SAGE A/N FSQ-7   NORAD Computer 

Can anyone tell us about 
this... is this really SAGE ?

Pete Karculias was an electronic technician/computer repairman for the Q-7 back in
1967-1969.  He was stationed at McChord AFB, WA. Thanks for sharing these  pictures with us Pete!

Introduction.

These are personal photos of the 25th Air Division HQ SAGE system.  They were taken 
in the spring of 1969 in the evening.  I could not photograph the operators on duty
as it would have interfered with their duties. Also, some areas were off limits due 
to security (ECM room, e.g.).

1. A_computer.jpg
Duplex Maintenance Console. This is the main control panel for computer operation.
The large box in the foreground is a line printer.

2. B_computer.jpg
The Duplex Maintenace Console for computer B. Part of the Long Range Inputs Console is 
in the left background. Magnetic tape drives can be seen in the right background.

3. Big_Mem.jpg
Magnetic Core Memory No. 1. This is a 256 x 256 array of 33-bit words (total 65,536 words).
Memory cycle time was about 3.25 microseconds. Actual operational time 6 microseconds.
Memory devices are iron ferrite cores.

4. CP_Console.jpg
This is the Air Division Commander's console in the Command Post of the Direction Center.

5. CP_Console_Close.jpg
Close-up of the Digital Display of the Commander's console.  The command post consoles did
not have the large situation displays found elsewhere.  They used a large projection system
to display the air situation on a central screen (this was not used except in an alert and so
was not on when the photos were taken).  The process used was to snap periodic photos of 
the situation display using a special blue-light console, develop the film and project it. 
So the projection display was always behind real-time by the film processing time (< 1 minute).

6. DC_Command_Post.jpg
This is the Command Post in the heart of the Direction Center.

7. Display_routing.jpg
Actually this is the Manual Inputs Frame. This is the variable wiring side.  Wires 
can be relocated to move signals to and from different consoles.  This provides system
reconfiguration capability.

8. Display_routing_2.jpg
This is the "permanent" side of the Manual Inputs Frame.  Signals from various system
components arrive on this side and are routed to their destination on the other (variable) 
side.

9. Drum_1.jpg
A magnetic drum assembly shown in position for maintenance access.

10. Drum_2.jpg
Another view of a magnetic drum in maintenance position. Technical information I recall: Each
computer had 12 drums. Six were for the display system and six for central computer use. Each
drum had fixed position read/write heads - at least 33 plus some spares. A word was 33-bits so 
you could read a word by reading all heads at the same time. The system had to deal with 
addressing by computing the angular displacement and timing the operation so the right data was
under the heads.  I don't recall the capacity but it was on the order of K words! The Central 
Computer stored it's system subroutines on drums and loaded them when needed.

11. Its_Crashed.jpg
This is for fun.  We kept a Polaroid camera in the operator's room to snapshot the console if
a computer crahsed. We could reboot and analyze the status later.  I am on the right. The
computer has just crashed with a memory parity error.

12. Large_frame.jpg
This is the Situation Display Generating Element (SDGE).  It processes information stored on 
the magnetic drums to generate the Situation Displays (SD) for over 100 SD consoles.

13. Lil_Mem.jpg
This is core memory No. 2. It has a 64 x 64 array of 33-bit words for 4096 total.  It has
the same operating speeds as Memory No. 1. The real reason there were two cores was to allow 
one memory to run diagnostic software on the other. 

There was also something called Test Memory.  This was an early equivalent to ROM.  There was
a patch panel that could be installed in the maintenance console.  It was hard-wired to logic 0
and 1 levels and was the first 32 words of the address space.  We kept a bootstrap loader
in it that would reload the system from tape. We also had different panels for maintenance
programs.  You can see three of them in the B-Computer.jpg photo. Just to the left of them
is the pull-out slot they go into.

14. Lil_Mem_Close.jpg
Close up of core memory 2 showing the X and Y plane drivers and individual iron ferrite cores.

15. Long_View.jpg
This shot shows the wall-to-wall distance of the area occupied by one computer.  Each computer 
occupied approximately a square of this size.

16. LRI_frame.jpg
This is the "tube-side" of the Long Range Inputs Frame.

17. Mag_Drum_Frame
This is a Magentic Drum Storage Frame.  It housed six drums and their control circuitry. There 
were two such frames for each computer.

