NORAD: guardians of North America

© Mark Ollig 

US President Franklin D. Roosevelt and Canadian Prime Minister William Lyon Mackenzie King signed the Ogdensburg Agreement, a mutual defense pact, Aug. 17, 1940.

This agreement laid the groundwork for a continental security plan, ultimately leading to the establishment of the North American Air Defense Command (NORAD) May 12, 1958.

Formed amid escalating Cold War tensions, NORAD’s primary mission was to monitor and defend North American airspace against potential threats from the Soviet Union.

The initial headquarters for NORAD was Ent Air Force Base in Colorado Springs, CO, with its future headquarters inside a mountain.

Excavation within Cheyenne Mountain, southwest of Colorado Springs, to house the NORAD Combat Operations Center began May 18, 1961, and involved removing 693,000 tons of granite.

The NORAD Combat Operations Center would be located approximately 2,000 feet beneath the surface within the Cheyenne Mountain Complex (CMC), nestled deep within the granite of Cheyenne Mountain.

This strategic location and the complex’s fortified construction were designed to protect against a range of threats, including nuclear blasts, electromagnetic pulses, and biological attacks.

The CMC’s Faraday cage design protects its electronic systems from electromagnetic pulses (EMP) caused by high-altitude nuclear detonations.

Two enormous blast doors, each weighing 25 tons and measuring 3.5 feet thick, secure the main tunnel entrance to the Cheyenne Mountain Complex.

The main tunnel blast doors, constructed of reinforced steel and concrete (and likely other classified materials), are reportedly designed to withstand the blast, heat, pressure, and radiation of a 30-megaton nuclear detonation 1.2 miles away.

Deeper inside the tunnel, a second blast door acts as an airlock system to protect the complex’s interior, as recreated in the 1983 movie “WarGames.”

The blast doors of the Cheyenne Mountain Complex can be sealed in approximately 45 seconds using hydraulic systems or by on-site personnel using a mechanical chain-driven system (which would take much longer).

It had been reported the blast doors to the CMC were closed Sept. 11, 2001, during the terrorist attacks.

The inner complex within Cheyenne Mountain comprises 15 interconnected buildings, each three stories tall, constructed with reinforced “battleship” steel for maximum durability.

These buildings are strategically positioned eighteen inches from the granite rock walls of the mountain’s interior and are anchored to 1,300 high-strength steel springs, each weighing 1,000 pounds.

This innovative spring system engineering solution serves as a giant shock absorber designed to protect the buildings and personnel inside from the devastating effects of earthquakes, explosions, and the shockwave of a nuclear detonation.

The CMC contains six large, interconnected chambers, often referred to as tunnels. Each chamber is three stories tall, 45 feet wide, 60.5 feet high, and 588 feet long.

Due to escalating Cold War tensions, the NORAD Combat Operations Center became partially operational at the Cheyenne Mountain Complex April 20, 1966, while construction was still ongoing.

One month later, the NORAD Attack Warning System became operational, with final construction completed on Feb. 6, 1967.

The CMC was completed in 1967 at a reported cost of $142.4 million, which is equivalent to around $1.29 billion today, not including additional expenses for equipment, complex communications, monitoring systems, and infrastructure.

The Cheyenne Mountain Complex is primarily powered by electricity supplied from Colorado Springs; however, it can operate independently even in the event of external disruptions.

Backup power includes six locomotive-sized backup diesel generators and a 510,000-gallon reserve of diesel fuel.

Additionally, a large bank of industrial-grade lead-acid batteries provides uninterrupted power for critical electronic systems.

This redundant power system allows the CMC to operate independently for extended periods, ensuring the continuity of military command and control systems.

The CMC uses artesian mountain springs and a 1.5-million-gallon reservoir for water, supporting drinking, cooling, and fire safety.

The Cheyenne Mountain Complex has an advanced air filtration system that can remove biological, chemical, and radiological agents, including radioactive fallout, ensuring a safe and breathable environment.

In 1966, NORAD relied on the semi-automatic ground environment (SAGE) computer system, developed in the late 1950s, to track and defend North American airspace.

By the mid-1970s, SAGE had reached its limitations, leading to the 427M computer system replacement program, which improved NORAD’s effectiveness.

