ATLAS-I

ATLAS-I, also known as the Trestle, was the world's largest non-nuclear electromagnetic pulse (NNEMP) generator and testing facility, designed to simulate the effects of high-altitude nuclear detonations on military aircraft and electronic systems during the Cold War.^[1]^[2] Constructed at Kirtland Air Force Base in Albuquerque, New Mexico, it served as the Air Force Weapons Lab Transmission-Line Aircraft Simulator, enabling full-scale testing of strategic bombers like the B-52 and B-1B to ensure their resilience against electromagnetic pulses (EMP).^[1]^[2] The facility's design emphasized non-conductive materials to avoid interfering with EMP waves, making it the largest all-wooden structure ever built, comprising 6.5 million board feet of glue-laminated southern pine and Douglas fir timbers fastened with 60,000 wooden and fiberglass bolts.^[1] Spanning a 20-acre site with a 12-story platform, a 1,300-foot-long deck, and a 127-foot-tall pulse terminator tower, ATLAS-I resembled an elevated railroad trestle to accommodate aircraft placement for testing.^[1] Its construction, overseen by designers including Dr. Carl E. Baum of the Air Force Research Laboratory and contractors such as McDonnell-Douglas Aircraft Company, began in 1972 and was completed in 1980 at a cost exceeding $60 million.^[1]^[2] Technically, ATLAS-I utilized Marx generators to produce EMP simulations delivering up to 0.2 terawatts of power and 200 kilojoules of energy per pulse, replicating the intense, short-duration electromagnetic fields from nuclear explosions without actual radiation.^[2] Operational from 1980 to 1991, it played a critical role in U.S. defense strategy by hardening electronics against potential Soviet nuclear threats, contributing to advancements in EMP mitigation for aviation and satellite systems.^[1]^[2] Decommissioned in the early 1990s following the Cold War's end, the structure now stands abandoned, occasionally used for military training like rappelling, though it has not been designated a National Historic Landmark despite its engineering significance.^[2]

History and Development

Background and Purpose

During the Cold War, the United States harbored significant concerns regarding the electromagnetic pulse (EMP) effects from potential Soviet high-altitude nuclear detonations, which could disrupt or destroy electronic systems across wide geographic areas by inducing damaging voltages and currents in conductive materials.^[1] These fears were heightened by historical events like the 1962 Starfish Prime test, which demonstrated EMP's potential to affect power grids and communications far from the blast site, prompting military planners to prioritize defenses against such asymmetric threats to national infrastructure and strategic capabilities.^[3] To address these vulnerabilities, ATLAS-I was specifically designed to replicate non-nuclear EMP (NNEMP) effects on aircraft and associated systems, enabling the validation of radiation hardening techniques essential for maintaining operational integrity in contested electromagnetic environments.^[2] The facility focused on testing Strategic Air Command assets, including bombers like the B-52 and B-1B as well as missile systems, to ensure their electronics could withstand simulated EMP assaults comparable to those from a high-altitude nuclear burst.^[3] This non-nuclear approach allowed for repeatable, controlled experiments without the risks and international repercussions of actual nuclear testing.^[1] The project was initiated in 1972 by the Air Force Weapons Laboratory (now part of the Air Force Research Laboratory) at Kirtland Air Force Base in Albuquerque, New Mexico, as part of broader efforts to bolster U.S. nuclear survivability. The project was led by figures such as Dr. Carl E. Baum of the Air Force Weapons Laboratory.^[1] The total cost of the ATLAS-I development reached $60 million, equivalent to approximately $240 million in 2025 dollars when adjusted for inflation using Consumer Price Index data.^[3]^[4]

