Mark 14 torpedo’s side view and interior mechanisms, published in “Torpedoes Mark 14 and 23 Types, OP 635”, March 24, 1945
A torpedo may take a long time before it settles on its final course. If the torpedo direction is still changing when the torpedo arms, it may set off the magnetic influence exploder.
by Frederick J Milford
Naval rearmament, which began in the mid-1930s, and WW II had dramatic impact on US torpedo programs. Three of the most significant changes were the enormously increased requirement for torpedoes, the urgent need for new torpedo types and the first use of US torpedoes against enemy vessels. The increased requirement was satisfied by expanding government facilities, the Newport Torpedo Station (NTS-Newport) was enlarged, the Alexandria Torpedo Station was reopened 1 and Keyport Torpedo Station began assembling torpedoes, and by initiating civilian production. Total production between 1939 and 1945, almost 60,000 torpedoes, was about equally divided between the torpedo stations and contractors. Mk.14 torpedoes were, however, in such short supply in 1942 that some fleet boats loaded out with Mk.10 torpedoes or even Mk.15s in the after tubes 2. New types of torpedoes are discussed in Part Three of this series. Firing warshots was an almost totally new experience for the US Navy. It seems probable that the number of warshots fired against enemy vessels in December 1941 was larger than the total number of warshot torpedoes fired for any purpose 3 in the entire past history of the US Navy. Perhaps not surprisingly, this intensive use of torpedoes revealed shortcomings that had been previously obscured, especially in the new service torpedoes and particularly in the Mk.14.
The trio of new service torpedoes, Mk.13, Mk.14 and Mk.15, which represented the bulk of the US Navy torpedo development in the 1930’s were on the one hand excellent weapons and had long service lives, Mk.13 remained in service until 1950, the Mk.14 was a valuable service weapon until 1980 and Mk.15 served as long as twenty-one inch torpedoes remained on destroyers. On the other hand they all had significant problems that were only fixed after wartime use began. The Mk.14, which was the principal submarine weapon, was plagued with defects that vitiated its use as a weapon until mid-1943. The conflict between the shore establishment and the operating forces over these problems was a very significant and much discussed factor in US submarine operations during WW II.
THE GREAT TORPEDO SCANDAL
The Great Torpedo Scandal 4 emerged and peaked between December 1941 and August 1943, but some of its roots went back twenty-five years. It involved primarily the Mk.14 5 and three distinct problems, depth control, the magnetic influence exploder 6 and the contact exploder, whose effects collectively eroded the performance of the torpedoes. The scandal was not that there were problems in what was then a relatively new weapon, but rather the refusal by the ordnance establishment to verify the problems quickly and make appropriate alterations. The fact that after twenty-five years of service the Mk.10 had newly discovered depth control problems adds weight to the characterization of the collection of problems and responses as a scandal. These comments should, however, be mitigated a little by the fact that each of the Mk.14 problems obscured the next. Although BuOrd did not identify the final problem, contact exploder malfunction when a torpedo running at high speed struck the target at ninety degrees, their response, once the difficulty had been identified, was notably prompt. In spite of the promptness of BuOrd’s response, by the time it reached Pearl Harbor a number of relatively simple solutions to the problem had been proposed, and modifications had already been designed and implemented. This was, however, almost two years after the United States entered WW II.
Torpedo Depth Control
The first of the US torpedo problems was deep running, which was a frequent torpedo problem in various navies beginning at least as early as WW I. The problem was not, however, always due to the same sort of defect 7. There are at least four distinct kinds of problems that impact depth control:
1) Differences between calibration shots and service/warshots
a) Torpedo weight or balance changed in converting to warshots, for example, warheads that were heavier than calibration heads.
b) Calibration firings failed to simulate service launch conditions, for example, calibration firings from barges or surface vessels rather than submerged torpedo tubes, and/or calibration shot launch speeds, i.e., the speed at which the torpedo leaves the tube, and accelerations during launch different from service conditions.
2) Design or manufacturing defects causing changes in calibration after proofing or effectively causing calibration to change with time or environment, for example, sensing water pressure where flow corrections were large, or depth spring fatigue, or leaky castings etc.
3) Erroneous calibration: failure to check against an absolute standard, for example, total reliance on hydrostatic depth measurement and failure to use nets, soft targets or other sensing systems to establish true depth.