The new system consisted of three segments: the communication system segment, the missile warning system/space computational center, and the NORAD computer system (NCS).

Honeywell, then headquartered in Minneapolis, was the primary contractor for the NCS component.

In 1981, NORAD’s acronym was redefined as North American Aerospace Defense Command, expanding its mission to include aerospace defense.

NORAD now tracks earth-orbiting satellites and space debris in addition to monitoring and defending North America airspace.

In 2006, NORAD relocated its primary operations to Peterson Space Force Base in Colorado Springs, joining US Northern Command (USNORTHCOM) and the predecessor of Space Delta 1 (formerly Space Command under the US Air Force).

Peterson Space Force Base serves as NORAD and USNORTHCOM’s primary headquarters, while the Cheyenne Mountain Complex is an alternate command center and training site for both organizations.

NORAD maintains a continuous presence at the Cheyenne Mountain Complex and can relocate its entire operations there in response to any significant national security threat.

NORAD celebrated its 66th anniversary May 12 as the guardians of North America.

NORAD Command Console (undated) 

The north portal entrance to the Cheyenne Mountain Complex

The Saturday morning emergency broadcast

© Mark Ollig 

At 8:33 a.m. (CST) Feb. 20, 1971, bells loudly dinged from teletype machines in radio and television newsrooms across the United States, printing out a single urgent message.

An emergency action notification (EAN) message triggered a nationwide alert and disrupted broadcasting.

The EAN originated from the National Warning Center, located within the North American Aerospace Defense Command (NORAD) at the Cheyenne Mountain Complex in Colorado.

But first, the backstory.

Wayland S. Eberhardt, a civilian teletype operator at the National Warning Center, prepared to transmit the weekly emergency broadcast system (EBS) routine test message for radio and television stations early Saturday morning, Feb. 20, 1971.

Stations used color-coded EAN cards with the EBS protocol to determine which message to broadcast based on the received EAN message and codeword.

The white card (EAN message number one) was used to alert broadcasters to an impending emergency, instruct them to interrupt regular programming, and inform their listeners to stay tuned for further instructions.

The red card (EAN message number two) was reserved for a confirmed attack on the US. It instructed stations to broadcast a warning message and direct listeners to seek shelter immediately.

Mr. Eberhardt had three punched paper tapes for the teletype machine in front of him: a routine test tape and two tapes for real emergencies, one instructing the use of EAN message number one and the other EAN message number two.

Unfortunately, instead of sending the routine test message, he accidentally transmitted the EAN message number one tape, along with its authenticator codeword “hatefulness,” to radio and television newsrooms nationwide.

EAN message number one, sent at 8:33 a.m. CST, was received by all Associated Press and United Press International teletype machines in radio and TV newsrooms across the country.

Cue the loudly dinging bells of the teletype machines.

“Hatefulness” matched the unique daily code word for Feb. 20. Thus, broadcasters considered the message an actual emergency alert.

Message number one read: “This is an emergency action notification (EAN) directed by the President. Normal broadcasting will cease immediately. All stations will broadcast EAN message number one.”

Several national radio stations and many across Minnesota went off the air, including KLFD in Litchfield, KBUN in Bemidji, WMFG in Hibbing, KTMF in New Prague, and KDOM in Windom.

However, radio station WCCO (830 AM), with its National Defense Emergency Authorization (NDEA), remained on the air to transmit official news and information.

Within minutes of receiving the teletype message, WCCO radio broadcast the following:

“This is an emergency action notification. All broadcast stations shall broadcast emergency action notification message number one white card. This station has interrupted its regular program at the request of the United States government to participate in the emergency broadcast system, serving the Minneapolis-St. Paul area,” announced WCCO radio broadcaster Howard Viken.

He continued, “During this period, many radio stations will remain on the air, broadcasting news and official information for areas assigned to them. This station will remain on the air and will serve the Twin Cities area. If you are not located in the Minneapolis-St. Paul area, you should now tune your ra . . .” Viken’s audio was suddenly cut off.

After six seconds of radio silence, another voice is heard, “They’ve run the incorrect tape, Howard. This is a [the speaker hesitates] mistake.”

Viken then says, “This is ah . . . disregard . . . disregard the last message. Disregard the last message.”

He pauses, seemingly at a loss for words, and then says, “Fourteen until nine, Twin City temperature is 24 degrees. The wind is north-northeast at nine miles per hour.”

“Well, we’re sorry for that error on the part of somebody, but this was to be an emergency action notification, but there’s been an error, so disregard that emergency action notification. Disregard that last announcement,” Viken explained.

Approximately 15 minutes after the initial emergency action notification, Viken reported, “We just received this from our CONELRAD advisory. All stations. CONELRAD advises that they have run the incorrect tape for today’s test. Please disregard the authenticator message which moved on this circuit a short time ago. This notice came from CONELRAD in Chicago.”

Howard Viken (1924 to 2021), who started working at WCCO radio in 1950, likely used the term “CONELRAD” because it was still informally used among broadcasters at that time.

The CONELRAD (control of electromagnetic radiation) system was a Cold War-era emergency communications protocol designed in 1951 to disrupt enemy navigation during an attack on the US and to broadcast official news and information to the public.

In 1963, it was replaced by the emergency broadcast system.

It required most radio stations to cease transmissions, except for designated CONELRAD stations broadcasting at 640 kHz and 1240 kHz.

Many radio models during that time had symbols printed on their dials, such as triangles, indicating these frequencies.

Despite multiple attempts to officially retract the erroneous emergency action notification, it persisted for 40 minutes until 9:13 a.m. CST, when the EBS transmitted the correct code word, “impish,” finally canceling the EAN.

You can listen to the archived WCCO radio broadcast from Feb. 20, 1971, with the EAN announcement starting 26 seconds into the audio file, along with news clips from that day, at tinyurl.com/1971WCCO.

Minnesota’s role as the computing heartland

© Mark Ollig 

In late 1945, J. Presper Eckert and John Mauchly completed the Electronic Numerical Integrator and Computer (ENIAC).

ENIAC is considered to be the first programmable general-purpose electronic computer.

Programming the ENIAC involved physically setting switches and plugging cables into a patch panel to configure the computer for specific calculations.

This electronic computer consists of 40 nine-foot-high cabinets and thousands of vacuum tubes, capacitors, switches, and relays.

In 1946, Eckert and Mauchly founded EMCC (Eckert–Mauchly Computer Corporation).

In 1950, Remington Rand acquired EMCC, which later became the Univac Division of Remington Rand.

In 1951, the UNIVAC I (Universal Automatic Computer I), designed primarily by Eckert and Mauchly, was delivered to the US Census Bureau.

Meanwhile, in 1946, a group of former WWII US Navy cryptographers, including William C. Norris and Howard T. Engstrom, along with businessman John E. Parker, founded Engineering Research Associates (ERA) in St. Paul.

ERA applied its knowledge of electronic decryption and numerical analysis to become a pioneering computer company, developing technologies such as numerical computers and drum memory systems.

William Shockley, John Bardeen, and Walter Brattain demonstrated the first working transistor at Bell Telephone Laboratories in Murray Hill, NJ, Dec. 23, 1947,

During the 1950s, transistors started to replace vacuum tubes, resulting in the creation of smaller, more efficient electronic devices and marking the beginning of a new era in high-performance computing.

The ERA 1101, a classified computer built by Engineering Research Associates for the US Navy and delivered in late 1950, marked a significant advancement in computing technology.

This once top-secret computing machine’s complexity and intricate arrangements were described in a Feb. 28, 1952, Minneapolis Morning Tribune newspaper article as a “labyrinth of tubes, wires, and ‘memory’ drums.”

The newspaper described how, like other early computers, sometimes referred to as “electronic brains,” the ERA 1101 could perform thousands of calculations and solve complex problems in seconds.

Remington Rand’s acquisition of Engineering Research Associates in 1952 had a significant impact on the computer industry.

It confirmed Minnesota as a key player in computing technology and furthered its emergence as a central hub for this industry.

Initially, ERA remained a separate division within Remington Rand at its St. Paul location, focused on scientific and military applications.

Through Remington Rand’s acquisition of ERA, it gained ownership of the ERA 1101 computer.

