Chapter 18: Diving and Diving Time
1. Originally the normal way of diving a submersible was by adjusting the amount of water ballast carried. Towards the end of the nineteenth century the most promising way was to get the boat to an 'awash' condition with a small amount of positive buoyancy and then drive it underwater. The main controversy was whether the boat should be dived horizontally or at an angle and this also applied to changing depth when submerged.
2. In his first boat in 1885 Nordenfelt submerged the vessel by using two vertical propellers placed in sponsons one on each side of the boat and driven by a single 6-hp double cylinder engine. In the bow on either side were balanced rudders (hydroplanes) on the same shaft always maintained in a horizontal position. In his second boat of 160 tons surface displacement, the two 'sinking screws' were each driven by a 6hp engine and were placed in the fore and aft line. 'With the vessel in the awash condition the vertical propellers were started and the boat pushed down so that it remained perfectly horizontal and was invariably kept so when moving under water by means of the bow rudders operated by a plumb weight. The mere arrest of the two vertical propellers sufficed to bring the boat to the surface, as it had a reserve of buoyancy'.
3. Three Nordenfelt boats were built at the Barrow Shipbuilding Company (now Vickers) during the period 1885-7, the first of which was the second boat mentioned In Paragraph 2, and they would all have been fitted with 'sinking screws'. The third boat went on trials on 7 May 1887 but it was not a success mainly because '(a) It had extremely small longitudinal stability when submerged and (b) a mistake had been made in the calculations which made the boat trim excessively by the stern. Ballast had to be added to correct the trim and the main ballast tanks could not in consequence be completely filled. The ballast tanks were not subdivided, the water surged about in them as the boat trimmed by the bow and stem and the surging of the water accentuated the trim. The slightest movement on board caused trim'.
4. It is seen that Nordenfelt's principle was to keep the boat perfectly horizontal when diving and to keep the forward hydroplanes always horizontal when running submerged so that if the boat developed a trim by the bows the hydroplanes were in effect put to rise and vice versa. It was the fact that the boats were longitudinally unstable when submerged that made them a failure and not Nordenfelt's principle of diving horizontally nor of using 'sinking screws'. It must however be recorded that the first two of the Barrow boats commissioned by Turkey and subsequently named ABDUL HAMID and ABDUL MEDJID. They left Barrow in sections for Turkey in April 1866 and the third was lost off the Home Reef, Jutland on 18 September 1888 on passage to Russia.
5. Mr Holland in the USA on the other hand believed in the principle of diving his boat at an angle-he 'steered his boat down and up an incline by the action of horizontal rudders placed in the stern'. Fyfe-states regarding the acceptance of the Holland design by the American Government in 1895 that:
The Holland design was accepted because it embodied the ideas of a fixed centre of gravity, of an exact compensation for expended weights, of a low longitudinal metacentric height and of quick diving and rising by the effort of the propeller pushing the vessel against the resistance of her midship section only down and up inclines, the angles of which were to be determined by horizontal rudder action'.
The statement about the low longitudinal metacentric height is extraordinary since it was well known at the time that the lack of good longitudinal metacentric height was one if not the main defect of previous submarine boats.
6. The controversy at this time was therefore on the question of whether a submarine should dive on an even keel or at an angle. Simon Lake-who submitted a design in the US Government competition in 1893-was a believer in even-keel diving. It is thought that when he failed to be awarded the contract he continued to press the US Government and US Navy for recognition of his method as being preferable to the Holland angled dive. In his book, Lake states that 'in 1902 a US Navy representative testified before the Committee of Naval Affairs to the effect that he did not think that Lake's hydroplanes would work and strongly contended that submergence by inclining the vessel itself was the proper method'. But it must be said that even at that time Lake had bow and stern hydroplanes and took his boat down horizontally using the hydroplanes. He considered it wrong to put a trim on a submarine purposely whether to dive or change depth and in those early days when submarines were very lively to changes of trim he was probably very right. The question of speed of diving was not then one of priority. Furthermore with both bow and stern hydroplanes he had the means of angled diving if he had wished with better control than Holland .
7. An Interesting point regarding depth control in the Holland boats, and it is understood to have been in the specifications for the RN Holland boats, is that an 'automatic control device was fitted to the after hydroplanes'. This device is stated to have been 'operated by the pressure of the water which was expected to automatically control the depth of submergence, it being only necessary theoretically to move a control lever to a point on a dial corresponding to the desired or premeditated depth of submergence, and the horizontal diving rudder would then be automatically manipulated to incline the bow of the boat down until the desired depth was reached and then to be manipulated (by the pressure of depth of water) to throw the bow up or down to maintain that depth'. Whether this device was ever fitted and worked satisfactorily is not known; it was certainly not fitted in the RN Holland boats on completion.