Construction and Timeline

The project for ATLAS-I was initiated in 1972 amid heightened Cold War concerns over electromagnetic pulse (EMP) vulnerabilities in military aircraft. Construction at Kirtland Air Force Base in New Mexico began following the 1973 contract award, with groundbreaking and site preparation occurring in July 1974 due to soil testing requirements.^[5] Major construction ramped up from 1975 to 1979, encompassing the erection of the wedge building by October 1975, completion of the 400-foot-long ramp by July 1977, and progressive assembly of the 115-foot-high test stand reaching 80% completion by July 1978.^[5] The facility achieved initial operational capability in February 1980, following the installation of pulser systems in December 1978 and final integration by January 1979, at a total cost exceeding $60 million.^[1]^[5] Engineering challenges were formidable, primarily stemming from the requirement for non-conductive materials to prevent interference with EMP fields during tests, necessitating the world's largest all-wooden structure at the time.^[1] Variable alluvial soils at the site demanded extensive geotechnical analysis, leading to adjustments in caisson diameters from 30 to 36 inches to ensure stability under heavy loads up to 550,000 pounds.^[5] Design iterations addressed wind loads up to 40 miles per hour and shifted from initial glued joints to bolted connections using dielectric fasteners for a 10-year lifespan, while minimizing electromagnetic perturbations through optimized placement of split-ring connectors.^[5] Cost overruns arose from inflation, underestimated lumber needs, and pulser development delays in achieving required frequencies.^[5] Key materials included 6.5 million board feet of glue-laminated timber, primarily Douglas fir for its superior tensile strength and EMP transparency, supplemented by Southern yellow pine for weather resistance, all pressure-treated with pentachlorophenol via the Cellon process.^[1]^[6] Structural elements comprised interlocking trusses with 12-by-12-inch columns up to 111 feet long, 52-foot deck planks, and over 150,000 Permali bolts—made of wood or fiberglass—along with 120,000 split rings and shear plates to secure joints without metal.^[5] Concrete caissons and minimal steel (690 tons) were confined to the ground plane wedge to avoid field distortion.^[5] The project engaged hundreds of workers through prime contractors such as McDonnell Douglas Astronautics Company for early phases and Allen M. Campbell Company for the test stand, ramp, and pulser supports, coordinated under Air Force Weapons Laboratory oversight.^[5] The overall footprint spanned approximately 600 feet in length and 120 feet in depth, including a 200-by-200-foot platform elevated 118 feet to simulate in-flight conditions, with over one million cubic yards of earth excavated for the bowl-shaped arroyo site.^[5] Component integration proceeded in phases to align with testing needs: the non-conductive wooden trestle and ramp were assembled first to support aircraft positioning, followed by erection of 185-foot dielectric towers for the 1,300-foot transmission line by mid-1978, and culminating in the placement of two 5-megavolt Marx generators in sulfur hexafluoride enclosures atop dedicated wooden stands.^[5] This sequential build ensured electromagnetic compatibility, with the resistive termination array on a 127-foot tower completing the pulse delivery system by late 1979.^[5]

Technical Design

EMP Generation System

The EMP Generation System of ATLAS-I utilized two Marx generators constructed by Maxwell Laboratories in San Diego, California, each designed to produce up to 5 MV of output voltage, which were combined to achieve a total differential output of 10 MV.^[7]^[8] These generators formed the core of the non-nuclear EMP simulation capability, enabling the replication of electromagnetic effects from high-altitude nuclear detonations without requiring actual explosions.^[9] The system generated pulses characterized by a peak power of 200 gigawatts, a rise time of approximately 10 nanoseconds, and a waveform specifically tailored to mimic the E1 component of a high-altitude electromagnetic pulse (HEMP), which features a rapid, high-frequency surge capable of inducing voltages in electronic systems.^[7]^[8] Operationally, the Marx generators functioned by charging high-voltage capacitors in parallel to store energy, then rapidly discharging them in series through spark gaps to produce the ultra-short, high-energy pulse required for testing.^[8] This mechanism ensured precise control over the pulse delivery into the transmission lines feeding the Trestle Test Stand.^[9] Integrated diagnostics, including field sensors and voltage probes, allowed for real-time verification and calibration of each pulse to maintain fidelity to HEMP specifications, while safety features such as sulfur hexafluoride (SF₆) gas insulation in the generator enclosures prevented unintended arcing and protected personnel from high-field exposures.^[7]^[8]

Trestle Test Stand

The Trestle Test Stand, a key component of the ATLAS-I facility, was a massive elevated wooden platform designed to position aircraft for electromagnetic pulse (EMP) testing while isolating them from ground-induced electromagnetic interference.^[8] Measuring 200 feet by 200 feet and rising approximately 115 feet high—equivalent to about 10 stories—it provided a stable, non-conductive surface for simulating free-space flight conditions during EMP exposure.^[8] The structure's capacity supported aircraft weighing up to 550,000 pounds, including fully loaded B-52 bombers, ensuring it could handle the weight of strategic bombers under test.^[8] Constructed entirely from glue-laminated Douglas fir lumber without metal fasteners, the Trestle prevented electromagnetic shielding or reflection that could distort test results, as metal components would interfere with the EMP waves directed at the elevated test subjects.^[8] This all-wood design, using dielectric bolts and pressure-treated wood for durability against decay and insects, incorporated innovative structural elements such as massive glued-laminated beams and arches spanning up to 126 feet for enhanced stability.^[8] Elevated 115 feet above the ground in a natural arroyo bowl at Kirtland Air Force Base, the platform minimized soil conduction effects that might otherwise alter the EMP field uniformity.^[8] Accessibility to the test area was facilitated by a 400-foot-long, 50-foot-wide towpath ramp ascending to the platform height, allowing aircraft to be maneuvered into position without metallic support structures.^[8] However, the untreated wooden surfaces posed weatherproofing challenges, as the exposed design required operations to halt during thunderstorms or winds exceeding 35 knots to protect the integrity of the structure and tests.^[8] Overall, these features made the Trestle the world's largest wooden structure at the time, optimized for precise EMP vulnerability assessments on full-scale aircraft.^[8]

