While the folks out on the computer floor were concerned with message switching operations, we in the technical control center were concerned with the data circuits that connected the switch to its tributary stations and other switching centers.
Autodin circuit facilities were built using the RED/BLACK concept, in which unencrypted data lines (the RED side) were kept physically separated from the encrypted circuits (the BLACK side). A typical data circuit was connected from the Accumulation and Distribution Unit to the RED side patch panel in tech control. From there the circuit went to the crypto center where it connected to a crypto machine. The encrypted output then came back to tech control where it was connected to the BLACK side patch panel, then on to a modem for transmission over an audio circuit.
Circuit access for troubleshooting purposes was made very convenient by banks of relays in the RED and BLACK patch panels. When the System Console operator reported a problem with a circuit (usually “no idles”) they reported it using the tech control jack number; that is, the number of the patch panel jack it appeared on. A tech controller merely had to punch the jack number into the tech control console to pull the circuit up for testing. You could view either the RED side signal or the BLACK side signal on an oscilloscope, and view the flow of control characters (on the RED side) using an integral character reader. If it was a teletype circuit, you could direct it to a teleprinter for monitoring, if required.
The synchronous crypto devices could be reset if they lost synchronization by pressing a button on the tech control console. This was the usual response to a loss of idle patterns. If a crypto unit failed, it could be replaced by plugging a patch cord into the RED patch panel to shunt the signal off to a crypto spare, plus a corresponding patch on the BLACK patch panel to connect the spare crypto unit back to the original modem.
Each tech control shift (24/7) had to have at least one person on staff who was technically qualified (demonstrated through a series of tests) to direct the telecom operations. Such a person was said to be “facility rated.” Again because of the Vietnam build up, communications personnel were being rotated on a frequent basis, so there was always a shortage of facility rated individuals to assign. Coming in with a computer background, I achieved full facility rating fairly quickly, which saved me from several hideous reassignments because of the lack of qualified personnel.
This led to a quirky work-around with dual shift supervisors. The individual with the highest rank was denoted the “military” supervisor for the shift, but if he (there were no female tech controllers in those days) wasn’t facility rated, another airman with a facility rating was denoted the “technical” shift supervisor and his word was binding when it came to determining what should be done to effect recovery of a failure.
I remember the confusion on the other end of the telephone when a controller at another facility, who had been speaking to sergeant such and so, asked for the shift supervisor and I answered, “Airman Ryals.” He said, “No, I wanted the shift supervisor.” I replied, “You’ve got him. What do you need?”
Because I had been facility rated for a while and generally understood the system, I was assigned to be the unit training supervisor. This didn’t sit too well with some of the troops who greatly out ranked me who were relegated to shift work while I got the cushy day job. My main task assignment was to rewrite the tech control training manual.
The two manuals used previously at Norton tech control were essentially just reprints of Western Union system documentation. They did a good job of explaining the various facets of the system, but they didn’t really tell you how to troubleshoot a failure and recover from it. I know many of the older controllers would have preferred to have a decision tree that they could follow, but there was insufficient time to develop one that would have been comprehensive. Instead, I opted for clear language explanations of how the parts worked, and hoped that coupling understanding of the system with controllers’ training and experience in trouble isolation would lead to a good result.
Long before word processing software was thought of, I was producing camera-ready pages for the manual by typing them on an electric typewriter. Neither of the two error-correcting methods I had available (white-out, and correction tape) produced a result that was clean enough for reproduction, so I had to type each page perfectly. One day my wife, who worked at the same facility (she was a WAF and operated the compound terminal), came by when I was away from my typewriter, and decided to “help” me by finishing the page that I had mostly completed. Instead, she wound up with a typo in the last line and I had to start the page all over. Thanks, dear.
Modems and communication lines for Autodin were provided by Western Union, which had a direct connection to their national microwave radio system for the long haul circuits. They also had a fat cable to the local telephone company for use if their microwave system failed for any extended period of time. Whenever a tech controller determined that the problem was with a circuit, he handed it off to Western Union personnel for resolution.
Western Union also had other government data circuits passing through their facilities, the most interesting of which was the bomb alarm. This was in the late 60′s, during the height of the Cold War, and NORAD had an elaborate matrix of low-speed data circuits connected to transponders mounted on telephone poles in various cities around the country. These transponders were polled every few minutes to see if they were still there. Because the location of the transponders and the route of the circuits were known, a computer at NORAD headquarters could figure out from which signals were missing what cities had been bombed in a nuclear attack. Pretty scary thought, isn’t it. We were aware of the bomb alarms because a bank of signal relays near our tech control console chattered in a distinctive pattern each time the transponders were polled. I am pleased to report that the bomb alarm system never saw any real use.
Probably my most exciting time in tech control was related to a forest fire that burned through the microwave repeater connecting our site to the Western Union system. The heat was so intense that it melted antennas and waveguides off the tower.
I was working the tech control coordination position and as a result of the Norton site being essentially cut off from the rest of the world circuit-wise, I had to send out bunches of failure report messages and call all over the place. I had never before in my short military career had the occasion to speak with so many colonels and generals. Our switch served SAC bases, so Strategic Air Command headquarters was most interested in tracking our outage and attempts to recover from it. Likewise for the Tactical Air Command and a bunch of other headquarters.
Western Union personnel ordered up a boat load of circuits from the telephone company and restored as many connections as they could. Sometime in the middle of the night a Western Union microwave installation team miraculously managed to hang large microwave dishes on a 50-foot tower, while being buffeted by 60 mph winds flowing across the ridge and into the base of the fire. Twenty-four hours later all was back to normal.
Have a question for Phil Ryals on this subject? Leave a comment below!
For more information on the Autodin: