Air University Review , July-August 1980

The Scorpion

a pictorial report

THE F-89 "Scorpion" was perhaps the most appropriately named fighter aircraft of the 1950s. With its decidedly radical, upward-swept cruciform tail and its large, missile-and-fuel-bearing wingtip pods, the F-89 bore a striking resemblance to its frightening arachnid namesake.

The development of Northrop Aircraft Corporation's Model N-24, as the F-89 was at first called, was initiated by the company in early 1945. Conceived as an all-weather ground attack fighter, it was to be powered by two of General Electric's new TG-180 axial-flow gas turbine engines. Developed from British jet engine patents, the TG-180 (or J35) was, in fact, one of the first production jet engines manufactured in the United States.

The first indication of the F-89's unusual design configuration came about when an engineering decision was made calling for the horizontal stabilizer and elevators to be placed high on the vertical fin. This arrangement was made in order for those surfaces to be clear of the turbulent and hot exhaust flow from the jet engines. Because of problems with stall blanketing, the high tail configuration would later prove to be more troublesome than anticipated.

The original Air Force design competition calling for an all-weather fighter brought submissions from Bell, Consolidated, Curtiss, Douglas, Goodyear, and Northrop. All six submissions were quickly found to be deficient in performance and, in effect, unsuitable for the called-for mission.

Rather than request new submissions from the various companies bidding on the program, a decision was made to proceed with the design that was considered least unsuitable. By chance, on 3 March 1946, Northrop was declared the winner and awarded a $4-million (later increased to $5.6 million) contract covering costs of two developed examples of their original design submission.

The new fighter, at the time still known as the P-89,* was to be a two-seat, twinjet, all-weather, day-or-night fighter. In consideration of the latter requirement, it was imperative that the airplane have the capability of carrying an exceptionally large and effective airborne radar.

*In 1948, all "P" designations were changed to "F."

By September 1946, the preliminary P-89 mock-up was available for Air Force inspection. Unfortunately, it got an unfavorable report. Northrop quickly returned to the drawing board and incorporated the many changes recommended by the Air Force. Among these were modifications for closer crew proximity, redesign of the canopy, and a change from magnesium to aluminum in certain wing structural areas.

Though further mock-up inspections eventually led to a tentative stamp of approval from Air Force examiners, faults were still discernible in a number of areas. These would later become more apparent after the P-89 entered the Air Force operational inventory. Not the least of the several, however, was questionable structural integrity in the empennage and vertical fin areas.

A number of design changes led to a nine-month delay in the completion of the prototype aircraft. On 16 August 1948, however, following roll-out ceremonies and several weeks of ground checks and taxi tests, the first XF-89 (AF. serial # 46-678), with test pilot Fred Bretcher at the controls, took to the air. As it turned out, the additional, unscheduled redesign efforts paid off, for preliminary flight evaluations by Northrop and Air Force test pilots found few problems with the airplane's general flight characteristics.

During the months following that first flight, the F-89 prototype was flown-off against competing designs from Curtiss (XF-87) and Lockheed (XF-90). (The latter, as point of interest, had been an unsolicited project primarily funded in-house by Lockheed.) It was eventually determined that the Northrop submission was the best of the lot, this conclusion based on its overall performance, its accommodations for on-board avionics, and its producibility.

In May 1949, the Air Force signed a cost-plus-fixed-fee contract with Northrop for a total of some $51 million. This sum would cover the production of 48 F-89As, one additional prototype, and a number of spare parts and static test articles.

The F-89 had been born when rudimentary developments in aerial electronic warfare were first achieved early in World War II. It was during this period that the first air intercept radars were used under experimental conditions. Production versions entered the operational inventory within a matter of months, and though big, heavy, and only partially effective, they did work. Most important, they proved that tracking and intercepting enemy aircraft at night and in bad weather were not only possible but quite feasible.

Because of the size of vintage radar units, World War II saw only large aircraft, such as the Northrop P-61 Black Widow and the British de Havilland Mosquito, used in the radar-directed air-intercept role. Electronics miniaturization was not long in coming, however, and by the end of the war tremendous strides had taken place toward smaller and more efficient units. Related postwar developments were a bit slow in getting under way due to the introduction of jet-propelled aircraft and their associated teething troubles, but by 1948 a number of preliminary projects had come to life with an orientation toward radar-directed intercept. The F-89 was a prime example.

