From Jutland to Junkyard: The raising of the scuttled German High Seas Fleet from Scapa Flow - the greatest salvage operation of all time

by George, S. C.

  A simple device to show how far men had winched a ship from the bottom was the attachment of an iron rod to the deck. The rod was marked to show the level when the lift began, and as it emerged during the winching, it could be seen at a glance how far the ship had been winched from the bottom.

  five ships were refloated between February and April 1926, the last of them, G104, on the last day of the month. In 20 months 23,000 tons had been raised.

  While the destroyers were being lifted, the positions of the capital ships had been charted. Those which were completely submersed were located by sweeping. The approximate positions of most ships were known, and an hour’s work was usually long enough for a ship in this category to be pinpointed. Then divers descended to fit marker buoys foreward and aft, and a tug was moored in position over the wreck. When Bertha was used for this purpose in the latter stages of the operation, eight 7½-ton anchors were needed to hold her in position because of the strong, variable winds encountered in Scapa Flow.

  7

  Hindenburg Fights Back

  WHILE THE DESTROYERS were being lifted, plans were well advanced for raising Hindenburg, a battlecruiser built between July 1913 and the middle of 1917. She was one of the three battlecruisers of the Derfflinger class, the only battlecruisers at the time of their completion to carry 12-inch guns, superimposed turrets and tripod masts, and to be flush-decked. Many experts considered these ships to be the best capital ships of their time.

  Hindenburg, the biggest ship of the fleet, had not taken part in the battle of Jutland. She lay, perfectly upright in 70 feet of water, half a mile west of the island of Cava, apparently presenting Cox with an easy task, though the experts again prophesied failure. Even at this depth, her great tripod mast, her funnels and a good part of her upperworks showed above water. At spring tides her deck was just awash. Ordinarily her flagstaff was almost covered, and nearly 28 feet of water lay over her aft quarter deck. Her huge size and weight made it impossible for her to be raised by the method so successful with destroyers. Cox’s original intention was to float her and use her as a pontoon for taking the guns and turrets off Seydlitz, and then for raising other vessels.

  Various plans for floating her had been investigated. One was a project for raising her by means of air-filled pontoons; another, submitted by a German firm of salvage engineers, involved freezing up all deck apertures and then using compressed air to expel the imprisoned water. Both were rejected in favour of Cox & Danks’ own ideas.

  The big dock, like the first one, was cut in two, and each part was fitted with workshops complete with generators, pumps, compressors and diving equipment, besides kitchens and mess rooms.

  Cox and McKenzie went down in diving suits to make their first survey of Hindenburg. It was decided to patch all the holes in her hull, position dock sections on each side, raise her by pumping out the water from her hull so that she could be lifted by her own buoyancy, and then to moor her to the four sections of the floating docks for her tow to the breakers.

  They had a stroke of luck when some of the plans of the ship were found in the control room, despite the fact that one of the divers said he had examined the ship for documents without success three weeks after she had been scuttled. Among plans now discovered was the pumping plan which showed the arrangement and positions of the control points for each valve. This enabled four experienced divers to examine the controls of all the valves which could not be reached from outside. They carried down no submersible lighting, for it would not have penetrated the inky darkness caused by mud and particles stirred up when they moved. Everything had therefore to be done by touch.

  Sixteen divers were employed, working in pairs. Sometimes they, too, had to work in darkness by touch, at other times by Siebe Gorman submersible lights operated from the docks. They reported that all Hindenburg’s open sea valves were so damaged that they could not be closed again. One diver said that when he was lowered into the ship, it looked like a vast submarine forest because of the density of marine growths, and for a time he did not know where he was. He found glasses and champagne bottles strewn about. and bunks undisturbed with mattresses still in them.

  William Hourston, a local photographer commissioned by The Times to take photographs of the ship, had a frightening experience when, five decks down, he saw that he had less than half an inch of candle to light his way back. The flame flickered out just as he dimly saw daylight far above him. On his next descent, to provide against a similar occurrence, he took with him a ball of twine which he unwound as he made his way about the interior.