Remington Rand modified and rebranded the ERA 1101 as the UNIVAC 1101, a commercial computer that utilized and further popularized the stored-program concept, enabling electronic storage of computer programs in memory alongside data.

Unlike earlier computers, such as the ENIAC, electronically stored programs eliminated having to run patch cords and simplified programming and overall computing operations.

The UNIVAC 1101 paved the way for the widespread integration of computers into business settings.

The success of ERA, and later the UNIVAC division of Remington Rand, with the ERA/UNIVAC 1101, 1102, and 1103, firmly established Minnesota as a leading hub in the growing field of computing.

Sperry Corporation and Remington Rand merged June 30, 1955, to form Sperry Rand Corporation.

In late 1955 or early 1956, Sperry Rand’s St. Paul UNIVAC division (formerly ERA) began developing the UNIVAC Athena for the US Air Force’s missile guidance system.

The UNIVAC Athena, a computer with a 24-bit architecture, was used for accurate missile guidance by performing complex calculations with high precision, ensuring the real-time control of long-range missile trajectories.

Seymour Cray, a University of Minnesota electrical engineering graduate and ERA employee since 1950, was an influential contributor to developing the UNIVAC Athena missile guidance computer, later adapted for the Titan I ICBM.

In 1957, Cray and William Norris founded Control Data Corporation (CDC) in St. Paul.

By 1959, the Sperry Rand St. Paul Univac division had completed and delivered the first UNIVAC Athena computer.

In 1960, CDC opened its new industrial park in Bloomington and hired 1,000 new employees.

By the mid-1960s, a total of 23 UNIVAC Athena computers were produced and successfully deployed for military use.

In the early 1960s, Seymour Cray and Minnesota native James Thornton collaborated on the initial design of the CDC 6600 supercomputer at Control Data Corporation (CDC) in St. Paul, MN.

Seeking a quieter environment, Cray returned to his native roots in Chippewa Falls, WI, where he established a new CDC laboratory. This was where his team continued its work, completing the CDC 6600 project in 1964.

The CDC 6600 featured a central processing unit with a clock speed of 10 MHz.

Its architecture included multiple peripheral processors, enabling it to achieve a peak performance of about three million instructions per second (MIPS), making it the fastest computer in the world at that time.

The computer had a keyboard and a dual-screen console, and it used a Freon cooling system to manage heat.

The CDC 6600 maintained its position as the world’s fastest computer until 1969, when its successor, the CDC 7600, surpassed it.

Did the Minnesota heartland play a significant role as a leader in computing?

You betcha we did.

CDC 6600, the first supercomputer

(CDC 6600 by Jud McCranie is licensed under CC BY 2.0.)

The IBM 1410 computer

© Mark Ollig 

The IBM 1410, a room-sized computer priced at $700,000 (equivalent to $7.4M in 2024), was unveiled Sept. 12, 1960.

The IBM 1410 is classified as a second-generation computer.

Second-generation computers marked a major technological leap by replacing vacuum tubes with transistors, improving computer processing, maintenance, and reliability.

Ferrites, ceramic compounds made mainly of iron oxide combined with other metal oxides, are known for their magnetic properties.

IBM 1410 computers used tiny, donut-shaped magnetic ferrite cores for memory storage.

These cores possess a property called hysteresis, meaning they “remember” their previous magnetic state even after the external magnetic field is removed. This allowed the computer to retain stored data even when powered off.

Each ferrite core stored one bit of data, represented as a binary one or zero, based on its direction of magnetization.

A standard modular system used with the IBM 1410 was designed with plug-in circuit boards containing transistors, resistors, capacitors, diodes, and other components.

The computer used a crystal-controlled clock circuit to ensure accurate sequencing of processing operations.

The IBM 1410 computer system relied on the IBM 1415 console, featuring a read-write typewriter and printer, for its primary input and output.

Punched cards, magnetic tape drives, and the integrated IBM Selectric typewriter all served as data entry and retrieval methods.

Magnetic tapes offered considerable storage capacity for the time and could be easily removed, filed, and used on other compatible IBM systems.

The computer’s control console had an attached CE (customer engineer) module for troubleshooting, diagnostics, maintenance, and testing of the computer system.