8. Regarding RN submarines the Holland boats were fitted with after hydroplanes only, and on the surface normally trimmed about 5° by the stern. When preparing to dive the vessel was stopped and the main tank flooded to bring the submarine to the 'awash' condition with about 3in to 5in of water over the hull and only the conning tower above water. Auxiliary ballast water usually had to be admitted as well to obtain this condition. The vessel then had about 300lb reserve of buoyancy and could be navigated on the main motor. On the order to dive the hydroplanes were put to about 8° dive and the boat taken down on the planes when sufficient speed, between 5 and 6 knots, had been obtained. The angle of inclination when diving was 'about 10° and after the required depth had been obtained this angle became so small as to be inappreciable'. With the 300lb of buoyancy when submerged the speed of 5 to 6 knots had to be maintained to keep effective depth control. It is stated that these boats could dive to 28ft in 8 seconds. This is extremely quick and could not have been achieved from stopped in the awash condition, but probably is measured from the moment of having attained sufficient speed to overcome the whole of the reserve of buoyancy. See para 14.
9. The same method of preparing to dive and diving was used in the A Class, B Class and C Class. In these larger boats the reserve of buoyancy in the awash condition was increased to 500lb in the A Class and 0. 5 tons in the B Class and C Class.
10. In the early spindle hulled boats diving from the full buoyancy condition was a lengthy process to be carried out with care. The main tank kingstons would have to be opened and closed a number of times or perhaps just partially opened. Too much water entering too quickly could be dangerous. This was shown in A4 whilst doing some sound-signalling experiments when the boat was being trimmed to the awash condition. 'The Kingston was opened and water taken in too quickly which gave too great a downward momentum in spite of the reserve of buoyancy. This was a snag in these vessels and the French had lost vessels in this way. One ventilator had been left open for signalling during experiments and this was dipped under-water. A considerable quantity of water came inboard, all buoyancy was destroyed and the boat reached 90ft before downward momentum could be checked. The supreme coolness of the crew brought the boat to the surface in spite of water in the batteries which put out the lights, short circuits everywhere and metal fittings electrified'. Furthermore, in certain vessels, there was a restriction on the use of main ballast water and some main tanks could not be filled completely.
This does not mean that the importance of diving as quickly as possible consistent with safety was not realised. A change from A1 to A2 in dividing up the forward main ballast tank to improve flooding time is a case in point.
11. Some interesting points on the hazards of diving and running on the surface and submerged are mentioned in Captain Bacon's paper before the Institution of Naval Architects in 1905, in a discussion on the loss of A8. Although the exact cause of the accident is not known, the following is taken from the records on what happened -'The boat was on the surface and gradually settling down by the bows through water entering at a leaking rivet into a forward tank. She suddenly plunged when going at 10 knots. Evidence showed that the Captain gave orders for more up-helm to be put on over and over again. It is possible the coxswain thinking the planes may have stuck worked them to hard dive with the intention of bringing them to hard up again to see if all was clear, a practice often done to test a steering wheel. The bow dipped and forced the boat under the water with the conning tower hatch open. Water entered and flooded the boat and closed the hatch. Submarine sank in 8 fathoms but was afterwards raised'.
12. In the discussion on the INA paper it appears that A8 was at only about one-third of her normal buoyancy, i. e. at about 6 tons, and steaming at a considerable speed (10 knots). The A Class when stopped trimmed 4° by the stern.
When running on the surface the action of the water on the fore end of the hull made the boat sink vertically and bring it down by the bow. At about 10 knots the trim decreased to 2° by the stern and the MCT1 in this condition with 6 tons reserve of buoyancy was 33ft tons. The question was how did the boat lose this 2° of trim and so get into a position to plunge? From calculations, and it is known that model experiments were also conducted at the Admiralty Experiment Works Haslar, it appeared impossible for this to happen with 6 tons reserve of buoyancy, but the pressure on the bows would cause the boat to dive if the reserve of buoyancy fell to 3 tons when the MCT1 was only 11ft tons. It was therefore concluded that water had somehow leaked into some internal tank probably forward.
