Sellner’s real interest was a dream he had first had as young man—to devise a submersible which would rise and fall according to a system he had invented using the liquefaction of air. In its gaseous state, air is diffuse and light, but when liquefied—as in a refrigeration system—it becomes comparatively compact and heavy. The change from one state to another creates the potential for a buoyancy device, first heavier than, then lighter than, water.
Here, a little scientific information is needed on the subject of buoyancy. The details are complex, but leaving aside variables such as the changing temperature and salinity of the surrounding water, the basic ideas are quite simple. In a normal submarine, such as a U-boat, vertical movement is controlled by allowing seawater into the buoyancy tanks placed on the outside of the hull in order to descend, and blowing out this water with compressed air in order to ascend. More water in the tanks makes the vessel heavier; more air, lighter. This system needs the submarine to store and then waste large amounts of compressed air. Piccard’s bathyscaphe, which of course went much deeper than a conventional submarine, worked instead by using a large buoyancy tank, full of gasoline, which being lighter than water caused the vessel to float. Metal weights attached to the underside of the bathyscaphe acted as ballast, making the whole vessel, now heavier than the water, descend; releasing the metal ballast when the desired depth had been reached allowed the gasoline-filled buoyancy tanks to carry the vessel upwards, much in the way that a hot air balloon carries a basket containing the passengers upward.
Sellner’s system is similar to a Piccard bathyscaphe in having a large buoyancy tank, but works more like the submarine in varying this buoyancy without the use of metal ballast. The compressed air system of the submarine is replaced with a liquefaction system. At the surface, the buoyancy tank is filled partly with seawater and partly with air, the proportion controlled so that the vessel just floats. The air is then cooled and liquefied, significantly reducing its volume and allowing more seawater into the tank through an opening in its bottom; this makes it heavier and causes the vessel to descend. When the pilot wishes to ascend, the process is simply reversed: the liquefied air is evaporated and expands, pushing the seawater out of the tank, and the vessel, now lighter than water, moves upward. The advantages of Sellner’s system are that no air is lost, since it is simply liquefied and then vaporized, and, more importantly, that the extreme change of density in the air gives the vessel much more lifting power than simply blowing out the buoyancy tanks with compressed air, making such a device suitable for lifting heavy objects such as undersea wrecks.
There is one complicating factor, particularly relevant for vessels descending into the depths of the abyss. The buoyancy tanks of both Piccard and Sellner’s bathyscaphes are open at the bottom to allow seawater to enter. The external pressure of the surrounding water increases as the vessel descends, compressing the contents of the buoyancy tanks and making it denser—this allows more seawater to enter, which in turn causes the rate of descent to accelerate. The vertical movement of the vessel has to be carefully controlled by continually adjusting the buoyancy, which requires considerable skill. The ascent has to be similarly controlled, because the reverse is occurring and the liquid in the buoyancy tanks is expanding as the external pressure of the water decreases, so that the vessel tends to ascend faster and faster. The whole issue of buoyancy is not just concerned with pragmatics, but also with a certain aesthetic; there is a beauty in the delicacy of these systems of balance and equilibrium, in the constant, minute adjustments made in response to small differences in the external environment.
Given the technical complexities involved in engineering submersibles, it’s even more remarkable that Sellner, who prior to 1946 had rarely even seen the sea, let alone design a ship, was able to convince his captors that his invention was viable. The Soviet supervisors, suitably bribed with vodka, decided to support his construction of a trial machine because if it turned out to be as buoyant as Sellner promised, it could be used in salvage efforts to raise ships sunk in the war. Sellner’s real aim, however, which he concealed from almost everyone except some trusted fellow prisoners, was to construct a primitive bathyscaphe capable of descending several hundred meters under the sea, a most unlikely ambition for a half-starved prisoner of war isolated above the Arctic Circle. Sellner must have known something of Piccard’s efforts in the late 1930s—halted by the war—to construct a bathyscaphe, but quite why he wished to follow the Frenchman’s example, rather than simply devise a submersible suitable for raising wrecks in a harbor, he never explains. In the transcript of the recording, he talks in great detail about the technical issues, but rarely about his desires. “Life had,” he recounted to Taillez, “no more meaning for me and no more value, apart from that absurd idea that I had the means of carrying out.”
Sellner selected an abandoned U-boat hull, some thirty meters long, to use as the buoyancy tank. He converted this hull for use with his gas liquefaction system, using the internal compartments between the bulkheads as variable buoyancy chambers so that the hull could be made to descend and ascend underwater. Sellner also discovered among the military junk a metal sphere some two meters in diameter, possibly originally a floating German observation post. This sphere was adapted and attached to the underside of the U-boat hull, as a chamber for the proposed one-man crew—himself. Searchlights, batteries, hatches, and so on were all also converted from available scrap. The whole vessel was constructed in different workshops within the camp so that no one except Sellner himself would be able to understand the ultimate aim.