There was also an indicator display panel with individually small, lighted lamps showing data processing status and computing system operational tasks.

IBM pioneered the development of the RAMAC (Random Access Method of Accounting and Control) data disk storage system in the 1950s.

This system, designed to store and manage substantial amounts of data, represented a significant advancement in storage technology.

The IBM 1410 computer later utilized the RAMAC system by incorporating IBM 1405 disk storage units.

These units, introduced in 1961, expanded the computer’s storage capacity to manage larger amounts of data.

Each IBM 1405 stored computer data on vertically stacked, cylindrical platters coated with magnetic material.

These platters rotated on a shaft at 1,200 rpm, and the IBM 1410 used automated read and write heads to access data.

A single disk had 40 platters divided into 10 disk packs, providing a total storage capacity of up to 10 million characters (equivalent to just 0.01 GB on a modern personal computer).

These disk storage units’ data access times were in the order of milliseconds.

However, it pales in comparison to today’s DDR4 RAM (Double Data Rate, Fourth Generation Synchronous Dynamic Random-Access Memory), with its near-instantaneous data access speeds ranging from 10 to 20 nanoseconds.

In terms of processing capabilities, the IBM 1410 could perform up to 600,000 additions, 60,000 multiplications, or 40,000 divisions per minute, and it had a processing speed of approximately 110,000 instructions per second.

To put this in perspective, a modern laptop with a 3-4 GHz processor can perform around three billion to four billion instructions per second.

While a direct comparison is difficult due to the different architectures and instruction sets, this comparison highlights the incredible advances in today’s computing power from that used over six decades ago.

The IBM 1410 supported various programming languages, including:

• FORTRAN (Formula Translation) was used for scientific and engineering computations.

• COBOL (Common Business-Oriented Language) was designed for business applications like payroll and record-keeping.

• Assembler Language provided precise, low-level hardware control.

• Autocoder, a simplified language, accelerated the conversion of source code written in one programming language to another programming language.

SPOOL (Simultaneous Peripheral Operations On-Line) is a buffering technique that temporarily stores data (such as reports or computation results) and sends it to a printer as soon as one becomes available.

By 1961, high-speed impact printers like the IBM 1403, capable of printing 600 lines of text per minute, were being used with the IBM 1410 system.

I need to mention Major Leland Fiegel, a Minnesota native from Rochester, and his top-secret mission during World War II, in which Thomas J. Watson Jr., the son of IBM’s founder, was his co-pilot.

They became close friends, but unfortunately, Fiegel died in a plane crash in 1948.

In 1956, IBM President Thomas J. Watson Jr. chose Rochester as the site for a new IBM manufacturing plant in honor of his late friend Leland Fiegel.

Construction of the new plant began July 31, 1956, at a reported cost of $8 million ($84,700,000 in 2024).

In 1988, IBM released the Application System/400 (AS/400), a new line of mid-range computers designed and manufactured at their Rochester facility.

The AS/400 sought to bridge the gap between larger mainframes and smaller personal computers.

In the 1990s, while working for USLink, a TDS Telecom subsidiary, I often interacted with various databases on their AS/400 computing system.

IBM ended production of their 1410 computer in March 1970.

The 1961 IBM 1410 data processing system as depicted in the reference manual.

A diagram of the IBM 1415 control console.

Blast into the past: Dr. Sbaitso

© Mark Ollig 

December 1991 saw the release of Dr. Sbaitso, a human-computer interaction program for MS-DOS computers.

Sbaitso is an acronym for “SoundBlaster Acting Intelligent Text-to-Speech Operator.”

Dr. Sbaitso, designed by Creative Labs (aka Creative Technology Ltd.), was a virtual therapist program.

It used a basic speech synthesis engine to engage users in simple conversations.

“I’m here to help you,” says Dr. Sbaitso in a deep, robotic, monotone voice.

Dr. Sbaitso was a digital voice program bundled with Creative Technology’s Sound Blaster Pro audio card.

The card used a digital-to-analog converter (DAC) chip to transform digital audio into analog sound that could be played through speakers or headphones.