  The most interesting object found was the beautifully executed model of a warship. It was two feet long and showed all the outside structure, deck fittings, gun turrets etc. One room was found clean and dry, without a drop of water in it, after seven years’ submersion.

  In the summer of 1926 work was stopped temporarily by the great general strike. Two hundred tons of coal were needed weekly for the boilers upon which so much of the equipment depended. The price of coal soared from £1 to £4 15s a ton, well beyond Cox’s means at that time as he was paying £1,000 a week in wages and another £1,000 in fuel. However, this time luck was with him, for through the exposed armour-plating of Seydlitz, whose side rose above the 65 feet of water in which she lay, they could see that the coal bunkers were full, and they were able to use this coal throughout the strike.

  Quick-setting cement was used to seal the damaged bottom valves from inside, and then the openings had to be patched. This was a colossal task as 800 holes had to be patched or plugged and made watertight. Some of the smallest patches were for portholes, and the largest patch was over a hole left by the removal of a badly corroded funnel. This needed a patch 40 feet long by 21 feet wide by six inches thick; it was made ashore and weighed 11 tons. Then it was towed out, lifted by crane on to a dock, lowered into position and used to plug the gaping hole. Five months passed in patching and sealing. Except where wooden plugs could be used, wood planks varying in thickness from one inch to six inches were employed. As the top deck of No 2 turret was just above high water at the spring tides, operations began at that point. All machinery was removed from the centre of the turret and, as the level of the water fell by pumping, all internal obstructions were cut away to give a clear passage right to the bottom of the ship so that it was possible to place submersible pumps at the lowest level.

  The sequence of pumping operations had been carefully considered when the work of patching was well advanced. There were two classes of pumping sets, the first being self-contained and driven by oil engines. These comprised several old engine-driven 12-inch Conqueror sets built by W.H. Allen, Sons & Co Ltd of Yoker, Glasgow and Lindsay, and by Swan & Hunter Ltd of Newcastle-upon-Tyne. The second class was mainly a series of 20 or more submersible pumps built by Submersible Motors Ltd of Southall. The pumps were connected by cab-tyred sheathed cables to the generating plant on the pontoons, and current was supplied by alternating-current generators driven by belting from steam engines. The steam was supplied by two water-tube boilers.

  The 18 pumps which were to be used were connected and tested. On 6 August pumping began, but many of the patches were not watertight. This was no fault of the workmen, for divers discovered that little fish, saith, had acquired a taste for the tallow used with the packing to make the patches watertight, so the tallow was made unpalatable by mixing it with cement.

  On the night of 24 August Hindenburg was flooded by a gale which drove destroyer G38, being used as a breakwater, on to the end of one of the floating docks, pounding it so severely that some plates burst and leakage to a tank was caused. The destroyer had to be removed immediately and the damaged dock repaired, during which time Hindenburg had to be completely flooded again. By the end of the month the patches had been re-packed and pumping began again. Considerable time was saved by making a template for each patch, getting the divers to fit it, and then making a patch from the template. The patches were invariably built plank by plank ashore a
nd completed ready for fixing before being sent down. This gave a much closer fit than would have been possible had they been made under water. Between patch and ship was a ‘pudding’ joint, which was a canvas bag, the pudding, padded with oakum, being lightly nailed round the outer edge of the patch. It was about three inches wide and from two-and-a-half to three inches thick, but when the bolts of the patch were fully tightened, it was squeezed to a thickness of about a quarter of an inch.