Support Facilities

The Wedge Building served as the primary control and support facility for ATLAS-I operations, located at the south end of the Trestle test stand. This steel structure measured 250 feet in length and reached a height of 240 feet, featuring a wire mesh enclosure that functioned as a Faraday cage to protect sensitive electronics from electromagnetic interference during tests.^[8] Internally, the building was organized across four levels accessible by elevator and two stairways, with the lower two levels (1 and 2) dedicated to walled offices, storage areas, and laboratories for personnel and administrative functions. The upper levels (3 and 4) provided semi-protected spaces for instrumentation, with Level 4 reserved exclusively for pulser operations and excluded from personnel access during EMP firings. Key components included EMP data acquisition systems, such as 12 analog fiber optic data channels supported by four Tektronix 7912 digitizers each, along with 32 analog fiber optic trunk lines offering a 250 MHz bandwidth and 20 dB dynamic range; these were processed by two DEC PDP-11/70 computers, enabling data quality checks in approximately three minutes per channel.^[8] Ancillary infrastructure encompassed robust power supply systems to support Marx generators rated at 5 MV output within SF₆-filled enclosures, extensive cabling conduits for transmission lines, and diagnostic equipment for real-time monitoring of test parameters. Integration with the Trestle was achieved through non-conductive pathways, including a towpath and ramp, which routed data via fiber optics to minimize interference while facilitating coordination with the adjacent test platform.^[8]

Operations and Applications

Testing Procedures

Testing procedures for ATLAS-I involved a series of standardized steps to simulate electromagnetic pulse (EMP) effects on aircraft and systems while ensuring safety and data integrity. Preparation began with the removal of aircraft fuel and the attachment of instrumentation at critical points, such as avionics and wiring harnesses, to monitor voltage and current responses. The test item, typically an aircraft, was then towed at approximately 3 mph along a 525-foot towpath onto the wooden trestle deck—a 200-foot by 200-foot platform elevated 115 feet above ground—positioned precisely within the working volume (a 248-foot diameter cylinder) using tiedown straps for stability.^[8] Once positioned, support systems like gaseous nitrogen and air conditioning were connected to maintain operational conditions simulating flight. The trestle structure, constructed from dielectric materials including glue-laminated lumber and fiberglass bolts, minimized electromagnetic interference to replicate in-flight isolation from ground effects. Final checks included verifying instrumentation integrity and test orientation, such as nose-to-tail or wingtip-to-wingtip alignments.^[8] Pulse delivery utilized the facility's dual Marx generators, each capable of producing approximately 5 megavolts and storing 210 kilojoules of energy, housed in SF6-filled enclosures to prevent arcing. The generators were charged for about 2 minutes before sequential or simultaneous firing, launching nanosecond-duration pulses into parallel-plate transmission lines via wire arrays. Multiple pulses were often delivered per test to evaluate repeatability, with the EMP propagating as a horizontally polarized wave across the test volume; the process produced an audible sound akin to a muffled rifle shot.^[8] Data collection relied on high-speed instrumentation, including 12 analog fiber optic channels with 250 MHz bandwidth and 20 dB dynamic range, to transmit transient responses from sensors embedded in the test item to shielded control rooms. High-speed oscilloscopes and two DEC PDP 11/70 computers recorded electrical parameters, focusing on thresholds for system upset, damage, or failure, with quality-checked data available within 3 minutes for up to four channels. Post-pulse analysis emphasized peak fields, risetimes, and energy coupling into critical components.^[8] Safety protocols prioritized remote operation from EMP-hardened areas, such as the control room in the facility's wedge structure, with personnel strictly excluded from high-field zones exceeding 100 kV/m for single pulses or 300 V/m for repetitive exposures. Warning signals, including audible alarms and flashing lights, preceded each firing, and testing was suspended for adverse weather like thunderstorms within 5 km or winds over 35 knots. A comprehensive fire suppression system, delivering 8,000 gallons per minute from a 500,000-gallon tank at 75-100 psi, protected the combustible wooden structure, complemented by post-test inspections for structural integrity and any induced hazards.^[8]

Key Tests and Outcomes

ATLAS-I primarily tested Strategic Air Command (SAC) aircraft to evaluate their susceptibility to high-altitude electromagnetic pulse (HEMP) effects, focusing on full-scale platforms such as B-52 bombers, E-3 AWACS, E-4B airborne command post, EC-135, and B-1 bombers. These tests involved positioning aircraft on the elevated Trestle platform and exposing them to simulated EMP fields generated by the facility's Marx generators, replicating nuclear-induced pulses up to 50-100 kV/m in strength. Early testing in 1980 on a B-52G revealed performance anomalies, such as unexpected electrical responses, underscoring the need for enhanced diagnostics in future evaluations.^[10] Additional subjects included the B-1 bomber, which demonstrated effective 'designed-hard' features with minimal added cost (less than 2%), and the E-4B airborne command post, recommended for proof-of-principle simulations using trailing wire antennas. Missiles and avionics subsystems were also subjected to scaled EMP exposures to assess component-level resilience, often in conjunction with aircraft integration tests.^[10]^[8] Key outcomes highlighted significant vulnerabilities in aircraft wiring and electronics, where induced currents could overload unshielded circuits and cause system failures. For instance, B-52 tests identified issues requiring extensive shielding retrofits, while avionics subsystems showed susceptibility to pulse-induced transients that disrupted navigation and communication functions. These findings prompted redesigns, including improved EMP-resistant shielding enclosures, fiber-optic signal lines to reduce electromagnetic coupling, and filtering on power and control lines to mitigate voltage surges. The B-1's success in achieving hardness through integral shielding and penetration controls served as a model, influencing subsequent hardening efforts across SAC assets with cost-effective adaptations. Validation of computational models for EMP coupling in unshielded versus hardened configurations was a critical byproduct, enabling predictive simulations for future designs.^[10] From 1980 to 1991, ATLAS-I conducted tests, including 13 near-threat-level simulations (four by the Air Force Weapons Laboratory and nine by systems program offices), providing empirical data on EMP propagation and mitigation. This testing supported the evolution of military hardening protocols for aircraft and related systems, contributing to advancements in EMP protection during the Cold War era.^[1]^[10]^[8]

Decommissioning and Legacy

Shutdown and Technological Replacement

Operations at ATLAS-I ceased in 1991, coinciding with the end of the Cold War and a diminished perception of nuclear threats that reduced the urgency for large-scale EMP testing of aircraft.^[1] This closure marked the termination of destructive physical EMP simulations on full-scale strategic assets, as geopolitical shifts lessened the need for such intensive hardening efforts.^[11] The facility's decommissioning facilitated a transition to more efficient alternatives, including computer-based simulations and smaller-scale EMP generators. Numerical modeling techniques, such as the finite-difference time-domain (FDTD) method, emerged as primary tools for predicting EMP effects, offering cost-effective analysis without the need for physical test structures.^[12] These computational approaches, validated against historical test data from ATLAS-I, enabled accurate vulnerability assessments for modern systems.^[13] High operational costs, estimated in the tens of millions of dollars for reactivation alone, contributed significantly to the decision to shut down, alongside escalating maintenance challenges. The massive wooden trestle structure, constructed from over 6 million board feet of lumber treated with creosote, suffered degradation in the arid New Mexico environment, drying out and increasing fire risks after the automatic suppression systems were deactivated.^[11]^[14] Following closure, demilitarization efforts involved dismantling key equipment, including the Marx generators and associated pulsers, while the trestle structure was partially repurposed for storage and administrative use by programs such as the U.S. Army's Big Crow initiative.^[5] This repurposing reflected broader post-Cold War resource reallocations at Kirtland Air Force Base.

Current Status and Preservation

As of 2025, ATLAS-I remains decommissioned and non-operational following its shutdown in 1991, when advancements in computer simulations rendered large-scale physical EMP testing obsolete.^[1]^[15] The facility, located on Kirtland Air Force Base in Albuquerque, New Mexico, is under restricted access, with public viewing limited to aerial perspectives or approved vantage points from non-restricted areas via coordination with base public affairs.^[1] Parts of the structure have been repurposed for military training, such as rappelling exercises by soldiers, while other sections house offices for programs like Big Crow and serve as a habitat for local wildlife.^[16] The site's fire suppression systems have been disconnected, increasing vulnerability to environmental risks.^[16] Preservation efforts have focused on recognizing ATLAS-I's historical value, though it has not been nominated to the National Register of Historic Places or granted formal protection status.^[1] Documentation initiatives, including video recordings archived at the National Atomic Museum, aim to educate on its role in Cold War-era defense research.^[16] Potential future uses include cultural or artistic projects to highlight its unique wooden architecture.^[16] ATLAS-I's legacy endures in the field of electromagnetic hardening, where its tests on full-scale aircraft like the B-52 and B-1B bombers advanced U.S. military resilience against high-altitude EMP threats during the Cold War.^[1] As the world's largest non-metallic wooden structure—spanning two football fields and standing 115 feet high—it exemplifies innovative engineering for EMP simulation, designed by electromagnetics expert Dr. Carl E. Baum.^[16]^[1] The facility's contributions, tied to broader nuclear research at Kirtland AFB, influenced post-Cold War standards for protecting electronics from electromagnetic interference.^[1]