18. Op_Console.jpg
This is the computer operator's console (left).  This is in the same room as the Maintenance 
Consoles. It includes a Situation Display (SD) a 19 inch CRT, a Digital Display (DD), a 5 inch
CRT.  Also there is a push-button panel which controls SD and DD information.

On the right are the Duplex Switching Consoles.  These provide switching functions to make
either A or B computer "Active".  One computer was always active and the other standby.  If
the standby computer was running the standby Air Defense program (rather than maintenance) the
system was in "Full Duplex" mode.  If the Active crashed, switchover could be done in a 
couple of seconds.  In Full Duplex Mode, the Standby computer received periodic data transfers 
from the Active so it was "ready to go" if needed.

19. Pwr_Control.jpg
This is a panel on the power control system.  The power plant (we had our own) generated a million
watts for the computers.  We also had our own air-conditioning system.  Chilled air was pumped
through the consoles at the bottom and out the top.  The exhaust air was warmed by the frames
and was recycled to heat the building in cool weather.

If air conditioning failed, an extremely loud klaxon over the operator's station (you can see 
it to the right on the wall in Op_Console.jpg) would go off.  The operator immediately called
the power house.  If AC could not be restored in 60 seconds, an emergency power down  was done.
The reason was that the magnetic core memories would begin to melt down if left without cooling
air for much longer than that. It went off once while I was on duty.  It turned out to be a
false alarm though.

20 SID_Close.jpg
Close-up of a Situation Display (SD) console.  You can see the outline of Washington, Vancouver
Island and British Columbia, Canada to the North.  The 25th Air Division covered all this 
territory.  Each Direction Center (DC) overlapped coverage with the adjacent DCs. The DC to the
South was located in California.  When it was shut down, parts of the machine were sold to a
Hollywood movie company.  For many years later, we saw them in movies and TV shows, wired up
to make the lights blink and flash.  Most notable were the Voyage to the Bottom of the Sea
(the TV series) and the "Kellog Computer" commercial which spewed cereal boxes out of the
test memory door!

21. SID_Console.jpg
This is longer shot of the SD console.

22. Simplex.jpg
This console provided operation and maintenance of the Long Range Inputs and Outputs. Simplex
because there was no redundant hardware.  Each radar station fed digitized data to the DC over
public telephone lines (a first - they had to invent the modem!). The DC also sent data to the
neighboring DCs and to the Command Center (the AN-FSQ-8 computer). 

23. Status_Board.jpg
This is located in the command post.  It provided Air Division Status data to the commanders.
It was updated by the WW II method of guys writing backwards on the back of the Plexiglas.

Two Tags off 'Clyde' the AN/FSQ-7 Outside Luke AFB Given to me back in the 80's
but... I just found them while digging though Paperwork left from the old days at
Computer Exchange Inc.  - -  Ed Sharpe

Sage Logistics Handbook  from March 1960. This copy belonged to  26AD At Luke AFB.
Had another binder stuck to it. Shall try to clean it up a bit. This would be a great 
candidate for a PDF file.

Nice general NORAD Brochure

26AD Cap, Hard to  do this  in a scanner folks!

                                                                                    October 15, 1997

Bunny White

Archivist – Lucent Technology

600 Mt. Avenue

Rm 3A-302

Murray Hill, NJ  - 07974

Bunny,

Attached is a map showing the principal installations of the SAGE System  (Semi-Automatic Ground Environment ) a USAF/RCAF project that comprised much of NORAD’s air defense of the US and southern Canada back in the late 50’s through the early 80’s. Interceptor air bases of the Air Force and Navy are not all shown.

The NIKE anti-aircraft systems, Ajax and Hercules, (the Army’s systems designed and built by Bell Labs and Western Electric) was integrated with this system. NIKE locations are not shown.

This system was significant in many ways:

1.       It was the first big integrated communications network. High speed data ran at 1300 BPS on the 1A Digital Signaling System between the worlds largest vacuum tube computers (AN/FSQ 7s and 8s) which were duplexed for reliability at 23 direction centers (sectors). The sectors were fed radar data from apx. 250 radar sites, Texas Towers, picket ships, airborne radar, and weather stations. The 23 sectors fed summarized track data to seven regional combat centers, in turn to the NORAD center at Colorado Springs.

2.       It gave the computer industry a big boost. The IBM 7090 was essentially a solid state version of the AN/FSQ 7/8 which was also converted. Burroughs also had a pre-processor at the radar sites.