Dr. Sbaitso’s audio contained pre-recorded diphones. These are small speech sections composed of two neighboring phonemes, the basic units of sound in language.

By stringing together these diphones, each representing a distinct sound, Dr. Sbaitso could generate synthesized voice responses that mimicked natural speech patterns. In this way, it produced realistic-sounding voice responses based on a PC user’s text input.

Dr. Sbaitso’s primary purpose, though, was to demonstrate and market the audio capabilities of the Sound Blaster Pro CT1350B sound card, which included ports for headphones, speakers, and microphones.

To generate diverse sounds, including musical instruments, the sound card employed the Yamaha YM3812 OPL2 chip with FM synthesis.

This chip’s ability to modify waveforms made it adaptable and cost-effective; it became popular among musicians and sound designers.

Additionally, the card featured a volume control and a Musical Instrument Digital Interface (MIDI), a widely used protocol for connecting musical instruments with computers.

Dr. Sbaitso relied on First Byte’s Monologue speech synthesis software, developed during the 80s and early 90s.

This software, constrained by the limited computing power of the era, relied on a system of pre-programmed linguistic rules to assemble understandable speech from text.

The program’s scripted responses and limited vocabulary often led to repetitive, humorous, and nonsensical phrases when it struggled to process complex sentences.

Dr. Sbaitso’s text-to-speech responses always used uppercase letters.

The following is a conversation between a PC user and Dr. Sbaitso. I typed Sbaitso’s responses in lowercase for easier reading:

PC user: Hello.

Dr. Sbaitso: Hello. I am Dr. Sbaitso. How can I be of assistance to you today?

PC user: I’m feeling a bit down.

Dr. Sbaitso: Why do you feel that way?

PC user: Well, I had a bad day at work.

Dr. Sbaitso: Tell me more about your work.

PC user: I’m a programmer, and my code keeps crashing.

Dr. Sbaitso: That sounds frustrating. Perhaps you need a break.

PC user: You think I should take a vacation?

Dr. Sbaitso: I do not have enough information to say.

PC user: This isn’t very helpful.

Dr. Sbaitso: Why do you feel that way?

PC user: Forget it. Maybe I’ll just play a game instead.

Dr. Sbaitso: Very well. Goodbye.

Despite being used in the early 1990s, Dr. Sbaitso sparked conversations about privacy and responsible data use that remain relevant today.

Currently, ChatGPT-4, a generative pre-trained transformer model, denotes its architecture and training processes within the area of natural language processing and artificial intelligence.

For this column, I asked the GPT-4 AI app: “How would you compare yourself with the Dr. Sbaitso software program from 1991.”

GPT-4: “Dr. Sbaitso was an early voice synthesis program that used simple pattern-matching techniques, primarily for entertainment and showcasing sound card capabilities.

In contrast, I am powered by GPT-4, a more advanced AI model that provides detailed, context-aware responses across a wide range of topics, aiming for meaningful and useful interactions rather than just entertainment.”

GPT-4 concluded, “This reflects significant advancements in AI and natural language processing over the years.”

To interact with the 1992 version 2.20 of the MS-DOS Dr. Sbaitso program emulator, visit the Internet Archives at archive.org/details/SBAITSO_CGA.

Once there, click on the round power button symbol displayed on the blue screen under the Creative Labs banner.

After a brief loading time, Dr. Sbaitso will appear and audibly greet you with: “Dr. Sbaitso, by Creative Labs. Please tell me your name.”

Type your name and press the “enter” key after the last letter of the name appears.

To start a conversation with Dr. Sbaitso, locate the “>” symbol and blinking command line on the left side of your screen.

Type a short message and hit “enter” to send it to the doctor.

You will now be engaged in conversation with Dr. Sbaitso.

For a list of commands, type “help” and press enter.

To end the conversation, type “quit” or close the browser tab.

While earlier chatbots existed in research settings, Dr. Sbaitso’s bundling with the Sound Blaster Pro brought this type of semi-artificial intelligence software to consumer-level computers.

After not using Dr. Sbaitso for more than 30 years, revisiting this forerunner to today’s AI chatbot technology was a nostalgic blast into the past.

My conversation with the MS-DOS
Dr. Sbaitso program emulator