DESIGN, CONSTRUCTION, AND OUTFITTING 69 



to increase negative buoyancy. To eliminate this unwieldy method of ballasting, Sealab II was 

 designed along submarine principles. The variable ballast would be water, stability would be 

 maintained during all phases of the submerging operation, and negative ballast on the bottom 

 to insure firm seating would also be water. NOTS Pasadena developed a winch-counterweight 

 lowering system (Chapter 18) that made lowering against negative buoyancy feasible and de- 

 sirable. Flooding had to be controlled simply and externally, since Sealab in this phase of 

 operation was unoccupied and sealed. 



The condition requiring full working pressure internally at the surface made the Sealab 11 

 cylinder an internally pressured nonfired vessel under the ASME Boiler Code. The code gov- 

 erns the structural design, construction, tests, and inspection. The tables in the code indicated 

 that one-inch-thick mild steel was sufficient for a working pressure of 125 psi, ample for the 

 desired 250 ft. A structural- strength test, hydrostatically, to 1-1/2 times working pressure was 

 also required. The end cappings for the cylinder were required to be ellipsoidal dished heads 

 of proper curvature and depth. Their unique method of fabrication will be discussed later. 



The use of water as variable ballast and the desire for reduction in internal volume to save 

 helium combined to provide internal ballast tanks. These were built into the overhead of the 

 cylinder with sufficient capacity to allow proper reserve buoyancy on the surface and adequate 

 negative buoyancy on the bottom, as previously discussed. The structural details necessitated 

 making these tanks "soft", i.e., incapable of withstanding pressure differentials in excess of 

 15 psi across their lower boundary. 



To preclude the necessity for a porthole (viewing port) capable of withstanding full internal 

 pressure and to allow large (24-in.) ports, structural covers were provided internally to con- 

 strain the pressure. When opened on the bottom they then exposed the port viewing glass to a 

 pressure differential of slightly over 6 psi. This allowed the use of 1-in. plexiglas as the 

 viewing-glass material. 



Previous data on equipment behavior in helium existed only in what could be obtained from 

 Mine Defense Laboratory observations during Sealab I. Many commercially obtainable items 

 functioned well, and this fact was accepted, tempered wherever possible by actual tests prior 

 to any operational certification. As an example, commercial dehumidification units used ap- 

 parently successfully in Sealab I. The same type units were procured and tested in helium at 

 the Sealab II working pressure. It was noted during operational test that the compressor motor 

 did not function properly. The malfunction was traced to a metallic relay in the motor start 

 circuit that apparently changed its characteristics when operating in helium. Replacement 

 with a sealed-unit relay restored normal operation. The rated capacity of 47 pints/day was 

 never conclusively checked, however. A standard Navy-type refrigerator-freezer was pro- 

 cured, additional insulation added, and the unit was tested in helium at working pressure. The 

 thermal sensors in the refrigeration compartments were of the fluid-filled bulb type and would 

 have crushed under the extreme pressures. Once these were replaced with thermocouple type 

 sensors the refrigerator and freezer functioned properly. 



No data were available on heat losses and heat input during Sealab I, although qualitatively 

 the aquanauts seemed comfortable at a temperature of 85° - 90° F at relative humidities between 

 60 and 70 percent. It was obvious that these observations were all that was readily available to 

 design the heating dehumidification, and insulation systems. A psychrometric chart for a He- 

 Nj - O2 atmosphere was nonexistent. A qualititive analysis indicated that since ambient tem- 

 peratures for Sealab 11 would be 20° to 30° F cooler than for Sealab I a much higher heating 

 capacity (needed also to allow for many more men and a larger volume) and more insulation 

 were required. Consequently, 25 kw of heat were provided and 2 in. of cork insulation on the 

 inner surface of the shell were installed. These perforce were based on the most rudimentary 

 qualitative analysis. A concrete deck was used for several reasons: 



1. Structurally simple and economical 



2. Provided additional ballast 



3. Reduced further the internal cubic 



