It was twenty minutes past eight. The pressure in the fuel tank was –0.4 psi. At least that’s what the gauge said. The PTPMU hadn’t been calibrated for negative readings, and the actual pressure could have been even lower. The pressure in the oxidizer tank had risen to 23.4.
The PTS team chief, Heineman, asked if they could be evacuated.
Childers and Holder finished shutting off the power to the missile and, at the direction of the command post, turned off the air-conditioning in the silo, too. Although the air conditioner cooled the silo, it could also produce a spark and ignite the fuel. Childers didn’t want to evacuate, and neither did Holder. They wanted to stay put. They were good friends, discussed the issue quietly, and agreed about what should be done. Mazzaro’s wife and Fuller’s wife were pregnant; Mazzaro’s was due to have the baby any day. We ought to let the other guys leave, Childers and Holder decided. We’ll stay here and ride this thing out. They volunteered to remain in the control center. It was important for someone to stay there. The two could monitor the PTPMU, keep an eye on the hazard lights, or even open the silo door. They felt confident that the blast doors would hold. “If the missile blows,” Holder said, “I think we’ll be OK.”
The strength of a blast wave is measured by the overpressure it produces — the amount of air pressure greater than that found at sea level, measured in pounds per square inch. An overpressure of 0.5 psi shatters windows. An overpressure of 2 psi destroys wooden homes, and an overpressure of 8 psi knocks down brick walls. The Titan II silo door was designed to withstand a nuclear detonation with an overpressure of 300 psi. The underground blast doors were even stronger. They were supposed to protect the crew not only from a nuclear detonation outside but also from a missile explosion within the silo. The enormous doors on both sides of the blast lock, theoretically, would survive an overpressure of 1,130 psi.
At half past eight, about two hours after the accident, the wing commander ordered everyone to evacuate the complex. The pressure in the stage 1 fuel tank had fallen to –0.7 psi. The safety of the crew could not be guaranteed. The missile could explode at any moment.
While Mazzaro and Childers stuffed top secret documents into the floor safe, Holder and Fuller put on gas masks and went downstairs to level 3 of the control center to open the emergency escape hatch. It wasn’t easy. The hatch was a metal dome attached to the wall by thick screws. The men took turns unscrewing them with a large ratchet. The hatch opened a little more with each twist of the ratchet. Holder took off his gas mask. He was out of breath and didn’t think the mask was necessary — yet. He’d opened the escape hatch a few times during inspections. But he’d never been inside the narrow, ten-foot tunnel beyond it. The tunnel led to a steel ladder, embedded in the concrete wall of an air shaft, that traveled about fifty feet upward to the surface.
Childers couldn’t close the door to the safe. There were too many documents crammed inside it. The command post told him not to worry about them, to leave the door open. But he felt uncomfortable leaving it that way. Although the launch keys and the cookies were securely locked in a different safe, the emergency war order checklists were among these documents. Someone who got hold of them would learn a great deal about how to issue a launch order and how to countermand one. The issue soon became moot. The safe wouldn’t close, the crew had to evacuate, and nobody else was likely to be entering the control center soon.
Once the escape hatch was open, the PTS team went down to level 3, wearing gas masks. The missile combat crew members grabbed their handguns and put on their holsters. Before departing from the control center, they took the phone off the hook, so that officers on the line at Little Rock could hear if any Klaxons, alarms, or portable vapor detectors went off. And the crew switched the diesel generator to manual. That way the generator wouldn’t automatically turn on if power to the entire complex was later shut off, an option being considered. Motors and pumps in the equipment areas of the silo were still running — because the circuit breakers to shut them off were inside the silo. Ideally, the crew wouldn’t have left anything running that might cause a spark. But they’d done the best they could. They put on their gas masks and hurried downstairs.
Fuller entered the hatch first, carrying a flashlight. It was pitch black in the narrow tunnel as he crawled toward the air shaft on his hands and knees. The PTS team and Serrano went next. Childers told them to look after the trainee.
“Put him in the middle of you guys,” Childers said, “because I’m not going to have him hurt.”