Air Force officials believe that the problem of Northrop's late delivery of MX missile guidance systems is being corrected but acknowledge that there has been no improvement since last June in the percentage of MX missiles deployed without any guidance system.
Twenty-eight MX missiles have been put into underground silos at F. E. Warren Air Force Base in Wyoming, but only 19 have guidance systems and could be launched, Air Force officials said in recent interviews. Last June, only 12 of 18 deployed missiles had guidance systems.
In addition to Northrop's late deliveries, a second problem has emerged involving repairs on failed guidance devices, according to Col. Thomas Speed, MX program manager at the Air Force's Ballistic Missile Office in San Bernardino.
Speed disclosed that it is taking more than twice as long as originally expected to repair the guidance systems--formally known as inertial measurement units--when they break down. More often than expected, repairs must be done at the factory instead of a repair depot, Speed said. It is not clear who is bearing the cost of the additional repair time, however.
Although Northrop has delivered 66 IMUs since production began in 1984, the Air Force has enough properly operating IMUs for only 19 missiles, a ratio that has not improved in recent months, Speed said.
"I am not trying to tell you . . . everything is wonderful," he said. "It has taken us more than twice as long (as expected) to get them repaired, for a variety of reasons."
Speed said each IMU, on average, is failing every 3,006 hours of operation--about 10% fewer hours than had been expected.
Northrop is responsible for assembling the guidance unit from parts that it makes and that are supplied by other contractors.
Northrop's portion of the system is doing substantially better than expected, he said, but instruments supplied by Honeywell, a Minneapolis-based defense contractor, are failing twice as often as expected at this point in the program.
As a result, the overall guidance system has gone from a better than expected reliability last June, by Air Force standards, to a slightly lower than expected reliability now, he said.
The Honeywell instrument, called an accelerometer, is used to calculate the missile's acceleration and thereby its velocity.
Debris on Balance Wheel
The problem with the accelerometer is microscopic debris on its "balance wheel," which is about the size of a grape and is made of machined beryllium. The device fits so precisely within its housing that the contaminants are jamming the wheel against the housing.
After extensive investigation, Speed said, the Air Force now believes that the problem resulted from Honeywell failing in early 1986 to change a filter in a tank used to ultrasonically clean the wheels at its factory in Clearwater, Fla.
Since the Honeywell instruments are in the core of the IMU, virtually the entire guidance device must be torn down to repair it. "You get back into a manufacturing process," Speed said about the IMU repairs. "We did not expect that.
The reliability of the IMU has been the subject of wide debate ever since three of Northrop's own engineers alleged that the IMU has serious technical defects that make it unreliable and inaccurate. The engineers are suing the company under the federal False Claims Act.
"Obviously Northrop has recognized they have serious problems with the IMU," said Robert Kilborne, an attorney representing the engineers. "That is why they failed to deliver any of them in November. They are working very hard to correct the problems so that past problems don't get repeated in new units, but that doesn't address the IMUs in the field. All of the IMUs in the field need to be torn down and retested."
The Air Force has long disputed critics who have alleged that the MX guidance system is grossly unreliable due to technical flaws in Northrop's products.
The MX was intended to achieve a "revolutionary improvement" over the guidance systems on existing long-range ballistic missiles. And, by Air Force thinking, it has not only achieved that but exceeded it.
"We think it is an incredibly accurate system," said Maj. Gen. Edward P. Barry Jr., commander of the Ballistic Missile Office. "We are tickled to death."
The MX carries 10 nuclear warheads capable of hitting separate targets at ranges of 6,000 miles.
Its degree of accuracy is classified, but one military official said the missile is accurate enough to consistently put warheads into an area about the size of the Los Angeles Memorial Coliseum.
Barry said the MX is already meeting accuracy requirements that were not expected to be met until more than three years from now. So far, 17 tests have all been scored as successes by technical Air Force standards.
But critics remain unconvinced. Rep. Ron Wyden (D-Ore.), a member of the House Energy and Commerce Committee, rejected the Air Force's system for scoring accuracy at a hearing last October, terming it "technical-legal lingo Air Force definitions."
Under Air Force definitions, the missile meets its accuracy requirements if half of the missiles fall within a certain circle and if all of the missiles fall into a larger outer circle about twice the diameter. In either case, the warhead would destroy its target.
But in Wyden's view, the MX could not be as accurate as the Air Force claims, he said, because three of the last eight test shots missed their targets and, in the only test of a production guidance system, every re-entry vehicle missed its target.