4) Inadequate understanding of the technology involved, for example, failure to recognize the importance of hydrodynamic flow in sensing the pressure at the skin of a fast torpedo; lack of understanding of the feedback loop and depth control dynamics 8.
Amazingly, US torpedoes, especially the Mk.14, demonstrated that most of these possibilities could, in fact, occur.
Depth control problems with US torpedoes were suspected by the Newport Torpedo Station (NTS-Newport) and BuOrd even before the United States entered WW II. On 5 January 1942 BuOrd, based on earlier (1941) testing, advised that the Mk.10 torpedo, which had entered service in 1915 and was still used in S-class submarines, ran four feet deeper than set 9. NTS-Newport tests on the Mk.14 torpedo in October 1941 had been interpreted as indicated that it too ran four feet deeper than set, but this was not reported to the submarine commands at that time. War patrol experience led to fleet suspicions that the torpedoes ran deep and these thoughts were communicated to BuOrd. In response to a direct order from the Chief of the Bureau of Ordnance, additional NTS- Newport tests in February-March 1942 “confirmed” the four-foot error for the Mk.14. RAdm William H. Blandy, Chief of BuOrd, notified RAdm Thomas Withers, Jr., ComSubPac, of the problem in a letter dated 30 March 1942, but general notification to the submarine forces was not made until BuOrd issued BuOrd Circular Letter T-174 dated 29 April 1942. The language in correspondence between Withers and Blandy indicate that Newport and BuOrd believed that the four-foot error in Mk.14 depth was due to calibrating torpedoes with test heads that were lighter than the warhead. This would cause torpedoes with warheads to run deep both because of increased weight and a nose heavy trim. The Mk.14 depth control problem was, however, much more severe than the four feet acknowledged by NTS-Newport.
In a mood of desperation, the operating forces made their own running depth determinations, using fishnets for depth measurement, at Frenchman’s Bay in Australia on 20 June 1942. These measurements indicated that the depth errors were probably more like eleven feet 10. BuOrd and NTS-Newport criticized the methodology and were reluctant to accept the results of the
Frenchman’s Bay firings and it was not until August of 1942, after intervention by the CNO, Admiral Ernest J. King, that they re-investigated and agreed that there was a ten-foot depth error in the Mk.14 system. Interim instructions for fixing the problem were issued very quickly and kits to effect an official alteration were distributed in late 1942. As near as we have been able to determine, there were two independent problems: Trim change due to warheads heavier than calibration heads and sensing the water pressure at a point where the velocity head was significant and consequently the measured pressure was low. The fix for the latter moved the pressure sensing port to the interior of the free-flooding midbody where the pressure was close to the true hydrostatic pressure and so reflected the true depth. The modified torpedoes were identified by the suffix A added to the Mod. with the most famous being Mk.14 Mod. 3A
Since the hydrodynamic problem has seldom been explained in readily accessible documents, we give a brief summary here. The pressure along the length of a torpedo varies because the velocity of the water relative to the surface varies. The pressure at the nose is higher than the hydrostatic pressure, which is proportional to depth, by an amount proportional to the square of the torpedoes speed. This corresponds to a depth of 39 feet of seawater for a torpedo moving at 30 knots or 88 ft for a 45 knot speed. As the measuring point is moved back along the skin of the torpedo the pressure decreases rapidly and becomes substantially less than the hydrostatic pressure. The pressure subsequently rises but remains slightly less than the hydrostatic pressure along most of the cylindrical section. Finally along the conical afterbody the pressure again drops and then rises though, since the actual flow is not streamline, not to the values found at the nose. The critical point is that the pressure at the skin of a torpedo is generally different from the hydrostatic pressure corresponding to the torpedo’s depth. The deviation is substantial in the nose and tail cone regions. A depth error due to the measurement of the wrong pressure would, of course, be detected in any calibration process that used an absolute depth measurement for reference. Unfortunately the Torpedo Station used a depth and roll recorder which determined depth by measuring the water pressure and was thus subject to the same kind of error as the depth gear. Furthermore, the depth and roll recorder was placed in the test head at a point where the hydrodynamic pressure was less than the hydrostatic pressure by almost the same amount as at the location, in the afterbody, of the sensing port for the depth gear. Thus both the recorder and the depth gear sensed essentially the same pressure, though not the hydrostatic pressure, and the torpedo appeared to be running at the set depth. The depth engine, however, responded to the lower pressure by adjusting the horizontal rudders to correct this “error” and the torpedo ran deep. The hydrodynamic theory needed to understand this problem was readily available in the 1930s but most design engineers were quite probably not acquainted with it. In consequence, it was assumed that since the depth recorder showed the correct depth, the torpedo was running at the correct depth. There are other insidious aspects to this problem. One of these is that a depth recorder checked against depth by static immersion in water to various depths or in a pressurized tank of water reads correctly since the error described above is due to hydrodynamic flow. Further the error is proportional to the square of the torpedo speed and is thus almost twice as important for a 46 knot torpedo as it is for a 33 knot torpedo. None of these comments, however, justify or excuse the failure to use an absolute standard to verify the results obtained with the depth and roll recorder or the obdurate resistance to complaints from the operating forces.