  It had been assumed that Hindenburg’s stern lay on sand and shingle into which it could be forced as the bows were raised, thus providing stability. But in fact the propellers were from ten to 12 feet above solid rock on which the ship lay, and not supporting it as they had expected. It was apparent later that they had been trying to balance Hindenburg’s 100 feet of width on a keel only three feet wide, for as the days passed, no matter which end was first lifted, Hindenburg listed and began to turn over. On the Saturday she suddenly rose 20 feet between 07.00 hrs and 08.00 hrs, her main deck rising well above water and exposing her battery deck, the flagpole on her stern and the starboard quarters. Acting against all advice, Cox decided to maintain the 25,000-ton mass of metal in an upright position by attaching two five-inch steel hawsers from the top of her mast to a destroyer sunk off the island of Cava. As might have been expected, the hawsers snapped like violin strings and once again Hindenburg heeled over and went down.

  A total of eight 12-inch and 30 six-inch pumps which could discharge 3,600 tons of water per hour were placed aboard. The diving team lashed back the bulkhead doors in the forward compartment so that water could drain towards the stern and bows by their own buoyancy. Otherwise water would have been imprisoned in compartments, upsetting the trim of the ship and creating undue stresses on the hull. Divers had already commented on the heavy corrugation on some hulls due to such pressures. But when pumping began, the rate of fall in the level of water was far less than had been calculated owing to the number of small leaks which now became visible. After four or five months’ hard work these were repaired and the bows began to rise. However, the hull had to be flooded again to find new holes through which suction pipes could be passed.

  The men were busy aboard lowering the big pumping sets as the water level sank inside the ship. For this to be done, steel deck had to be cut through with oxy-acetylene burners – dangerous work on slimy steel plates. Pumps were kept going night and day, and on the fifth day the bows rose from the depths. Excitement exploded into a great burst of cheering although the men had seen 25 destroyers raised from the sea bed and might well have regarded this as ‘just another ship’.

  Hindenburg was afloat again when on 2 September a considerable amount of work was undone by a fierce northwesterly gale which, lasting throughout the night, set the dock rolling and Hindenburg heaving. There was trouble with pumps and dynamos, and the big water-tube boilers broke down. By their failure the most important and flexible pumping units were completely deprived of current, while the remaining pumps were totally inadequate for the task of righting the vessel. Cox then tried to obtain steam for the engines of the generating sets from the large tug which was moored alongside the lifting-pontoon, but because of the poor quality of the foreign coal, and the ship being stationary, a sufficient head of steam could not be raised. So many patches had been smashed that Hindenburg began to sink fast. Another £10,000 had been lost in addition to the £20,000 already spent. Cox was not a man to accept defeat, but conditions were against him, and further attempts to raise Hindenburg that year were abandoned, partly, too, because the weather at that time of year was so uncertain.

  8

  Moltke’s Last Voyage

  ON THE DAY AFTER the temporary abandonment of the attempt to raise Hindenburg, undeterred by his costly failure, Cox climbed on to the bottom of the battleship Moltke which lay awash at low tide between RYSA Little and Cava. She was bottom-up in 78 feet of water and, of all ships in Scapa Flow, Moltke was the most dangerous to navigation. She was on a fairly level, but slightly soft, bottom, with a list of some 17 degrees.

  Moltke was an enlarged version of von der Tann and had the same protection as a battleship. Five pairs of 11-inch guns were set in five turrets on the centre line of the ship. This turret arrangement also followed that of Kaiser class ships, but her low freeboard had caused her forecastle to be awash in anything like a heavy sea. Her crews had always looked upon her as being a lucky ship.

  Of the £100,000 with which Cox had begun operations, only £10,000 now remained, but Moltke would be worth a considerable sum as scrap. A surveying crew had reported that her bottom valves and other openings could be sealed from the outside. Still attached to her was a large boat made of two-ply oak in wonderfully good condition. The boat was detached, floated and towed ashore in good shape. Water inside Moltke’s hull could be driven out by pumping in compressed air after sealing apertures. The deck openings would be convenient vents through which expanding air could be released as she surfaced. Petersen, Cox’s chief diver, reported that there was ample room to blast away funnels, the crashed foremast and all other hindrances which might prevent the ship from floating. In this, as in several other ships, the gearing for the operation of the armoured doors weighing several tons had been sabotaged, giving the salvage crews much additional work. Twelve inches of solid steel in the centre sections tapering off to about four inches fore and aft, provided the main side armour, and 12-inch armour encircled the main conning tower. Seaweed and other marine growths on the bottom and hull were as high as a diver’s head and so thick that they could be removed only with knives and axes, some of the stems being as thick as a man’s wrist. This growth had to be cleared away before bottom valves could be located.