The prototype F-89, following its first flight, had continued its flight test program at a steady but rather cautious pace. Following a minor landing accident that caused some slight belly and wing damage, it was rebuilt and used by the Air Force as a proof-of-concept airframe. Unfortunately, on 22 February 1950, it was totally destroyed in an accident.

The postaccident investigation revealed a number of major structural design failings in the basic F-89 airframe. Northrop reacted with a program to correct these faults, but terminal solutions were many years in the making.

The May 1949 contract eventually resulted in the completion of 37 F-89As. These aircraft were plagued with a number of major problems, not the least of which was a marked tendency to disintegrate when undergoing certain high-g maneuvers! The first production F-89As entered the operational inventory in limited numbers in 1952. These aircraft were used primarily as F-89 trainers and were quickly superseded on the Northrop production line by the slightly improved F-89B. This newer model had numerous minor internal changes and additional mission-related equipment that included a Lear F-5 autopilot, a Zero reader gyroscope, and a limited capability instrument landing system.

The first unit to receive the F-89B was the 84th Fighter Interceptor Squadron, at the time flying out of Hamilton AFB, California. Most of the thirty-seven F-89Bs eventually completed went to the 84th. It is interesting to note that total flyaway costs for the F-89B, $1 million per airplane, were considered quite high for the period!

While the Air Defense Command was absorbing the limited production F-89As and Bs, a further improved model, the F-89C, entered production. Shortly afterward, the operational Air Force inventory began receiving them. This was the first up-to-standard variant of the F-89 to reach quantity production, and 163 were eventually completed. Because of the number built, unit costs were significantly lower than those for the F-898, amounting to slightly less than $800,000 per airplane.

It was while the F-89C was entering the operational inventory that two major Scorpion design faults came to light. One, related to the airplane's engines, was brought on by deicing system failures and a propensity for "ramp sweeping"--a phenomenon peculiar to the F-89's low-slung engines (they literally sucked trash from the runway).

The other problem was far more serious. On 25 February 1952, an F-89 disintegrated during flight maneuvers and all its crew were lost. This was followed, over a period of several months, by a disturbing number of similar fatal accidents, a total of six by 15 September. On 22 September, all F-89s, with the exception of a select number to be used for test purposes, were grounded.

Examination of F-89 wreckage, coupled with an intensive test program, revealed the cause of the failures. Poor pitch stability coupled with a structural fatigue problem were determined to be the culprits. During the course of high-g maneuvering, common during intercept missions, the poor stability characteristic and the structural fatigue problem could, in combination, lead to catastrophic airframe failure.

The Air Force and Northrop began to collaborate on a program to modify and improve all F-89s in the inventory. Additionally, all aircraft on the Northrop production line were to be similarly modified and improved. The cost of this program was no less than $17 million. Unfortunately, by the time work had been completed in January 1954, there was still a 20 percent limitation on the F-89's performance envelope. This limitation would remain with the affected aircraft throughout their operational lives.

The structural problems manifest in the basic F-89 design had come to light just as the Air Force began accepting the first of the improved "0" series. Though a number of "Cs" had been lost under similar circumstances, the problem had never been fingered as being endemic. The "0" series accidents changed all that. Five F-890s had been accepted by the time the structural problem came to light. An additional 120 "Os" had also been produced, though not yet accepted. On discovering the wing spar fatigue problem, Northrop initiated a major modification and "beef-up" program. All completed F-89s were affected, including a total of 170 F-890s and 194 F-89As, Bs, and Cs. Nor was the F-89 particularly healthy in the armament department either. Though advertised at the time as being the world's "most heavily armed interceptor," it was, in truth, one of the least effective aircraft ever to operate in the interceptor role. Problems with the F-89's sting, in fact, were sufficient to lead to a number of performance restrictions, the majority of which were never removed. Armament for the standard F-890 consisted primarily of fifty-two 2.75" folding-fin air-to-air rockets in each wingtip pod or provision for three Hughes GAR-1, GAR-2, GAR-3, or GAR-4 Falcon air-to-air missiles in each of these same pods. Early F-89A models had provision for six 20-mm T-31 (M-24) guns in the nose with 200 rounds per gun. It was also possible to mount external free-falling stores under the wings of most F-89 models.

