F-111 Mark II Avionics System
Applying "graduated pressure" in limited war put a premium upon the accurate delivery of conventional munitions. Between 1945 and 1965, the techniques for hitting targets with iron bombs underwent little improvement. The launching of ROLLING THUNDER, combined with interdiction of the Ho Chi Minh Trail, highlighted the lack of night and all-weather attack capabilities. A quantum jump appeared feasible, however. The Mark I, consisting of an attack radar, a navigation-attack system, and a lead computing optical sight, was to be installed on the F-111's A, C, and E models.
The Air Force wanted reliable systems, minimum maintenance per flight hour, and high sortie rates from austere bases. Gen. Frank F. Everest, who had been Commanding General, U. S. Air Forces, Europe during 1959-61, strongly resisted any proposals that would push the Mark I beyond the limits of proven technologies. But many civilians in the office of DDR&E saw aircraft mainly as vehicles for avionics and considered the Mark I to be austere, useful primarily for delivering tactical nuclear weapons from a low level at supersonic speeds. Supported by the President's Science Advisory Board as well as the Air Force Scientific Advisory Board, DDR&E pressed for a much more sophisticated Mark II and Secretary McNamara agreed.
Sperry, Hughes, and North American Rockwell's Autonetics Division competed for the Mark II contract. On 27 June 1966, Autonetics was chosen. Theoretically, the Mark II would be able to separate a target from all ground clutter and maintain a fix on it, regardless of the plane's speed and altitude and even whether the pilot could see it. A Mark II contained seven major components. An IBM computer, weighing 47 pounds and filling less than one square foot, would present computations-numbers shown on a panel-responding to any question about navigation or weapon delivery. A Horizontal Situation Display about the size of a small radar screen, to be developed by Astronautics Corporation of America, would show the pilot where he was at every second. Under a tiny aircraft silhouette, a map would move in any direction as the silhouette continued to show the aircraft's actual location and heading. The IBM computer, knowing the aircraft's location, speed, and altitude, would move the map.
A Doppler processing system, contained in the Autonetics radar, would create a completely new ability to hit moving targets automatically. It did so by eliminating all fixed targets from the display, leaving only the moving ones and tracking them wherever they went. To sum up, the Mark II would do away with the Mark I's limitations of "knee-pad navigation" and pre-canned mission constraints. Autonetics' navigation system would cut the "specified drift error" to .5 nautical miles per hour, compared to 2.0 in the Mark I. Instead of depending on the pilots themselves to integrate successive sensings, the Mark II would answer each operational problem with inputs that were correlated directly from all components.
On 1 July 1966, Autonetics entered into Purchase Agreement 181 with the F-111's prime contractor, General Dynamics. A fixed-price incentive contract put the target price at $145 million-an estimate of $183 million minus a $38 million "management reduction." General Dynamics could negotiate downward adjustments to the target price, but Autonetics could not negotiate any upward adjustments to the Statement of Work. Purchase Agreement 181 established a baseline for performance rather than design specifications for individual end items. Later, Autonetics would claim that the contract definition phase had not established a design baseline. Autonetics had agreed, however, to develop and produce the Mark II according to its own performance baseline, incorporating whatever changes the periodic deficiency reports showed to be necessary.
Time pressures shaped the project. To make the replacement of Mark Is with Mark IIs in F-111As cost effective, installation of Mark IIs would have to begin with the third wing of aircraft. Stripped-down Mark IIBs were planned for FB-111s, the strategic bomber variant. To meet delivery schedules for FB-111s, though, the first Mark IIBs would be needed six months ahead of the Mark IIs. Consequently, thirty-three months were available for contract definition, development, and initial production of Mark IIs but only twenty-eight months for Mark IIBs. Merely three months were allowed for preparation of subcontractors' proposals and six weeks for their evaluation and source selection, creating a high degree of concurrency. In the schedule drawn for F-111Ds carrying Mark IIs, flight tests would start only six months before production deliveries so test results could not be used to improve the initial production units. The schedule drawn for FB-111s was even tighter, with the first test unit slated for completion four months after a production go-ahead.
Autonetics acted promptly to adopt the findings from deficiency reports, making substantial changes in many designs. By September 1966 a new set of performance specifications-called the "A-1 Specs"--had been drafted and some degradations in performance accepted. On 15 March 1967, Autonetics asked for a price increase to $367 million. When Secretary of the Air Force Harold Brown rejected that figure, Autonetics on 15 April proposed $297 million. The F-111's System Project Officer, Brig. Gen. J. L. Zoeckler, told General Dynamics that $297 million was a "not to exceed" price about which negotiations could begin-provided that Autonetics could trace when and why the increase had occurred. Autonetics countered that traceability was impossible. Instead, it attributed the $163 million increase to government Contract Change Notices 1 though 12, which had been issued before the A-1 Specs appeared. At Deputy Secretary Vance's direction, the Air Force established its own technical trace team directed by Gen. Zoeckler. The team identified $20.2 million as the maximum amount of government responsibility. Design changes had been made to meet performance specifications, not to implement Contract Change Notices. Specifications had been revised at Autonetics' request to relax those that it was unable to meet, not to accommodate changing Air Force requirements.