3.       The first ground control of interceptor aircraft was carried out by this system. Ground to air radio provided digital vectoring of the aircraft to the pilot via a fire control CRT on the aircraft.  A wide variety of jet aircraft and air to air missiles were used.

4.       Digital Display Technology was promoted. Light pens, random access digital display CRTs were developed by Hazeltine and IBM and used by the thousands in this system.

5.       Simulation of hostile aircraft was developed by WE during the implementation of the system.  This was done to cut down the use of large aircraft as targets during training exercises, and system integration testing of ground to air radio and interceptors.

      (Live aircraft were used in the final system tests but not actually shot down.)

6.       The first in line microfilm fast processed 35 mm projection displays were incorporated.  These preceded the printer plotters that followed in the 60s and 70s and were able to project a large screen display within 30 seconds of the computer generated display.

7.        WE was able to obtain a large number of computer trained personnel.  At a time when very few computer trained people existed, close to 500 engineers and technicians were trained at MIT’s Lincoln Labs in digital techniques, computer programming and radar and other systems integration areas. These people often formed the cadres of computer personnel at many locations and other projects throughout the company when the SAGE implementation phase ended in the 1962-3 period.

Although the SAGE system design was led by the MITRE Corporation, Western Electric’s Air Defense Engineering Service (ADES), headquartered on Church Street NYC, was awarded the systems integration contract to coordinate the installation, and perform the subsystems and systems tests. (The good work on the DEW Line around the Arctic Circle helped getting the award.) It checked out hardware and software of many subcontractors (IBM, Borroughs, RAND/SDC, GE, Philco, Raytheon and others including  AT&T )

WE hiring started in 1955 and hit full stride in 1957 with training started at Murphy Army Hospital in Boston and then moved to MIT Lincoln Labs, Lexington MA. Class groups of about 15 started about every three weeks. Training lasted nine months for the first three years, afterwards which, with trained people in the field, it was cut to six. Students were paid a base salary but not paid expenses until they went to the field and were expected to find their own housing with many going to boarding houses and apartments in Cambridge.

Upon graduation they were assigned to one of five field test teams that would move about every nine months if the sector tests went well. The first five sectors to be checked out were on the East Coast.  Test teams were comprised of about 50 to 65 persons; half at the direction center, half at the radar and ground to air radio sites. At the end of a sector about ten persons were left behind to handle retrofit changes and upgrades. The others would be split into two groups with half matching up with half of another group at a new sector; the other half with another at another new location. Retrofit personnel losses were replaced with “rookies” just out of school. (The old men were those with two or three year’s experience.) It made for close friendships between couples that shared the hardships of the road, moving and looking for a new place every nine months.. Essentially your company work mates and their spouses were your family. Those friendships lasted for many decades after the project winded down.

The first sector (New York Air Defense Sector) was cut over in late 1958. The entire 23 SAGE sector and 7 combat center job was completed in 1962 (on time) with smaller scale retrofits taking place for a few more years. A few of the experienced people went to Bell Labs Whippany where they worked on a series of very involved tests of SAGE and NIKE Hercules air defense effectiveness in 1961-62. An experimental SAGE system was maintained in Montgomery, AL for testing of new sub-systems through the 60’s.

In the early 60’s, some of the personnel of WE ADES were phased into Bell labs/WE projects such as MAR-I, NIKE X/Safeguard, Underwater Defense, and ESS. Some went to WE factory or service locations, Bell operating companies and AT&T. Many, longing to be closer to home, now with children, or seeking another venue went with NASA, Jet Propulsion Labs, or other companies.

SAGE was a great project for the country, the company, and most of the people who worked on it. For many it was a way of learning a new field, getting to see the country, and make a few dollars in the process, - a great adventure which led to many good things.

Ed,

You have a wealth of info on your site. It was a long time ago and my memory
is poor. I'm not sure I could contribute anything you don't already have. I
have no physical mementos either. However, I may try to recall what I can.
Some points (thinking as I write):

1. Vacuum tubes didn't have a very long life. With all those tubes in the
computer (I Don't remember how many), the mean time to failure was a few
minutes. Solution: Marginal testing where voltages would be systematically
lowered and tubes that were on the verge of failing would be pushed to fail
during testing. Diagnostic marginal testing software automated the process
and used the computer to pinpoint its own failures.

2. The computers used a lot of power. I heard a rumor once that while
testing the system in Newburg, NY, a computer was hooked up with the wrong
phasing and it blew line circuit breakers all the way to Pennsylvania.

3. Each site had two computers (A and B). At any time one was Active and the
other was Standby, periodically switching the roles of the A and B
computers. We performed diagnostic testing and maintenance on the Standby
computer.

4. At the time I was there, IBM had a maintenance force of about 60 people
at each site for round the clock maintenance. They were divided into 3
specialties: Central Computer, I/O, and Display.

Perhaps I can think of other stuff, too...

Tom

The Last of the SAGE Techs

   I arrived at McChord AFB in Tacoma, WA, in January of 1983, after finishing tech school at Keesler AFB in Biloxi, MS.  At Keesler, the systems we trained on seemed a bit behind the times, including the BUIC (BackUp Interceptor Control), based on the Burroughs D825.  I figured though, that once I reached my permanent duty assignment, I'd get to work on some more modern equipment.  When I received my assignment, I was quite distressed to discover that I'd be working on the AN/FSQ-7 and its 50,000 vacuum tubes, installed in 1958!  McChords' Q7 was nearing the end of its service life, and I found myself in the last group of technicians to be trained and qualified on the Q7.  Things weren't as bad as I anticipated, though, and I was eventually selected for one of the four remaining maintenance slots on the new system, the AN/FYQ-93 (ROCC).  I spent 8 months working on the Q7, before heading for Tyndall AFB in Panama City, FL for training on the new system.

   Working on the Q7 turned out to be an interesting experience.  A computer big enough to walk around in!  I trained under TSgt Larry White, in Central Computer (CC).  We were short handed on the Q7 because most of the techs had been moved over to the new system (ROCC), but we still had to keep the Q7 running.  I completed OJT training on the Q7 in record time, and was soon working shifts as a qualified CC troop.  The final part of Larry's training program involved pulling a random tube from the standby system, and having me deduce its location using the Duplex Maintenance Console (DMC) and available diagnostic procedures.

   Some of the maintenance procedures were particularly memorable, such as changing heads on the drums.  Because it took in excess of 20 minutes for a drum to spin down, you couldn't often afford the down time and you changed the head while the drum was turning at speed.  It was a nerve wracking process which involved using an oscilloscope to monitor the signal amplitude on the head.  You cranked the head down towards the drum surface until the signal reach the proper level.  If the head went too far and actually touched the surface of the drum, it would score the surface and make that track unusable.  At least one of our female techs refused to change heads on a live drum, after  qualification.

   Running "margins" on the system meant using the Marginal Checking System to help detect system components, especially tubes, which were weakening and about to fail.  Voltages were varied by small amounts to specific parts of the system, while running diagnostic programs to check for errors.  It was also used to keep the core memory systems in adjustment.  I still have a printout which lists the ideal readings for margins on Big Mem.  Adjusting Big Mem meant using a flat blade screwdriver to make minuscule adjustments to a pot with a quarter inch shaft.  A tiny tap was all it took to bring it in or out of adjustment.

   One of the annual maintenance procedures involved powering down one of the duplexed systems.  For the annual PMIs (Preventive Maintenance Inspection), all of the available techs would be called in to help with this all day project.  One of the jobs was inspecting the power supply units for defective selenium rectifiers and electrolytic capacitors, and replacing them.  The capacitors were installed in "cap trays", with metal bus bars connecting the capacitors, and a neon indicator on the front which would indicate a problem with that tray.  With the power off, you had to discharge the cap trays with a special tool that looked like a gigantic, two pronged plug.  There was a lot of power stored in those units, and I never liked working around them.  Bringing the system back up after a power down was an adventure, since something would invariably fail when power was applied.

   On the lighter side, programmers over the years had written some interesting "diagnostic" programs for the system.  One was the Hula Girl, which was an animated drawing of a hula girl on the display scope.  If you pointed the light gun at her navel and pulled the trigger, her skirt would fall off!  Another was an interactive game of baseball, played on the display scope by two players.  A baseball diamond was drawn, with some crude base runners.  One player would press a button on the scope to pitch the ball, and the other would press a button on a cord attached to the scope to swing.  If the batter connected, the men would advance around the bases.  One night, after the PMIs were completed, two of our techs were playing baseball on the standby system's maintenance scope.  The red phone rang, which was connected to the Senior Director (SD) upstairs in the display room.  Expecting a problem, one of the techs answered the phone, to hear the SD ask, "Who's winning?"  Apparently, one of the display operators had switched over to the standby system and saw the game in progress.  Fortunately, the SD had a sense of humor.

