (scale 1, Fig. 5), will fall at the point marked E.P. This is beyond all reason, and, even with the most extreme assumptions, it is evident that this ship will have a much greater radius than that admitted in the official statement. The radius indicated for the 31-knot cruiser at twenty knots is 16,700 miles. This is materially less than would be possible for a ship with a maximum speed of twenty- eight knots, owing to the lower value of the mean effective pressure, at twenty knots, of the higher powered ships. With the improvements in economy which have recently been made, and which are still being made, it should be possible to build a naval engine which, with all propulsive auxiliaries, would show a consumption of not more than 0.40 lbs. per S.H.P.-hour. With this figure the radius at 20 knots, with 3,500 tons of fuel, would be something better than 20,000 miles.
Conclusion
In view of the manifest advantages of the Diesel engine, it is a singular fact that, outside of Germany, there is no evidence of any serious effort being made to adapt it to naval use, outside of submarines, where its use was forced from the beginning due to the fact that no steam plant could be made to stand up under the exacting conditions existing on board that type. That the early submarines were unreliable is undoubtedly true, and it is probable that the memory of this fact, without a correct appraisal of the reasons, is very largely responsible for the distrust which still exists as to the reliability of the Diesel engine, in the face of overwhelming evidence to the contrary. All of the world’s yachtsmen, and a majority of the ship owners outside of the United States, have found out that it is fully as reliable as the steam engine, and that it has intangible advantages fully as great as that of the low fuel consumption possible. These advantages in a naval vessel may be stated:
- Greatly increased aero capacity, without sacrificing gun power.
- Better protection for machinery. Each engine is an entirely separate unit which can be entirely separated from the other units. The controls can be entirely separated from the engines, as has been done in the Ham- burg-American, St. Louis and Milwaukee, and placed in a central compartment. Since the Diesel can be run under water, the flooding of a compartment need not put an engine out of action.
- Better protection against gas attack. The induction air being taken from outside, the low heat of the compartments, containing water-jacketed engines, makes it possible to seal the engine compartments long enough to run out of any gas area.
- Full speed available at all times.
- Greater safety to personnel, due to absence of high-pressure boilers and pipes.
- Greatly increased crew space due to the absence of the large boiler uptakes passing through the living spaces, and greatly increased comfort due to the absence of heat from boilers and steam pipes.
- Economy of space. This is due to the fact the over-all length will be somewhat less and the athwartship dimensions will be very materially less. In the plan shown in Figs. 3 and 4 the total length of the machinery space is only 156 feet, and between the forward engines there is a space, between the fore-and-aft bulkheads, 90 by 26 feet.
- The practically complete elimination of dependence on fuel ships, due to a radius at twenty knots equal to about six crossings of the Atlantic Ocean.