How a failed escort fighter concept evolved into one of the Cold War’s most capable—and dangerous—tactical jets
James S. McDonnell’s company, reformed in St. Louis in 1939, cut its teeth on naval jet fighters—the FH-1 Phantom and F2H Banshee—before turning its ambitions to the USAF. In April 1946, designers Edward ‘Bud’ Flesh and Kendall Perkins began work on the XF-88, a long-range “penetration fighter” intended to escort Convair B-36 bombers on 10,000-mile sorties. The USAF demanded 600 mph, a 40,000-ft ceiling, and a 900-mile combat radius.
The XF-88 introduced a 35-degree swept wing—proven on the F-86 Sabre and validated by captured German research—along with an all-moving tailplane, irreversible powered flying controls, and twin Westinghouse J34 engines fed by wing-root intakes. McDonnell developed its own afterburners after engine manufacturers declined to supply them, a first for an aircraft manufacturer, incorporating the first multi-element jet nozzle for a turbojet and pushing maximum speed to 708 mph. The aircraft was named Voodoo, continuing the company’s tradition of occult-themed names.
The prototype flew on 20 October 1948. Performance with non-afterburning engines disappointed, but the afterburner-equipped XF-88A reached Mach 1.18 in a dive on 12 May 1949 — still a rarity at the time. A three-way fly-off at Edwards AFB in June–July 1950, against Lockheed’s XF-90 and North American’s YF-93A, saw the XF-88A ranked first by the USAF’s Air Materiel Command, with experienced pilots including Capt. Chuck Yeager and Lt. Col. Pete Everest were among the evaluators. Nevertheless, procurement was ruled out — the Korean War had shifted priorities, and established types like the F-86 and F-84 were already rolling off production lines. Both XF-88As were scrapped at Langley AFB by 1958.
Neither the XF-88A nor its rival, the XF-90, could realistically have fulfilled the escort mission. The era’s immature jet engines hamstrung both: poor thrust-to-weight ratios, sluggish acceleration, and high fuel consumption left little margin for combat performance.
Enter the F-101 Voodoo
The arrival of the Pratt & Whitney J57 in 1952—offering up to 15,000 lbs of afterburning thrust against the J34’s 3,600 lbs—transformed what had been an unreachable concept into something viable. Flesh’s team returned to the XF-88 drawing board, scaling up the airframe significantly: fuselage length grew from 54 ft 2 in. to 67 ft 5 in., gross weight doubled to around 48,000 lbs, and internal fuel capacity reached 2,340 gallons—five times that of an F-84 Thunderjet. The redesigned aircraft received the designation XF-101 following a USAF Letter of Intent in January 1952.
External changes included larger variable-ramp air intakes, an extended inboard wing chord, and a repositioned horizontal stabilizer set high on a tail fin twice the size of the XF-88’s. Armament comprised four 20 mm cannons, Hughes AIM-4 Falcon missiles, unguided rockets, and provision for a Mk 5 nuclear weapon delivered via a “toss-bombing” technique. A retractable inflight-refueling probe and AN/APG-37 radar were housed in the forward fuselage.
The first F-101A flew on 29 September 1954, with company Chief Test Pilot Bob Little outrunning his F-100 chase aircraft and reaching Mach 1.07—the first time any aircraft had gone supersonic on its maiden flight. However, serious problems quickly emerged.
The Pitch-Up Problem
As engine modifications allowed higher angles of attack, a dangerous aerodynamic phenomenon appeared: airflow disrupted over the high-set tailplane caused loss of pitch control, the nose would rise uncontrollably, and the aircraft could enter an unrecoverable spin. The condition occurred rapidly and without warning at supersonic speeds.
Capt John Dolan was killed when his aircraft entered a flat spin at Edwards AFB. Korean War ace Maj Lonnie Moore lost control on takeoff at Eglin AFB in a fatal crash. McDonnell test pilot Bill Ross entered a flat spin but ejected safely. After 200 spin-recovery tests, McDonnell established avoidance and recovery procedures, though these required significant altitude—recovery from a stall at 30,000 ft could consume up to 12,000 ft. A stall-warning system with an automatic “stick knocker” was eventually fitted, though it addressed symptoms rather than causes.
AFCS technician Neil Byrd recalled the system: “At a certain AoA, a warning would show and sound in the cockpit, and beyond that, at a defined angle, the AFCS would drive the stick forward regardless of pilot input, which seemed like a rather drastic solution. When the aircraft had entered a pitch-up situation, the only recourse was to deploy the drag ‘chute and hope there was 10,000 ft altitude for recovery.”
The F-101A and RF-101A airframes were also found to withstand only 6.33 g rather than the specified 7.33 g—a deficiency corrected in the heavier F-101C and RF-101C variants, which also benefited from improved afterburner cooling, allowing virtually unrestricted reheat use below 20,000 ft.
Service Career
SAC cancelled its F-101 order on the same day as Little’s first flight, having concluded the bomber escort mission was impractical. Tactical Air Command took over, and 77 F-101As were delivered to the 27th Strategic Fighter Wing at Bergstrom AFB from May 1957. The unit transferred to TAC as the 27th Fighter Bomber Wing. The F-101A proved less versatile than the Republic F-105B Thunderchief and was replaced by the F-100 Super Sabre after just 18 months.
The 27th FBW’s aircraft, along with many F-101Cs, subsequently transferred to the 81st TFW at RAF Woodbridge and RAF Bentwaters, where they maintained nuclear Victor Alert until January 1966. Col Boyd Van Horn, who flew the type there alongside Robin Olds and ‘Chappie’ James, recalled the mission approvingly:
“I enjoyed my tour at Bentwaters more than any other in my 29 years. Robin, with his background and leadership abilities, put us at the top. The F-101 was a great airplane, very stable and perfect for the nuclear mission we had. My target was four hours away with air refueling over the North Sea at very low altitude, radio out… Our approach was at 500 knots at 50 ft, and when we got the pull-up signal, we would pull 4 g’s. The bomb was released almost straight up and went to 40,000 ft. We would do an Immelmann turn and dive back to the deck to get at least five miles away because of the over-pressure of a one megaton bomb.”
The reconnaissance RF-101C emerged as perhaps the most operationally significant variant. Ninety-six were built from the F-101C production batch, with the type eventually operating over North Vietnam in 1965, where speed—its sole defense—proved essential. Later, 29 F-101As and 32 F-101Cs were converted to RF-101G and RF-101H reconnaissance platforms, respectively.
The XF-88B variant, rebuilt with an Allison XT38 turboprop in its nose for a NACA program, became the world’s first official supersonic turboprop aircraft on 27 June 1953. The basic design also spawned proposals for strike fighters, two-seat all-weather fighters, and carrier-based derivatives, though none advanced beyond the concept stage. Many of the XF-88’s features appeared, refined, in the definitive F-101 a decade later.
This article is from the book “RF-101 Voodoo units in combat” by Peter E. Davies, available here.