The operational aspects of the depth control problem have been recounted many times 11. The Mk.10 problem, which was probably dominated by the error caused by the change from exercise heads to warheads, was handled by simply setting the torpedo to run at a shallower depth and this procedure was implemented in January 1942, over twenty-five years after the weapon entered service. The Mk.14 problem required both a calibration modification and a modification to sense water pressure in the midships section and the latter was implemented beginning in the last half of 1943.
The Magnetic Influence Exploder
The second problem with the Mk.14 torpedo was the erratic performance of the magnetic influence feature of the Mk.6 exploder. Magnetic influence exploders had great appeal as proximity fuses for torpedoes offering the possibility of detonating the warheads under the vulnerable bottoms of warships. This potential advantage led most of the major navies to attempt to develop such exploders and generally these first attempts were not successful in service use.
The basic idea of a magnetic influence exploder is to sense either the field due to permanent magnetization of a ships hull or the perturbation of the Earth’s magnetic field caused by the large quantity of relatively high permeability ferrous metal in the ships structure. This is a sound and workable idea, but early simple attempts did not take adequate account of the nature of the perturbation. The Mk.6 device in particular relied on the variation of the horizontal component of the magnetic field as the torpedo approached the target. This field variation induced a voltage in a sensing coil. The voltage triggered a thyratron which discharged a capacitor through a solenoid. The solenoid, in turn, operated a lever that displaced the inertia ring thus triggering the mechanical exploder. This complex arrangement was presumably designed so that an exploder, Mk.5, without the magnetic influence portion, but otherwise identical to the Mk.6 exploder could be produced and issued to the fleet in peacetime. Security was apparently the overall motivation for this convoluted approach.
The perturbation of the Earth’s field by a ship naturally depends on the inclination of the Earth’s field to the horizontal. This inclination varies from zero at the magnetic equator to ninety degrees at the magnetic poles. At NTS Newport it is about sixty degrees. Regardless of the inclination of the Earth’s field, a ship, because of the ferrous metal in its structure, causes both horizontal and vertical perturbations of the Earth’s field which vary with distance and direction from the ship. The closer the Earth’s field is to vertical the greater the rate of change of the horizontal perturbation field with distance and the closer to a point directly below the keel the maximum rate of change occurs. Thus a device that senses the rate of change of the horizontal component of the perturbed field works best where the Earth’s magnetic field has a large vertical component. Unfortunately, a device that works well at high magnetic latitudes may not work at all well where the Earth’s field is nearly horizontal. Thus, the performance of a simple magnetic influence exploder is significantly dependent on the latitude at which it is operated.
Exactly this problem affected the magnetic exploders developed by the Royal Navy, the German navy and the US Navy. The Royal Navy quickly abandoned magnetic influence devices and relied on contact exploders. The German navy provided a sensitivity adjustment that would, in principle, compensate for changes in latitude. This was unsatisfactory and it too was abandoned fairly quickly 12. The BuOrd/Naval Torpedo Station Newport response was first denial that there was a problem, then a complicated set of instructions for setting the exploders for different latitudes.
The magnetic influence exploder was unquestionably responsible for sinking some, perhaps even a large fraction, of the 1.4 million gross registry tons of Japanese merchant ships sunk by submarines between December 1941 and August 1943. Reports from submarine commanding officers of apparent magnetic influence exploder failure, mainly duds and prematures, finally led to CinCPac ordering the disabling of the magnetic influence feature on 24 June 1943. ComSubSoWesPac reluctantly followed suit in December 1943 13. CinCPac’s order was issued eighteen months after Jacobs, on Sargo’s first war patrol, ordered the deactivation of the magnetic influence portion of the Mk.6 exploders in his torpedoes and incidentally got into considerable difficulty for doing so. Magnetic influence exploders were not used by US Navy submarines through the balance of WW II.