   There were also programs written to play music on the Q7.  The DMC had a speaker on it, with a volume control and a switch to select one of four register bits to monitor.  During normal air defense operation, the speaker was usually turned to a low level to monitor the operation of the system.  The air defense program had a certain rhythm to it while cycling normally, and an experienced tech could often catch a problem early by hearing a change in the pattern.  In standby mode, by toggling the register bit on and off at a defined rate, you could produce definite musical pitches through the speaker.  The problem with the existing music programs was that it was a complicated process to write a program to play a tune, and each program could only play one tune.  Over the course of several mid-shifts, I wrote a music interpreter, which would read a short data deck (punch cards) telling which notes to play and for how long, making it easy to program new tunes.  After one productive mid-shift, the day crew arrived to hear the standby system playing "The Liberty Bell" march, with big block letters on the display scope spelling out "Monty Python's Flying Circus".  I wish I had saved that program!

   When McChord's Q7 was dismantled in 1983 after 25 years of operation, the crew saved a couple of pieces for me: a switch/indicator assembly from the DMC and a 64Kbit core plane from Big Mem.  I've also acquired a few other parts over the years, including two PUs, some panel faceplates and a LRI test panel assembly.  I have a page dedicated to my SAGE artifacts, here:

http://ripsaw.cac.psu.edu/~mloewen/Oldtech/SAGE/

   Most people don't realize it, but if they've watched any number of Science Fiction movies and TV shows, they've probably seen parts of the SAGE system.  From about 1964 on, parts of the system were used as props in many productions, including Lost In Space, The Time Tunnel, The Man From U.N.C.L.E. and Battlestar Galactica, to name a few.  Irwin Allen in particular was fond of the hundreds of blinking lights and switches, and many of his productions used Q7 parts as props.  I've been a Science Fiction fan since I could read, and used to watch all the Irwin Allen shows.  It wasn't until after I left the Air Force that I realized that the "computers" in the Time Tunnel complex were actually sections of the DMC and SMC from the Q7.  Some of these props have been used as recently as 2006, in the ABC production, Lost.  I've created another web site with a list of productions with Q7 props, along with screen captures of the equipment:

http://ripsaw.cac.psu.edu/~mloewen/Q7/

Mike Loewen

USAF 1982-1986

mcl8@psu.edu

We were constant having "unknowns" in our sector NYADS. This was usually
the Navy with their carrier ops. Sometimes they would let us know most
of the time not. We used to have to track them down to find out. One day
we had an unknown pop up out of sector center proceeding straight North
at mach 3. We scrambled the interceptors to find out what was going on.
and when they got there, no one was seen. So they turned around an came
back home. We checked with the site programming folks. System
Development Corporation, and told them what had happened. One of the
programmers when chasing through the code and came to the conclusion
that the only way that might have happened was if we had an instruction
failure on the B register. He proceeded to write a 1 card diagnostic
which exercised the B register when run and sure enough it failed. which
when on to prove that things  weren't  always  as they  seemed.


Another time a USAF LTC came down from North  Bay , Canada. He was
attached to the Atomic Energy Commission.  He was on a fact finding
mission. I met with him and  the programming site manager. He was 
trying to find out if the maintenance folks could launch Bomarc
missiles. I said yes. And I  proceeded to tell him how. The Senior
Director had a keyed console. The maintenance  people had a console
which had a plug board to make it look like any other console in the
building  so we could diagnose  problems  ahead of going to fix them.
that console DID NOT have a key. Going through the process of plugging
up our console to look like Air Surveillance, Identification, And Weapons
we could have launched a Bomarc. The LTC asked the site manager if that
was right and he said yes. That was not the answer he wanted to hear. I
guess maybe I was the only one who figured it out. I seems we were not
on the Human Reliability Program. No one had checked the maintenance
people to find out if any one had a screw loose. I don't know what the
ever did because I separated shortly after that.

On of the features of the AN/FSQ-7 was a speaker attached to a bit in
the accumulator. We listened to the program cycle and after a day or so
you consciously forgot about the noise but unconsciously you knew it was
there. You could be doing work, but the minute the sound changed. you'd
run to see what was up. That would be the time we'd switch over to the
standby machine


When I first got to the site my chief NCO gave me a tour of the second
floor and the computer room.  One of the things in the computer room was
a Polaroid camera on a roll around stand. I asked what it was for and he
told me it was a piece of test equipment. I said you have to be kidding.
He said no. He said we used it to take pictures of the console lights.
It helps us to trouble shoot problems.

