The quench circuit lost its reference probe on the inner branch as the collar distorted, and the board went blank on that channel.
It happened between one reading and the next. The quench had six reference probes, one on each of the six branches nearest the collar, set into the channel walls where the stepped lining met the body: thermocouples, reading the gradient along each channel as the quench water came down the jackets. Without them, nobody could tell whether the grain was setting evenly.
The probe on branch three, the inner branch, beside jack four, had been reading steadily since the seating.
Then it read nothing.
On the board, its line simply stopped, neither falling nor rising. A line drawn across the slate, and then no more line. Where the next reading should have been, a blank.
“Probe three,” Eshel said.
He read the rack. The probe’s channel. Nothing on it. He checked the cable, at the rack’s end. Sound. He looked at the board.
“Gone,” he said.
