The actual casing for the device was already fabricated. Ellipsoidal in shape, it was 98 centimeters in length by 52 in extreme breadth. Made of steel one centimeter in thickness, it had to be strong, but not grossly so, just enough to hold a vacuum. Also ready for installation were curved blocks of polyethylene and polyurethane foam, because a device of this sort required the special properties of both the strongest and the flimsiest materials. They had gotten ahead of themselves in some areas, of course, but there was no sense in wasting time or idle hands. On another machine, workers were practicing yet again on a stainless-steel blank that simulated the folded-cylinder plutonium. reaction-mass primary. It was their seventh such practice session. Despite the sophistication of the machines, the first two had gone badly, as expected. By number five, they had figured most of the process out, and the sixth attempt had been good enough to work - but not good enough for Fromm. The German had a simple mental model for the overall task, one formulated by America's National Aeronautics and Space Administration to describe the first moon landing. In order for the device to perform as desired, a complex series of individual events had to take place in an inhumanly precise sequence. He viewed the process as a walk through a series of gates. The wider the gates were, the easier it would be to walk through them quickly. Plus/minus tolerances reflected slight closure of the individual gates. Fromm wanted zero tolerances. He wanted every single part of the weapon to match his design criteria as exactly as the available technology made possible. The closer to perfection he could get, the more likely it was that the device would perform exactly as he predicted or even better, part of him thought. Unable to experiment, unable to find empirical solutions to complex theoretical problems, he'd over-engineered the weapon, providing an energy budget that was several orders of magnitude beyond what was really necessary for the projected yield. That explained the vast quantity of tritium he planned to use, more than five times what was really needed in a theoretical sense. That carried its own problems, of course. His tritium supply was several years old, and some parts of it had decayed into 3He, a decidedly undesirable isotope of helium, but by filtering the tritium through palladium he'd separate the tritium out, ensuring a proper total yield. American and Soviet bombmakers could get away with far less of it, because of their extensive experimentation, but Fromm had his own advantage. He did not have to concern himself with a long shelf-life for his device, and that was a luxury that his Soviet and American counterparts did not have. It was the only advantage he had over them, and Fromm planned to make full use of it. As with most parts of bomb design, it was an advantage that cut both ways, but Fromm knew he had full control over the device. Palladium, he told himself. Mustn't forget that. But he had plenty of time.