The effectiveness of the Hughes Falcon series in the air-to-air combat role was very questionable throughout the late 1950s and early 1960s, when it first entered operational service. Dependability was marginal, at best, and the sensitivity of its optional guidance systems to countermeasures was extreme. In retrospect, it is not unfair to say that the missile would have proved itself almost completely ineffective in a real-war scenario.

Similarly, the F-89's AN/APG-33 intercept radar (and associated systems) was also riddled with problems. It took an inordinate amount of time to warm up; it was quite susceptible to countermeasures; and downtime and maintainability showed poor performance.

From its conception in 1945 to the final model's production run, the F-89 went through a number of major and minor engine changes. The original project, mentioned at the beginning of this story, powered by the General Electric TG-180, had not lasted long. By the time of the prototype airplane's first flight, the power plants had been changed to Allison J35s. This power plant was to remain the standard F-89 engine throughout its operational career. Various models were used, the majority rated in the 7200-pound thrust category with afterburner.

It should be noted that one F-89, the YF-89E, was test flown with Allison J71 s in place of the normal J35s. This program was relatively successful but offered few significant performance improvements over the standard airplane.

ALTOGETHER, 682 F-89Ds were completed by Northrop before production was terminated. During the course of the F-89D's development and production run, a proposal was made by Northrop outlining a program wherein a number of F-89Ds would be modified to carry the then state-of-the-art Hughes Falcon air-to-air missile. This missile was considered the most effective weapon of its kind in the world at the time, and until the advent of the F-89, an effective delivery vehicle had not been found for it. The Falcon/Scorpion integration program was initiated in January 1954, approved the next month, and terminated in March. It would soon be reinstated under the F-89H program.

The penultimate F-89, the F-89J, was not really a new model at all; rather it was a modification of an older model that had suffered through a number of serious problems and had emerged as a useful, albeit dated interceptor. The F-89J was simply an updated and improved modification of 350 older F-89Ds. The "J," however, offered a number of advances over the earlier configuration, not the least of which was the ability to carry a total of two Douglas MB-1 Genie unguided, nuclear-tipped, air-to-air missiles.

The Genie was an awesome bit of weaponry. It was one of the first examples of miniaturized nuclear capability and as such was possibly the first weapon of its type to enter operational service. Powered by a solid-fuel rocket engine that could propel it horizontally for as much as six miles at speeds approaching 2000 mph, the Genie was considered a quantum jump forward for the Air Force when it entered service for the first time in January 1957.

The last of the production F-89 models was the F-89H. The "H" was the result of an armament system first test flown aboard an F-89D in the mid-1950s. The new development consisted of wingtip pods that could house, internally, up to three Hughes Falcon air-to-air missiles each, along with 21 folding-fin air-to-air rockets. Additional weaponry could be mounted under the wings. Altogether a total of 156 F-89Hs were completed before the Northrop Scorpion production program rolled to a halt.

The standard F-89D carried a pilot and radar operator seated in tandem, had a wingspan of 59'8", a length of 53'10", a height of 17'7", and a gross takeoff weight of 42,250 pounds. Maximum speed was 636 mph, service ceiling was just under 50,000 feet, and ferry range was about 1350 miles. Propulsion was provided by two Allison J35 turbojet engines rated at 7200 pounds thrust each with afterburner.

In spite of its many problems, the F-89, during its later years with the Air Defense Command and the Air National Guard, proved to be a moderately good performer. As much as anything else, it served to prove the viability of electronic warfare in a fighter-versus-fighter environment.

Jay Niessen Miller (B.A., University of Texas at Austin) is Editor of Aerophile and Associate Editor of Wings/Airpower: He has been Editor of Texas Aeronautics Commission Bulletin and curator of the University of Texas History of Aviation Collection. He is a contributing columnist to several publications including the Air Force Times. Miller is a professional aviation photographer and active in many aviation organizations.

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

Air & Space Power Home Page | Feedback? Email the Editor