John P DeWerth
CPT USAF

Airman First Class Trent Bailey, Weapons Controller Technician at 26th NORAD Region/Air Division

(click SAGE artwork to open in a separate large window)

"Airman First Class Trent Bailey, Weapons Controller Technician at 26th NORAD Region/Air Division points out a target for Airman Toni Bailey, a Air Surveillance Operator new to SAGE Operations, who is also his spouse." Picture was taken on February 12th, 1975 for an article in the Division's newsletter.  As I recall, there were 3 couples that worked together in the 26th Air Division and two of them were stationed at the NORAD Region Control Center at Luke AFB, AZ.

Trent Bailey tells us "The 26AD crest is actually printed on adhesive-backed paper and was used to mark manuals, books, etc.    I remember a sheet of them had a huge one, like 8X10, then smaller ones, down to an inch high.    This is an 3" original from there, and I've had it since my days there.   Yellowed with age, it is stuck to a mirror on my workstation at home."

AIR NAVIGATION ATTACK COMPUTER TYPE CPU-73A/P

Trent Bailey tells us: "Ed, it was a mechanical Attack Computer that enabled Weapons Controllers to calculate intercept vectors for engaging enemy aircraft. 
 
When the SAGE Direction Center determines that a 'track' (or computer-generated radar return) cannot be positively identified as a friendly aircraft, interceptor aircraft are scrambled.  In 26AD, this usually meant that something was coming toward the west coast of the United States from the Pacific Ocean.    While these tracks are usually off-course airliners or private aircraft, they are within an ADIZ (Air Defense Identification Zone) and, as such, require positive identification.   
 
As interceptor aircraft kept on Cold War alert take off, usually from active or Air National Guard bases in California, a team consisting of a Weapons Controller (at this time, early 70s, WCs were always commissioned officers) and a Weapons Controller Technician, usually an enlisted airman, we being assigned to the intercept by the Weapons Director on duty at the SAGE Direction Center.
 
Usually the WC Tech starts the ball rolling by contacting the appropriate FAA Control Center to coordinate the 'handoff' of the interceptor, now flying through civilian airspace under ATC control, to the control of the Weapons Controller.  
 
As the incoming track has been entered into the system, the Controller Team has a few facts about it, it's location, it's heading and it's altitude.  If the Unknown, as it is logged, has a IFF/SIF (Identification Friend or Foe/Selective Interrogation Feature) transponder on board, it is either 'squawking', or transmitting, an invalid code, or is silent.    
 
As the WC Tech finishes the handoff from Center, the Controller has been using his manual Attack Computer (similar in form and function to a civilian Flight Computer) to determine the best heading to assign the interceptor, based upon several things, the speed and heading of the Unknown among them.    He decides upon one of four attack options;
 
* FSR/A -  This is an initial attack from in front of the Unknown that continues into a re-attack from his rear.  This is being done with a significant speed advantage on the part of the interceptor, and is usually the most effective attack to use, if conditions allow.
 
* FSR/D - This is also an initial attack from the front, followed by a stern re-attack, but here, the interceptor has a speed disadvantage, or is on a par with the Unknown.  This changes the attack geometry itself, to enable the attack, so is classed as a different form of attack.    
 
* BEAM - A Beam attack is, as the term might suggest, a attack from the side, or 'abeam' of the Unknown.  These are shorter-ranged attacks, and are carried out with 'secondary' weapons, such as the FFAR/HVAR (Folding Fin Aerial Rocket/High Velocity Aerial Rocket, an unguided munition) or guns.  As I recall, both the F-102 and the
 F-106 had the capability to carry up to 24 FFARs within their missile bay doors.   
 
* PURSUIT - This is both self-explanatory, and the simplest attack.  Just point the Interceptor at the target and go. 
 
Now, while I have characterized this whole exercise as a defensive attack upon an enemy, of course we don't go shooting down innocent civilian aircraft without a really good reason and a lot of authorizations.  However, these intercept profiles are still run, just in case this Unknown turns out to be an actual Hostile aircraft."

For BUIC, The SAGE Backup Please see the BUIC section under Burroughs Click ---HERE 
 www.smecc.org/burroughs_buic_-__an_gsa-51__sage_backup.htm