We’ve talked about the early warning system in place for detecting any Cold War Soviet bomber attack, such as the DEW Line. The next key element of any successful air defense network is a method of controlling interceptions, to ensure that every attack is met, and that no more than the minimum force is dispatched to any one attack. Ground Controlled Intercept (GCI) was the most visible part of this, but the allocation of resources was every bit as important.
Readers of the history of air warfare in World War II have probably heard of the control measures used by the RAF in the Battle of Britain to defend against German bomber attacks. Conversely, the Luftwaffe devised an even more sophisticated air defense network to oppose both the RAF Bomber Command and the US Eighth Air Force raids upon Germany. Radars were integrated with a communication system that linked fighter airfields, GCI stations, flak batteries and civil defense organizations.
When the US faced the specter of possible Soviet bomber attacks, the basic template of air defense was relatively well understood. But the absolute necessity of stopping every nuclear armed bomber, the vast area to be protected, and the increased speeds of modern aircraft meant that no manual system of managing the engagements would be feasible.
The Air Force turned to George Valley, and his proposed solution, computer assisted automation, lead to an integrated system that became known as SAGE, or the Semi-Automatic Ground Environment.
When we think of a GCI site, or air traffic controllers or whatever, we tend to think of a controller hunched over a radar scope with the circling bar of light and the blips of contacts glowing as the radar sweeps. But SAGE controllers weren’t looking at a radar scope. What they used was actually among the very first Graphic User Interfaces (GUI) ever tried for a computer system.
Remember, when SAGE was designed, the computer was in its infancy. It was rapidly becoming accepted as a critical tool for many tasks, but in those days, when you used a computer, you input your program, sat back an hour or two, and eventually the computer printed out the results, or spat out a stack of IBM punch cards. Pretty clearly, that wouldn’t be much use in an air defense system. The Air Force, partnered with IBM and MIT, took a chance that they could build a large enough computer to provide information processing in real time. Having processed the information, the computer also needed to present it in a way that was instantly useful.
Here’s roughly how it worked. Air Force surveillance radars (as well as FAA air traffic control radars) operated across the country, theoretically covering all the airspace over the US. The feed from these radars was converted to a digital format, and fed via modem over telephone lines to one of 22 regional SAGE air defense centers. At the SAGE center, the massive FSQ-7 computer converted the data to a graphic representation. It also kept track of each contact, assigning it a track number, and projecting its likely course. Prior to SAGE, radar controllers followed contacts by marking the screen with a grease pencil, then doing some math to figure out likely course and speed. SAGE did that math for the controller.
In addition to tracking any inbound raid, SAGE also evaluated which interceptor bases were best positioned to launch against a particular target, how many and what type of interceptors they had on alert, and could send the launch message to the interceptor base.
After launching the interceptor, SAGE also tracked the interceptor, via transponders on board. Knowing the speed, course and altitude of both the target and the interceptor, SAGE was able to quickly calculate the best courses and speeds for the interceptor to fly to attack the target. In fact, in later years, the SAGE could actually be coupled to the interceptor’s auto-pilot and steer it to intercept, hands off. The pilot, freed from having to hand fly the plane, was then free to concentrate on operating his own radar and weapon system.
Each of the 22 SAGE centers had up to 150 controllers, and each controller might be in charge of several intercepts simultaneously. The controller’s station, in addition to being one of the very first uses of a GUI also utilized a forefather of today’s computer mouse. A “light-gun” was used to point at and select a track the controller was interested in. And a sign of just how different things were back in the day, each controller station came equipped with an ashtray and cigarette lighter, standard!
The FSQ computer itself was something of a wonder. It was the largest compute built. How big? Well, each SAGE center had two complete computers, and took up about half an acre. Each computer had 55,000 vacuum tubes. And the system drew up to three megawatts of power.
A small part of an FSQ-7 computer at a SAGE site.
By having two computers at each site, there was a back up system ready in case one computer failed. But diligent maintenance, such as replacing tubes before they failed, meant that availability was surprisingly good, as high as 99%.
Construction on SAGE began in 1957, and was complete by 1963. Ironically, just as the system became fully operational, the threat of Soviet bomber attacks was diminished by the threat of Soviet ICBMs. SAGE was not equipped to track ICBM attacks. There was still some threat of Soviet bomber attack, and so SAGE remained in service- clear up to 1983! Further, it was the model for todays FAA air control system.
Not only did SAGE serve as a great leap forward in the use of graphical interface and real time computing, and what we now call “sensor fusion”, it inspired the SABRE airline ticket/reservation system still in use today.
Next up, we’ll take a look at some of the interceptors and weapons used in continental air defense.