  After careful consideration of all possibilities, Cox decided to use compressed air. There was nothing new in this. Compressed air had been used in 1909 to raise SS Fleswick which sank after a collision in Cork harbour. In 1925 Captain John Iron used it to raise the monitor Glatton which had rested upside down since 1918 in Dover harbour. General Ferranti and Major Gianelli had used compressed air to float Leonardo da Vinci, a 24,000-ton Italian battleship carrying 13 12-inch guns which had been blown up and sunk in Taranto harbour on the night of 2 August 1916 by an enemy bomb, secretly introduced into one of her magazines. She had come to rest in 36 feet of water, and her gun turrets and funnels, resting on the bottom, had been forced into her hull until the upper deck was buried. When she had sunk 30 feet into the mud, her funnels had reached a bed of clay which had arrested further sinking.

  Yet, though the general principles of using compressed air had been established, the method was new to Cox. Moreover each ship presented its own particular problems. Cox had read an account of the Taranto operation, and he was further assisted by Major Gianelli who visited Scapa Flow to see his work there and exchange information with him on their respective methods of approach to salvage tasks. One major difference between the situations of the Italian ship and Moltke was that while the former lay in a sheltered position with a list of only eight degrees, Moltke’s list was 17 degrees and she lay upside down in water often whipped to a fury by storms. A ship with the lesser degree of list would right herself when raised by compressed air, but one with a much greater list could not do so.

  Figure 2. General arrangement of 100-foot airlock.

  After a week of bleak, wintry weather and October gales which stopped all work, concrete was used to block the bottom valves and lower torpedo tubes. Connections for air-pipes were made along the hull, and the compressors were run night and day. To lighten the ship’s weight, the giant propeller was loosened, worked off the shaft and hoisted out. Ten days from the commencement of pumping in air, the fore end broke surface, but the stem was still down. She also developed an alarming list, and when it was found that she was not watertight she was allowed to settle again to the bottom.

  Divers now had to work in compressed air inside the ship to seal the transverse bulkheads and thus divide the ship into watertight compart
ments. No drawings of the ship were available, so there was no alternative to piercing the hull and fixing an airlock on that part of the bottom exposed at low tide. The divers commented that under the grey, northern sky and its bitingly cold winds, working under the water was the best place and they were comparatively comfortable. Up above, seas frequently washed over the boat from which operations were conducted.

  An air-lock is an airtight, steel-plated chamber employed to prevent the escape of compressed air while men enter the hull to work in the space created by air-pressure above the level of the water. A hatch, or door, at each exit from the lock permits the safe entrance of men and materials, one hatch being opened while the other leading to the high-pressure area is shut. When the open-air hatch is shut, a man opens a valve which admits high-pressure air to the lock, and when the pressure is the same as that in the hull, he can open the lower hatch and descend into the hull. When he climbs out, the procedure is reversed. A man’s first entry into an air-lock could be a terrifying experience, for when the valve in it was turned there was a sound as though every engine in the world was letting off steam, while compressed air gushing in from the ship’s interior thickened the air like an old-time London fog, and one’s ear-drums seemed to be about to burst.

  As Cox had not used compressed air, he had no experience of air-locks either, although they had been in use for 70 years or so.

  The position for each air-lock was marked by divers on the ship’s bottom. Angle plates were then fixed round the place marked on the low side so as to be able to locate the air-lock accurately when the time came to fit it. Meanwhile other divers attached eyeplates to take the steadying guys.