13. Be that as it may some further interesting information came out in discussion:
- The danger in running on the surface at less than normal reserve of buoyancy was not realised before this accident since, as Captain Bacon stated,
With any tendency of the boat to dive the hatch merely had to be shut to render the boat absolutely safe. If the boat had been at full buoyancy when steaming fast the water would never have approached the hatch.
In future, boats will only be run with their proper amount of buoyancy.
- There was a definite rule never to open a Kingston unless the conning tower hatch was shut.
- There was no tendency for an abnormal dive to happen when running submerged 'and up to the highest inclinations we ever go, about 10° by the bow'.
- When submerged the boats always went along with a certain inclination bows down to overcome the tendency to rise due to the reserve of buoyancy. There is a suggestion that this inclination down by the bows might have been as much as 5°.
- On the question of movement of the crew in the boat upsetting trim when submerged and putting the vessel into a dive, and of course the longitudinal stability was small. Captain Bacon stated
There is no danger in moderate movement. But the time of duration of a dive in these boats is at the outside 3 hours and there is no more difficulty in sitting still for 3 hours in a submarine boat than there is in sitting still for 3 hours in a railway train.
14. Going back to the time taken to dive Sueter3 gives some interesting figures for the USN Adder Class boats which were the same as the RN Holland boats. He states that 'the actual time taken to dive with main ballast blown and other tanks full is about 4 minutes'. The key statement here is underlined because the boat is not then in the normal surface condition. He goes on to say 'to form a correct estimate of rapidity of any submarine diving it must be remembered that the time varies with the amount of water in tanks. 29 minutes is a maximum with everything empty in the Holland type. The following three boats from the light condition took Adder 29 minutes, Moccasin 25 minutes and Pike 15 minutes. When trimmed for diving, to dip the boat under the surface takes only a few seconds'. The fact is that the RN Holland boats on the surface when underway were never in diving trim.
15. An interesting remark is made by Captain Bacon about the American method of diving their boats. He states that they 'actually have a little tank representing exactly the amount of buoyancy they require when diving and actually let the boat begin to sink and then blow the water out of this tank and then buoyancy is exactly right-Porpoise sank in 120ft of water-they were not quick enough on this occasion'. This sounds very much as if the Americans were using the buoyancy tank to achieve quicker diving.
A statement has been seen that the capacity of the buoyancy tank was equal to that of the conning tower which means the positive reserve of buoyancy when submerged. It is perhaps relevant that from A3 onwards in the spindle hull boats the buoyancy tank was not fitted with a Kingston possibly to prevent it being used for this rather rash operation.
16. With the growth In size of submarines it must have become obvious that the method of diving on after hydroplanes alone was obsolete and preparing for diving by coming to the awash condition and diving with some reserve of buoyancy was unnecessarily complicated. It may be also that Lake's idea of forward and after hydroplanes was by now considered to have attractions.
17. During the period 1905 to 1907 amidships and/or bow hydroplanes, in addition to the after hydroplanes, were tried and experience showed the combination of bow and stern hydroplanes was the best for 'keeping the boat horizontal when changing depth. They also added to the navigational qualities submerged'. Bow hydroplanes were fitted in the C Class during building and probably in the A Class and B Class retrospectively. The fitting of both bow and stern hydroplanes now became standard practice.
18. From the D Class onwards the submarines had no reserve of buoyancy when dived. As far as is known the diving procedure adopted in these boats is that which is still used in the latest conventional submarines. There seems to be no reason why those earlier spindle hull boats fitted with bow hydroplanes should not have adopted the same diving procedure and they probably did.
19. Experience on service at the beginning of the 1914-18 war really brought out the importance and necessity of being able to dive in the quickest possible time consistent with safety. That is not to say no attention was paid to the time taken to dive in designs prior to that date. In the D Class design in 1906 consideration had been given to improving habitability and safety by increasing the reserve of buoyancy but it was considered undesirable 'to incur the great complication and bulk of flooding and venting apparatus necessary to deal with the additional ballast water with the rapidity which is essential for the efficiency of any submarine boat'. This was undoubtedly a reference to the double-hulled vessels which some foreign nations had adopted and which were notoriously slow diving boats. In the Laurenti design boats the spaces above the baling flats in the external tanks were flooded through valves fitted just above the flats. Because of the time taken to flood these tanks it was necessary later on to remove the flooding valves and replace them by open holes.