The Impact Exploder
Once the depth problem had been fixed and the magnetic influence feature of the Mk.6 exploder deactivated, it came the turn of the impact exploder to demonstrate its merit. Unfortunately the initial result was a plethora of duds, solid hits on targets without warhead detonations 14. This problem was suspected earlier, but it was not until the other two problems had been eliminated that there was unequivocal evidence of a problem with the impact exploder. This difficulty was a further frustration for the operating forces, but fortunately it was quickly diagnosed. The key to the problem was again the increased speed of Mk.14 15. The impact portion of the Mk.6 exploder was exactly the same as that which had been used in the Mk.4 and Mk.5 exploders. The Mk.4 worked entirely satisfactorily in the 33.5 knot Mk.13 torpedo. What was overlooked was that in going from 33.5 knots to 46.3 knots the inertial forces involved in striking the target at normal incidence were almost doubled. These greatly increased inertial forces were sufficient to bend the vertical pins that guided the firing pin block. The displacement was sometimes enough to cause the firing pins to miss the percussion caps, resulting in a dud. In cases of oblique hits, the forces were smaller and the impact exploder more often operated properly. Several war patrols, especially those cited above, convinced ComSubPac, VAdm Charles Lockwood, that there was a problem and he again resorted to experiment. Firings at a cliff in Hawaii demonstrated that some torpedoes did not detonate when they hit the cliff. A rather risky disassembly of a dud revealed the distortion of the guide pins. It was a simple solution to make aluminum alloy (rather than steel) firing pin blocks and lighten them as much as possible thus reducing the inertial forces to a level that did not distort the guide pins. Another solution was to use an electrical detonator and a ball switch to fire the warhead. This too was relatively easy to implement and soon became standard.
Once these and other less significant problems were solved, the Mk.14 torpedo became a reliable and important weapon. After WW II, it was modified to accommodate electrical fire control settings, gyro angle, depth and speed, and as Mk.14 Mod.5 remained in service until 1980.
HOW AND WHY
It is worth asking how these three problems might have come about and presented such a refractory situation early in WW II. It is easy to identify several contributing factors, but it is unlikely that any one of them alone was the deciding factor. One of the first factors was the economy. These torpedoes were developed during the great depression, the total US Navy budget from 1923 through 1934 averaged less than 350 million dollars per year and total personnel stood at about 110,000. In that environment a torpedo was valued at around $10,000 (about the same as a fighter aircraft airframe complete except for engine) and destroying one in testing was a risk that only the fearless were willing to run. The result was that testing and proofing were done in such a way as to avoid risk of damage either to expensive torpedoes or scarce targets. As is often the case, constrained testing failed to reveal certain critical problems. It is, however, difficult not to believe that deep running, in particular, should have been discovered. There were well documented reports of German and British problems during WW I. It appears also that impact exploders were not tested in high speed torpedoes or at least not tested in impacts of well simulated warheads with hard targets. Such tests were undoubtedly omitted in an effort to avoid destroying useful materiel, exploders in particular, and perhaps further justified by the fact that the exploder performed satisfactorily in lower speed tests and by its primary role as a back up to the magnetic influence exploder. Thus we conclude that with respect to these two problems, depth control and the impact exploder, the poor state of navy finances and the concomitant lack of realistic testing probably played a significant role.
Another aspect of the situation was the almost total isolation of NTS Newport from the larger US technical and engineering community especially after 1923 when the station secured a monopoly on torpedo development and production. Political and labor interests in keeping jobs in New England probably encouraged the isolation. The net result seems to have been a lack of expansion of the scientific basis for torpedo technology at Newport at a time when dramatic changes in engineering were taking place elsewhere. No one was thinking about torpedoes from different perspectives and asking hard questions about design details. The isolation was exacerbated, especially in the case of the Mark 6 exploder, by draconian security, which in some cases even excluded the operating forces from full knowledge of the weapons they were expected to use. In this isolated environment, NTS-Newport developed an arrogant `We are the torpedo experts.’ attitude and when problems began to arise, the response was denial–`there is nothing wrong with the torpedoes’- -with the result that problems were identified and fixed slowly.