By January 1968, Autonetics anticipated an increase of $600,000 per production unit. Worse, the normal "learning curve" in which costs decreased as quantity production proceeded would not occur. In fact, follow-on costs for the first year probably would exceed the cost of units already under contract. The table below shows how far cost estimates had ballooned:
Date of Research, Development, Production Total Estimate Testing, and Evaluation June 1966 $39 million $386 million $425 million January 1968 $173 million $917 million $1,089 million
The Air Force did not want a major supplier to suffer heavy loss, but it saw no facts that would justify higher prices. Many in the Air Force viewed Autonetics as the "innocent victim" of a system that was changing from oral promises to written agreements. Others, however, felt that the contract must be enforced to maintain the integrity of the source selection process.
William W. George, who worked under Comptroller Robert Anthony, analyzed the Mark II procurement and in May 1968 reached a disturbing conclusion. All the reforms of the 1960s had been employed: meaningful competition by strong contenders; a fixed-price incentive contract with a tight ceiling and a cost-sharing pattern; adoption of total package procurement, the only variation being exclusion of support and logistics equipment; use of concept formulation and contract definition as set out by DoD Directive 3200.9; allowing the contractor to specify technical characteristics, schedule milestones, and cost targets; using performance rather than design specifications; signing a definitized contract immediately after source selection; and minimizing the number of substantive changes to requirements. Why, then, were the results still unsatisfactory? Hurrying the final phase of contract definition and forcing a high degree of concurrency, which George labeled "the biggest mistake of the Mark II procurement," did not by themselves explain such a severe overrun.
In fact, the usual causes of overruns (sole-source contracting, undefined requirements, numerous changes, undefinitized contracts, lack of risk placed upon the contractor, "gold-plating" by government engineers, and major unexpected problems) were conspicuously absent. Instead, George put forward two explanations. First, both Autonetics and the Air Force had been unrealistic about ultimate costs. Aggressive competition led bidders to be unduly optimistic. Autonetics, he surmised, knew what it was getting into but resorted to the familiar tactic of "buying in." The Air Force, anxious to sell the Mark II to OSD and Congress, made no independent cost assessment at the time of source selection. Second, at source selection time and for months thereafter, Autonetics believed it would not be held to written agreements and tried to use Contract Change Notices as a means of "getting well." The Air Force, however, made no changes of any magnitude. It did rush the final stages of contract definition but was encouraged to do so by what George called Autonetics' "unguarded optimism" that deficiencies could easily be set right.
Cost controlling incentives were "vastly overshadowed" by Autonetics' desire to fulfill performance and schedule specifications, thereby assuring the profits from follow-on contracts. Finally, in March 1967, Autonetics realized that it would have to redesign substantially and adopted a strategy of contending that the original contract had been invalidated by Contract Change Notices. This was exactly the strategy that Autonetics had employed successfully in its Minuteman program. Unfortunately, Autonetics failed to appreciate that the procurement system was shifting from oral promises to written agreements.
Autonetics delivered the first Mark IIs to General Dynamics on 21 November 1967. Flight testing by an F-111A began on 31 March 1968. Then the "unanticipated unknowns" appeared. During the first full system test, in June, components started interfering with each other. Radar being tested in a B-66 bomber was found to operate twenty decibels below its specified performance, about half the deficiency resulting from loss of transmitter power. Because of transmitter instability as well as a need for major redesign of the Doppler processing system, Autonetics had to postpone tests of radar in the "look-down" or "squint" modes. In mid-September 1968, Charles Fowler (Deputy Director, Tactical Warfare Systems, DDR&E) warned Dr. Foster and Secretary Brown that these problems looked "formidable." Fowler also cited, as possible areas of major trouble, high-power traveling wave tubes and storage tubes. Already, cracking had occurred at the windows of traveling wave tubes. An alternative tube was being developed but, "as you know, Murphy's Law applies to new tubes more than anything else."
By late 1969, the Mark II's snowballing cost reduced the F-111D program from 315 to 96, and component costs swelled as mass production slumped. The Air Force accepted one F-111D in June 1970 but none followed for the next twelve months because of the Mark II's unavailability. Facing a possible loss of $128 million, Norden on 31 October 1970 stopped producing Integrated Display Sets, assembling only five more for testing. After prototype tests yielded much better results than expected, the Air Force promised Norden an extra $63.2 million and output resumed. Still, by June 1972 only 24 F-111Ds were available-two years beyond the time when a 72-plane wing should have been operationally ready. Even then, F-111Ds were crippled by lack of spares; the Horizontal Situation Display had a field reliability life of only fifty hours.
Commonality (with FB-111A avionics, in particular), a prime requirement of the Mark II systems envisioned by Secretary McNamara in 1966, had long disappeared. Technical problems, remedial cures, and expedients had left the F-111D with a complex, highly integrated, one-of-a-kind avionics system.
The Mark II stands as an example of technological over-reach that no acquisition reforms could remedy. There was a basic incompatibility between air-to-air systems needing extremely fast data-processing rates and variable waveform flexibility, and air-to-ground ground systems needing very large data storage and processing capabilities for high-resolution mapping. Autonetics' strategy of buying in, accomplished with the complicity of its backers in the Air force and OSD, only compounded these difficulties.
|Join the GlobalSecurity.org mailing list|