20. Little actual information can be given about actual diving times in any boats up to the H Class. Crash diving trials carried out in the first L boat in December 1917 showed that she could crash dive from full buoyancy to 30ft in l½ minutes. The disadvantage of the K Class was that they took about 5 minutes to dive although the specified time to close down and secure the boiler room, funnel etc was only 30 seconds. The M Class were quick diving undoubtedly because of the large buoyancy of the gun chamber high up in the vessel which accelerated the downward movement when main tanks started to flood, and it is stated that when the gun became submerged its buoyancy checked the boat at about 20ft.
21. By the late 1920's the ability to dive quickly had become a point of major consideration. The need was that a submarine should be able to submerge from any state of buoyancy in the minimum time compatible with safety and of course the complaint was that none of our submarines dived quickly enough. The writer took considerable interest in this subject at the time and wrote as Paragraphs 22-24 below in 1935. A lot of this may now be obvious but is perhaps worth repeating.
22. In war patrol condition a submarine will be trimmed down to the lowest possible buoyancy to decrease silhouette and time of diving, and the calmer the sea the more will she be trimmed down. When under way the actual minimum time that can be accepted to submerge until the conning tower becomes awash is that taken to shut off the engines and close the conning tower lid. In no case will this be much less than 10 seconds. In many cases it will be more. From observations made in Seahorse and Porpoise I think it is impossible to make a submarine submerge from the condition with the conning tower hatch awash faster than at about 2 feet per second. In Seahorse from full buoyancy on the surface with Q tanks flooded before commencing the dive, planes hard over and Q not blown until after passing periscope depth, the submarine passed from 20 feet to 35 feet in 8 seconds. The inclination bows down was over 6°. Fifty-four seconds had elapsed before 20 feet was reached so there was no question of a hang-up through main tanks or casing not being flooded. From these figures therefore the minimum time to dive to periscope depth that can be expected under ideal conditions (i. e. in a calm sea with submarine trimmed down to the fullest extent, engine clutches out and conning tower hatch closed in 10 seconds) is in Swordfish Class with 34ft periscopes 20 seconds, and in Odin Class with 40ft periscopes 22 seconds.
Table 18. 1 gives a summary of actual times to dive to periscope depth for various classes of submarines reported between 1931-1934. The figures quoted for diving from a low buoyancy condition appear to be high, but it must be remembered that in all these cases it is not the war condition low buoyancy. Furthermore, from Odin Class onwards the forward hydroplanes are out of the water and In general most of the casing was out as well.
Diving trials were carried out to obtain diving times under war patrol conditions of buoyancy with the following results:
|Conning tower hatch covered in|
|(a)||At cruising stations with Nos 3 and 6 main tanks blown for 15 seconds and blowers on all tanks for 4 minutes. Water level 3½ft below top of casing,||52||55|
|(b)||At diving stations Nos 3 and 6 main tanks blown for 10 seconds. Top of casing awash.||16||26|
|(c)||As (b) but water level 1ft below top of casing||40||42|
In cases (b) and (c) the vessel was stopped. It took 22 seconds in Orpheus to drain the clutches, so it could have been possible to cover the conning tower hatch before clutches were drained in exceptional circumstances. It may not have been possible to cruise at such low buoyancy as in (b) but there is always the occasion of charging without being under way. If of little value in themselves, these figures do show that from the times quoted in Table 18. 1 considerable reductions can be made to give the times for diving from the war condition low buoyancy. I have been informed by officers serving in Odin Class and Parthian Class in China that these submarines can and do frequently dive from low buoyancy in 20-25 seconds. They appear therefore to be as near the minimum as can be accepted with safety.
23. The minimum time to dive from full buoyancy to periscope depth, which it is practicable to obtain, appears to be of the order of one minute. The times for the later classes mentioned are better than the earlier boats when allowance is made for the increased periscope depths.
24. The factors, which influence time to dive, are:
- (a) Capacity and size of main tanks and time to flood. The ideal is a large number of small tanks but this is impracticable in complication, cost and operation. The lower the tanks are placed the quicker they will flood. Trials in Oberon showed that the majority of the main ballast tanks flooded in times varying from 17 to 32 seconds. However, Nos 9 and 10 Port and Stbd tanks which extended above the normal WL took 50, 47, 40 and 38 seconds respectively to flood. The times for No 9 tanks are excessive and suggest that the venting and/or flooding was restricted. In Osiris in which all the main tanks were below the normal WL the maximum time to flood any tank was 30 seconds with an average time of 23 seconds.
- (b) Buoyancy high above the waterline is an advantage in decreasing diving time-It gets the submarine moving downwards sooner. A large conning tower is a case in point, but it has the disadvantage of increasing silhouette. However, superstructures and buoyancy tanks above the WL must be completely free venting.