Perhaps not surprisingly a very strong polarization developed between the operating forces and the torpedo shore establishment. The operating forces resented their exclusion from the torpedo development cycle and flaunted their successes in proving that there were problems with the Mk.14 torpedo. These strongly expressed opinions of the men of the operating forces did not tend to improve relations with NTS-Newport. The operating forces also tended to exaggerate their contributions to the solution of the problems and deprecate those of NTS-Newport. A distinguished and truly great submariner recently wrote: _So by the beginning of September 1943, the operating submariners had detected and solved three serious defects in the Mark XIV torpedo: its faulty depth setting, skittish magnetic exploder and sluggish firing pin. All three problems had been solved by the operating forces in their tenders and bases, without help from Newport or Washington. 16 This is certainly an overstatement, but what is most significant is that though written over fifty years after the events, it still reflects the intense polarization that existed between the operating forces and the torpedo shore establishment.
This spectrum of problems was not unique to the US torpedo establishment. Almost the same set, defective depth control, unsatisfactory and untested magnetic exploder and a contact exploder that did not work at certain striking angles, occurred in the German navy and many of the responses of the shore establishment to the problems were also the same. The situation is discussed in considerable detail by Doenitz in his memoirs 17. The German navy’s problems were closed out, however, with four senior officers being tried by court martial, on the orders of Grand Admiral Eric Raeder, found guilty and punished.
Lest there be any implication that the entire US Navy or even all of BuOrd was functioning in isolation, we note that at about the same time early experiments with what became radar were being conducted at the Naval Research Laboratory (only about 350 miles Southwest of Newport). In 1937 complete disclosure of the state of radar development was made to the Army Signal Corps and Bell Telephone Laboratories. Radio Corporation of America was brought into the fold in 1938. 18 The contrast of this approach to the Newport approach is nothing if not striking. BuOrd itself in the development of range keepers for surface fire control, in a comparably secret endeavor roughly contemporaneous with the Mk.14 development, co-opted Ford Instrument, ARMA and Sperry to assist with the development. A later dramatically contrasting development program was the development of the Mk.24 Mine (Torpedo) between December 1941 and May 1943, which is discussed in a subsequent part of this series.
This takes the story of U.S.Navy torpedoes through beginning of WW II. As the United States became involved in the war, it became apparent that new kinds of torpedoes would be useful and a multitude of programs to develop improved weapons for submarines, surface vessels and aircraft were initiated. The idea that torpedoes could be significant ASW weapons also evolved and was elaborated with considerable success. The wartime developments and the post war development of US Navy torpedoes are discussed in the third part of this series.
1 The Newport monopoly on the torpedo business had a significant effect on the development of torpedoes. The extent of the monopoly and efforts to preserve it are illustrated by opposition to the reopening of Alexandria, which was accomplished in the face of demands from New England politicians and labor leaders that Newport be expanded. Resuming torpedo work at Alexandria expeditiously was possible only because when it was closed in 1923 it had been incorporated into the Washington Navy Yard. Consequently, the torpedo station could be reopened without an Act of Congress.
2 This was mentioned by Adm. B.A. Clarey in a recent interview with John DeVirgilio and confirmed by RAdm M.H. Rindskopf who also supplied key parts of the following material. Mk.15 torpedoes were too long to be loaded through hatches or stowed in the torpedo rooms. They were also too long for either the forward or longer aft torpedo tubes. They were modified, probably by using shorter warheads, and loaded into the aft tubes through the muzzle doors. USS Drum, SS-228, sailed so loaded on her second war patrol from Pearl Harbor in July 1942. All four Mk.15s were fired.
3 This, of course, means self-propelled torpedoes and excludes spar and towed devices. Apparently, only eleven torpedoes fired by US forces against enemy vessels prior to WW II (AL boats against U-boats). The number of warheads used in training and test and evaluation was very small. US submarines made 54 war patrols in December 1941 and fired 66 torpedoes at enemy targets, quite possibly more warheads than had been fired in the entire previous history of the US Navy.
4 At least three MA theses have been written about the problems of the Mk.14 torpedo (Ingram (1978), Shireman (1991) and Hoerl (1991)); the problem was noted by Morison and is discussed at length in Theodore Roscoe, , _United States Submarine Operations in World War II_, Annapolis: Naval Institute Press, 1949; Clay Blair, Jr., “Silent Victory: The U.S.Submarine War Against Japan”, Philadelphia and New York: J.B.Lippincott, 1975; and Edwyn Gray, “The Devil’s Device: Robert Whitehead and the History of the Torpedo” (Revised Edition), Annapolis, USNI Press, 1991.. David E. Cohen has written a paper on the subject, “The Mk.XIV Torpedo: Lessons for Today”, Naval History, Vol.6, No.4, Winter 1992, pp.34-36″
5 Criticism of the destroyer launched Mk.15 is almost nonexistent. This is strange because the principal differences between the Mk.14 and the Mk.15 were in the size of the warhead, the fuel load, three speed vice two speed and slightly slower high speed, 45.0 k vice 46.3 k. One might speculate that it is even more difficult to distinguish misses from duds in a high-speed destroyer attack than it is in a more measured submarine attack. The Mk.15 did, in fact suffer from the same defects and they were rectified in essentially the same way that those of the Mk.14 were. The Mk.13 was a slower speed torpedo so it did not have the contact exploder problem and it used the Mk.4 exploder which did not have the magnetic influence feature.
6 Properly, the exploder is the entire Mk.6 assembly. It has an influence feature and a contact feature. This leads to awkward verbiage so we refer to the magnetic influence exploder and the contact exploder. Both are parts of the Exploder Mk.6, which weighs approximately 90 lbs, and some elements of the exploder function in both modes. The exploder also contains important safety features.
7 Some indication of the bewildering set of problems experienced by other navies can be found in Cdr. Richard Compton-Hall, RN (Ret) “Submarines and the War at Sea”, 1914-1918″, London: Macmillan, 1992; Karl Doenitz, “Memoirs: Ten Years and Twenty Days”, Annapolis: U.S.Naval Institute Press, 1990. Cajus Bekker, (pseudonym for H.D. Berenbrok). “Hitler’s Naval War” Garden City, NY: Doubleday, 1974.
8 The Summary Technical Report of Division 6 of NDRC, _Torpedo Studies_ Vol.21, Washington: NDRC, 1946, p.15, contains the following revealing comment: “The principal result of the study of Depth-keeping is the development of a theory … there is no longer any excuse for the laborious production of depth mechanism that cannot be expected to operate at all.”
9 Roscoe p.253
10 More detail can be found in any of the references cited above. Blair discusses the situation on pp 275 ff. It is not clear whether or not the eleven-foot error included the error due to changing from exercise heads to warheads. It is, however, interesting that BuOrd/NTS-Newport criticized the Frenchman’s Bay experiments on the basis of “improper torpedo trim conditions” (Quoted in Blair p.276).
11 Roscoe p.253; Morison Vol.IV p.221 in particular; Blair pp.169-70, 198; John David Hoerl, _Torpedoes and the Gun Club_, unpublished MA Thesis, VPI and State University, 1991, pp. 9-15
12 Successful magnetic exploders have, of course, subsequently been developed by many organizations.
13 ComSubSoWesPac (Christie) issued the deactivation order in response to an order he had received from the new Commander, Seventh Fleet (Kincaid). Blair “Silent Victory”, p.504. Christie had been heavily involved in the development of the Mk.6 exploder at Newport and was reluctant to see it abandoned.
14 Two of the best documented patrols that suffered duds were Wahoo 5 (April 1943) and Tinosa 2 (July 1943). The first of these is reported in O’Kane “Wahoo” and the second in Shireman “The Sixteenth Torpedo” unpublished MA thesis, U. of Wisconsin, 1991.
15 The literature on the Mk.13, Mk.14 and Mk.15 torpedoes focuses strongly on the Mk.14 and says almost nothing about either the Mk.13 or the Mk.15. This is understandable in the case of the Mk.13 since it was a slower torpedo and consequently had a smaller depth error and no major problem with the contact exploder. In the case of the destroyer launched Mk.15, which was a few feet longer than the Mk.14 and carried a larger warhead, but otherwise nearly identical to the Mk.14, I have found no references to unequivocal torpedo failures. This may be because during a destroyer torpedo attack things are too hectic to permit a careful evaluation of torpedo performance.
16 James F. Calvert _Silent Running: My Years on a World War II Attack Submarine_, New York: John Wiley, 1995 pp.96-97.
17 Karl Doenitz “Memoirs: Ten Years and Twenty Days”, Annapolis: USN Press, 1990. The bulk of the discussion of torpedo failures is contained in Chapter 7 and Appendix 3.
18 L.S.Howeth “History of Communications-Electronics in the United States Navy”, Washington: GPO, 1963 Chapter XXXVIII, and chronology pp. 540-41.