Regarding the need for free venting some interesting figures were obtained in Oberon which did (a) some diving trials in April 1927 and (b) further trials in October 1927 after cutting additional holes in the superstructure. The results were
|To 25 feet||3 mins 10 secs||1 min 50 secs|
|To 30 feet||3 mins 23 secs||2 mins 18 secs|
After further investigation and taking a mean of five dives Oberon got the following results:
|With superstructure covers closed.||To 24 feet||1 min 37 secs|
|Best dive||1 min 30 secs|
|With superstructure covers open.||To 24 feet||1 min 22 secs|
|Best dive||1 min 11 secs|
The above figures were all obtained without using the quick diving tank.
- (c) Above water hydroplanes are a disadvantage since they come into play so late in the diving. It was extremely difficult in the post war submarines with folding hydroplanes in the casing to get an angle on the submarine in the first 40 seconds or so.
- (d) Quick diving tanks help diving. They are however liable to be a menace from air-bubbles when blown and it is doubtful whether the complications Involved in trying to cure this menace justifies the decrease In diving time, especially In small submarines with a large reserve of buoyancy. However, it must be accepted that Q tanks had a big effect. In Otway it took 3 minutes to reach 40ft normally; with Q tanks it took only 1^ minutes.
In January 1930 RA (S) objected to using Q tanks when diving normally. He stated that 'the true function of the tanks called quick-diving tanks is to enable quick depth changing in an emergency either from the surface or submerged or to enable sudden changes of trim due to unforeseen circumstances to be quickly corrected. in my opinion they should not normally be used for submerging the vessel'. He considered that speed of submergence should not be judged from any other standpoint than that, when submerged, the submarine is at once in perfect trim and under perfect control. It was because of this that quick-diving tanks were called by a letter Q only. If only used in an emergency, then the menace from air bubbles giving away the position when tanks are blown becomes more acute. As far as can be judged if diving times from the low buoyancy condition comply with the requirements laid down, then Q tanks are undesirable. Their only redeeming feature appears to be to get a submarine down in a seaway when it is necessary to get the vessel bodily heavy.
25. Until the mid 1920's the range of sea water density in which submarines were designed to dive was quite limited. Very late in the building of the Odin Class (towards the end of 1928) a requirement to be able to dive in fresh water was brought in. In fact, these vessels had to be capable of diving in sea water with sg varying between 1. 00 and 1. 30. In a boat of the Odin size, the difference in the amount of internal water carried when diving in fresh water and sea water of sg 1. 030 was 53 tons. Odin had not been designed to meet these conditions and if diving in fresh water at the beginning of a patrol all compensating water and the fresh water in external tanks had to be discharged overboard. This left on board only 6 tons of water for trimming purposes. This was of course, insufficient for a boat of this size and in the Parthian Class which was not too far advanced an additional compensating water tank was fitted. It was still necessary to sacrifice part of the external fresh water in certain conditions. It is to be expected that later on modifications were made to overcome these restrictions.
This capability of being able to dive In waters from sg of 1. 00 to 1. 30 remained for all submarines until the Second World War when additional emergency fuel was required. The range of sea water density In which the boats were designed to dive has since decreased to 1. 005-1. 03, then to l. 01-1. 03 and finally to l. 015-1. 03.
26. Some interesting trials were carried out in Thames early In 1934 to obtain the deceleration in speed when diving. From the order to dive from full buoyancy running on the surface at various speeds the speed decreased in 30 seconds from:
- (a) Diving speed 18 knots to 10. 7 knots
- (b) Diving speed 15. 7 knots to 9. 8 knots
- (c) Diving speed 12. 0 knots to 8. 7 knots
The following times are relevant to each trial:
|Engines stopped revolving||3½||3||2½|
|Engine clutches out||2½||1½||2|
|First starting switch made||8||6½||8|
It took 40 seconds to turn out the forward hydroplanes.
27. It may be said that a high mean speed of attack is required in a submarine as much as a high maximum speed. Trials were carried out in Porpoise with the following results:
- (a) To accelerate from slow group-down (2. 6 knots) to 6 knots took 56 seconds.
- (b) From slow group down to 7 knots took 80 seconds.
- (c) From slow group down to full speed (7. 9 knots) took 134 seconds.
- (d) To decelerate from 7 knots to 4 knots took about 90 seconds.
28. The tactical diameter and advance of a submarine turning submerged is important and some turning trials results under full helm are: