Safety Margin Believed Ebbing for Air Traffic : Spending Cuts, Obsolete Equipment, Long Hours by Fewer People Cited by Controllers, Officials

“I have been thoroughly briefed by members of my staff as to the deplorable state of the nation’s air traffic control system. They have told me that too few people working unreasonable hours with obsolete equipment have placed the nation’s travelers in unwarranted danger.”

The man who wrote those words was Ronald Reagan. The year was 1980, and Reagan, seeking the presidency, promised that if elected he would take whatever steps were necessary to “provide our air traffic controllers with the most modern equipment” and make the necessary adjustments in staff levels and work hours.

But today, more than six years after Reagan was elected, there are public officials, labor representatives, controllers and maintenance personnel who say that even fewer people are working even less reasonable hours with even more obsolete equipment.

They say that our nation’s air traffic control system is still the best in the world. But they also say that White House efforts to cut spending across the board and growing lags in manpower and equipment replacement schedules are rapidly taking their toll--at a time when the volume of commercial air traffic continues to expand.

Margin for Error

While most agree that the system is still basically safe, they also agree that the margin for error is rapidly vanishing.

They cite repeated failures in aging Federal Aviation Administration computer, radio and radar equipment that can leave air traffic controllers deaf, mute or blind.

Many of them say there aren’t enough controllers to man the gear properly when it does work, and not enough technicians to fix it when it doesn’t. Some say that tiring work schedules and poor communications between controllers and their supervisors are creating serious morale problems.

The Los Angeles Terminal Radar Approach Control (L.A. TRACON) center and the control tower at Los Angeles International Airport--one of the busiest airfields in the world--are one case in point.

The Los Angeles Air Route Traffic Control Center at Palmdale--which controls aircraft over a 180,000-square mile chunk of airspace ranging across southern Utah, southern Nevada, western Arizona, Southern California and a stretch of the Pacific reaching 200 miles off the coast--is another.

L.A. TRACON, the LAX tower and the center at Palmdale are part of a network of control facilities that provides overlapping radar coverage of the entire nation. It’s a network burdened by commercial air traffic that grew from 275 million passenger boardings in 1978 to 380 million in 1985, with 610 million forecast by 1996. It is a network that is beginning to suffer from the corrosive effects of time.

“When I came here last December, my whole impression was that I’d never seen a facility that had as many equipment outages as here,” said Richard Cox, the FAA’s air traffic manager at L.A. TRACON, which is housed in an old hangar known as “The Trashcan.”

L.A. TRACON handles the approach and departure routes for the more than 1,700 flights, carrying more than 100,000 passengers, that take off and land at Los Angeles International (LAX) every day.

“You’d think this place would be one of the flagships,” Cox said. “Instead, it’s more like a dinghy.”

Although there has been no evidence that the aging equipment at L.A. TRACON was a contributing factor in the Aug. 31 collision between an Aeromexico jetliner and a small private plane over Cerritos that claimed 82 lives, the reliability of the facility’s radar, which was tracking the ill-fated jetliner, was questioned during hearings earlier this month by the National Transportation Safety Board.

L.A. TRACON depends on two radar systems--a 1960s-vintage ASR-4 on the north side of the airport and a 1970s-vintage ASR-7 on the south side.

The ASR-4 has been out of service 16 times during the last year, forcing controllers to rely entirely on the overlapping coverage provided by the ASR-7, according to Dick Muckle, who oversees the maintenance of air traffic control equipment in the Los Angeles area.

‘Not Been a Good Year’

“It’s not been a good year for the 4,” Muckle conceded sadly.

The ASR-7 has been out of service at least six times during the same period, forcing reliance on the ASR-4, according to FAA maintenance logs.

And on one occasion--last Sept. 24--with the ASR-7 already malfunctioning, the ASR-4 was knocked out by a power surge probably caused by lightning, leaving the controllers at L.A. TRACON momentarily blind.

“When one radar is down, it causes concern,” Muckle said. “When both go down, there’s really no impact on safety, but it sure slows things down.”

With radar working, controllers can allow aircraft to fly within three miles of each other as they approach LAX. Without radar, controllers must increase separation to 20 miles, falling back 40 years to the procedures their forerunners used--relying on position reports radioed by the individual planes. That is how air traffic control in most nations still operates.

While most controllers agree with Muckle that radar loss is more an inconvenience than a hazard-- reducing, but not erasing, a needed margin of safety--the loss of radar means the loss of “snitch” alarms, which warn controllers that planes may be on a collision course.

At the LAX tower cab, which handles takeoffs, final approaches and taxiway traffic, an old radar system used to track taxiing planes was out of service for seven months this year because replacement parts were no longer being manufactured, according to Jim Holtsclaw, traffic manager at the tower.

Holtsclaw said that the radar loss was mostly an inconvenience, resulting in delays moving planes in and out of loading gates.

The Los Angeles Air Route Traffic Control Center--one of 20 regional FAA control facilities that together provide blanket radar coverage of the entire nation--is housed in an attractive building in the open fields near Palmdale. It’s more of a showplace than its city cousins, but it, too, is showing its age.

“Our equipment isn’t worth a damn,” said Dave Allec, a controller at Palmdale. “It’s antiquated, and it’s rapidly reaching a point where if something goes out, they may not be able to make it come back.”

“I have some equipment with a date of 1945 stamped on it,” said Paul Beam, a repair technician at the center. “The insulation is getting so old that it’s just falling off.”

The center at Palmdale relies on 10 radar systems--some as young as the state-of-the-art equipment atop Mt. Laguna in San Diego County, but others as old as the 1950s-vintage gear at Las Vegas--to provide coverage of some of the most rugged mountain and desert terrain in the world.

While most of the coverage is overlapping, tall mountains and deep valleys mean some areas will be blind without a particular radar. That means some controllers have to ask pilots to report their position by radio if certain radars fail. And fail they do.

In the six months ending Oct. 1, radar system outages ranged from none at Paso Robles to eight at Boron, according to Bob Cox, the FAA’s airways facilities manager at Palmdale. FAA records indicate that the old ASR-4 at LAX, which feeds into the system at Palmdale, failed four times during that period. Even when the radars serving the L.A. TRACON center and the center at Palmdale were functioning, they may not have been functioning well, sometimes causing erroneous information to appear on the controllers’ screens.

Maintenance records indicate that the radar systems suffered variously and repeatedly from a series of ailments that reduced the effectiveness of either or both of their two fully redundant channels, each of which handles two kinds of coverage--primary and secondary.

Sometimes it was the primary circuits, providing simple echoes showing a plane’s location, that were affected. Sometimes it was secondary circuits that offer up a data tag on the screen showing a plane’s identification code, altitude and speed.

The ailments, while not knocking out an entire radar system, may force controllers to switch channels or limp though with impaired coverage. In either case, the margin of safety supplied by fully functional redundant systems simply isn’t there.

Among the ailments mentioned by controllers:

- “Weak returns” that can cause images on the display screen to fade.

- “Noise,” visual static that can obscure the picture on the screen.

- “Uncanceling” that causes stationary objects on the screen to look like airplanes.

- “Speed jumps” that make planes appear to leap suddenly ahead.

- “Anomalous propagation,” a reflection caused by heat inversion over desert valleys that can cause a plane to appear in more than one spot at the same time. Controllers then must try to figure out which image is the real plane.

- “Blind spots,” gaps in radar coverage caused by mountains or other prominent terrain features.

Most of the radar complaints involved gaps and jumps.

“There are blind spots near Fillmore and Point Mugu,” said Karl Grundmann, an L.A. TRACON controller who also serves as a regional representative for the National Air Traffic Controllers Assn. “You may lose the primary, you may lose the data tag, you may lose it all. It happens all the time.”

“We used to get a lot of speed jumps at Mt. Laguna, sometimes several times a day,” Bob Cox said. “We changed some components on the radar, and that finally fixed it.”

There have been other problems with the radar systems.

FAA personnel said the microwave, telephone and cable systems used to transmit radar data from the antennas to the control centers have occasionally failed. When that happened, technicians rerouted the transmissions through alternate systems.

And on Sept. 11, 1985, technicians at LAX reported that the ASR-4 radar beside a Westchester Club fairway had been dented by golf balls. When that happened, the technicians recommended a better swing.

Even if the radar systems themselves are functioning properly, the main computers that refine data sent from the antenna can garble the display on the controllers’ screens, making an already demanding job even more difficult. While the ‘70s-vintage ARTS III computer used by L.A. TRACON is the subject of occasional criticism, most of the problems involved the IBM 9020 used at the Palmdale center.

Bob Cox said the 9020 has failed 12 times since June, with an average failure time of about four minutes. When that happened, a fully redundant 9020 system usually took over. On the few occasions that the redundancy was not available, most of the 9020’s functions were taken over by a backup computer system. There have been no complete computer systems failures since 1980.

Controllers’ complaints about the 9020 involved malfunctions--rather than systems failure--malfunctions again characterized as more of an inconvenience than a safety hazard. Among the complaints:

- “Data tag swaps,” in which display screens switch the data tag from one plane to another aircraft near it.

- “Snitching,” in which hypersensitivity causes the computer to sound unnecessary alarms indicating that a controller has allowed aircraft to get too close to one another. The name is derived from the fact that copies of such warnings are given to supervisors.

- “Overload,” a situation in which the computer is asked to track more planes than it can handle. When this happens, the computer begins dropping data tag information, with the oldest information--that which has appeared longest on the screen--disappearing first.

The computer complaints seemed about evenly distributed among controllers.

“We often get data tag swaps when a couple of planes are in line with, say, one at 7,000 feet and the other at 5,000,” Grundmann said. “Sometimes one will pick up the other’s altitude, give you a reading of 12,000. Sometimes they’ll both read 6,000. That sort of thing happens every day.”

“Sometimes the snitch should go off, and it doesn’t,” said Kathy Heet, a controller at Palmdale. “And sometimes it goes off when it shouldn’t. It happens all day long.”

“Sometimes, when there’s a heavy load, the computer gets slow,” said Dave Whalen, another controller at the Palmdale center. “You may lose a target for a second or so. It’s not gone altogether, just dim. But that’s a problem.”

Another source of complaints is the digitizer computers at radar sites that the Palmdale center depends on. The digitizer computers encode the data from the antennas for transmission to Palmdale and eventual display on controllers’ screens.

When digitizers fail, the data appears as a meaningless scramble, forcing controllers to switch to alternate radar.

There are those who say the digitizers fail more often than any other component in the air traffic control system. “The digitizers may be the weakest link of all,” Bob Cox said.

But there are others who say the weakest link is the radio systems that controllers use to communicate with pilots.

Howard Johannssen, president of the Professional Airways Systems Specialists, a union representing most of the FAA’s repair technicians, says most of the radio gear is 1960s vintage, with some of it dating back to World War II.

Richard Cox, the FAA’s traffic manager at L.A. TRACON, said this aging radio gear fails almost every day.

“Sometimes a controller will start to transmit and find he’s no longer transmitting,” Richard Cox said. “Sometimes a controller will start to receive and find he’s no longer receiving. . . .

“There are standby transmitters and receivers. Sometimes they function, sometimes they don’t. When those fail, there’s one other backup (called a Back Up Emergency Communication, or BUEC), which is essentially a portable radio. . . . BUEC is like an inflatable spare. It meets your bare needs, but you don’t want to go too long on it.”

Some controllers at Palmdale who relied too long on the portable system had it fail, too.

“Every few months, we lose it all,” said Dave Alford, who has worked at the center for the last four years. “We had as much as four or five minutes with no radio at all. . . .

“What happens then is that pilots go back to a previous frequency they were talking on and tell the controller on that frequency what happened,” Alford said. “That controller contacts us (usually via telephone) and we tell him what to tell the pilot.”

Not Quite a Failure

Like the radar systems and computers, the radios sometimes don’t fail to work, they just fail to work properly.

“Some radios have areas of poor reception--little dead zones,” Allec said. “Sometimes you have to wait until the guy’s gone another 20 or 30 miles before you can talk to him.”

“For a while, when we were talking to planes over Ontario on the approach to LAX, we were getting a taxi company,” Alford said. “We’d hear things like, ‘Pick up the woman with the shopping bags in front of the market.”

There is little argument that the tired radar systems, the overloaded computer systems and the balky radios that make up so much of the FAA’s crucial air traffic control equipment all need to be upgraded or replaced. But that’s no easy task.

For one thing, there are about 23,000 individual pieces of radar, computer and radio equipment in the air traffic control system, and each must be compatible with whatever new gear is introduced--and be able to keep performing its intended function around the clock during the replacement process.

“It’s like painting a moving train,” said Steve Hayes, an assistant administrator at the FAA headquarters in Washington.

Recognizing the magnitude of the task, the FAA initiated a comprehensive National Airspace System plan in 1981 for modernizing and improving air traffic control and airways facilities services.

The original projected cost was $11.5 billion over a 10-year period, but Martin Pozesky, the FAA official in charge of the plan, now estimates that the bottom line will be closer to $15 billion.

Anticipating a 68% increase in aircraft operations between 1982 and the year 2000, the plan calls for the upgrading or replacement of almost all of the radar, computer and radio equipment in the air traffic control system with more modern gear--22,000 new pieces in all. The new equipment promises to work better and, because of its solid-state electronics, promises to break down less and require less maintenance.

The question is, when will it be in the field?

Even FAA Administrator Donald F. Engen admits that the plan “is not moving as fast as it can or it should.”

There are technological problems, to be sure. Problems with things like the software for the new computers that will replace the 9020s at the control centers. Problems with things like the new digitizer equipment to replace the weak links at the radar antenna sites.

But Engen, and a lot of other people, say that the basic problem is money.

It isn’t as though the money isn’t there.

The Aviation Trust Fund, established in 1970 to promote aviation safety and financed by a continuing inflow from taxes on airline passengers and airfreight, currently has a surplus of between $5 billion and $8 billion (depending on who is counting), with an annual income of about $1 billion a year.

The problem is that the money isn’t being spent.

President Reagan’s Office of Management and Budget says the money is being held up because of the technological problems.

But people like Rep. Norman Y. Mineta (D-Calif.), chairman of the House subcommittee on aviation, and Tom Tripp, an official with the airline industry’s Air Transport Assn., say the budget office is holding up the money because it helps to offset the Administration’s record budget deficit.

Administration Criticized

“The dollars are in place,” Mineta said. “But frankly, the Administration isn’t doing a damned thing.”

Engen, a member of that Administration, explained his position:

“The aviation trust fund should be spent to support the aviation system,” he said. “The monies have not been expended as rapidly as I would have desired. . . .

“But I don’t want to say that I feel shortchanged, because I think that would be disloyal to my President. I feel that I have had ample opportunity to explain the need for expenditure of that money. And I feel that the President, in his wisdom, has provided that which could be provided. . . . The President’s budget must meld together all the requirements of the United States. . . .

“You know, you’ve got to deal with shortages. . . . Our job is to put priorities where they belong and keep the important things moving forward. I am fighting very hard to keep the NAS (National Airspace System) plan moving forward. . . .

“I won’t deny that we have delays. But that doesn’t mean that we’re not going to get there, that we’re not going to get it done. We will.”

If equipment is one major problem Engen is facing as the top official of the FAA, staffing is another.

“The growth in air traffic activity has caused the controller workload to reach a point where controllers are stretched too thin,” J. Dexter Peach, director of the General Accounting Office (GAO), wrote to Transportation Secretary Elizabeth Dole last March after his staff had completed a yearlong study of FAA staffing.

The staffing troubles had begun in 1981, when 11,500 members of the Professional Air Traffic Controllers union went on strike.

Before the strike, the FAA had 16,200 air traffic controllers, of whom 13,200, or 80%, were “full performance level” personnel qualified to work any of the positions at an FAA control facility. After Reagan fired those who walked out, only about 3,400 of those highly qualified workers remained.

In a report to the House subcommittee on investigations and oversight last March, Herbert R. McClure, a deputy director of the General Accounting Office, said that after the strike the FAA set out to rebuild a staff that would have several thousand fewer controllers. The FAA had concluded that the work force could be streamlined despite a steady increase in air traffic.

The General Accounting Office report noted that for September, 1985, the FAA had set a system-wide goal of 12,500 controllers, of whom 9,375, or 75%, would be full performance level. As of that date, there were 12,500 controllers, but only 8,300, or about 66%, were full performance level.

In his report to the subcommittee, McClure discussed a recently completed General Accounting Office survey of 4,500 controllers, 1,000 first-line supervisors and the managers of all 20 of the nation’s air route control centers. Among the findings:

- “The FAA does not have as many fully qualified, experienced controllers at major air traffic control centers as managers, supervisors and controllers believe are needed.”

- “Air traffic activity has reached record levels and is at the point where . . . 70% of the controllers who work radar believe they are handling more traffic during peak periods than they should be handling.” About 50% of the controllers believe they have to spend too much time during a shift at a radar position. About 42% of the controllers said they were working more overtime than they wanted.

- “There are fairly widespread communications and other employee/management problems . . . that are adversely affecting the morale of controllers and (first-line) supervisors.”

- “The FAA faces some difficult obstacles in building toward its staffing goals. First, it takes time for a controller to qualify as an FPL (full performance level). Second, training attrition is higher than expected. Third, many of the experienced FPLs and supervisors have retired or are approaching retirement.” While the FAA expects that only 14% of those who are eligible to retire will actually do so, more than 80% of the controllers and supervisors scheduled for retirement within two years said they will retire when eligible.

McClure concluded his summation with the note that “most of the controllers, supervisors and managers who answered our questionnaires rated the overall safety of the air traffic control system as adequate.” But he also noted that the supervisors and controllers “also had serious concerns about their ability to maintain system safety.”

The findings in the General Accounting Office report reflected closely the comments of individual controllers and supervisors around Los Angeles.

“There aren’t enough people, not nearly enough,” said Kathy Heet, who has worked as a controller at Palmdale for more than three years.

“There were only two weeks this year that I didn’t have to work six days, and we don’t get all our breaks,” she said. “You get stressed and tired. You start cussing at everybody when you’re off the mike. . . . Morale is bad, definitely bad. Nobody is happy with management.”

“Our experience level is way down,” said Allec, a veteran Palmdale controller who was one of the relative few not to join the 1981 strike. “The new controllers just don’t have the flexibility. If something goes wrong, they may not be able to handle it, experience-wise.”

Larry Suppan, a first-level supervisor at the LAX tower, said he will be eligible to retire in February, and that is when he plans to go. “A lot of my friends are eligible, too,” he said. “Most of them plan to retire as soon as they can.”

But all the controllers and supervisors interviewed agreed that, basically, they like their jobs.

There are things like the special secrets they share with the government.

“We know every time a missile is fired from Vandenberg,” Whalen said. “We know how fast the SR-71 (America’s high-performance spy plane) goes. We know about it, but we can’t tell you.”

Prestige Factor

There are things like the special camaraderie, the respect and even awe some said they are afforded when they mention their jobs to outsiders.

Grundmann, who worked control facilities at Sacramento, Leemore and Burbank before moving to L.A. TRACON, seemed to speak for most of them:

“I love it,” he said. “There’s a certain air, a mystique. . . .

“You can’t just be adequate, you have to excel. You’ve got to know that you’re good, that you’re going to win. How else could you play the ultimate Atari game?”

When asked about the controller staffing situation, FAA chief Engen emphasized the sort of thing Grundmann was talking about.

“Controllers have job satisfaction,” he said. “People are pleased, they’re challenged, they’re enthusiastic. You walk around and you see happy people. . . . I don’t see that many retiring. . . .

“But people did feel that their supervisors were authoritarian,” Engen conceded. “People were dissatisfied with that.” He said the FAA is conducting a series of seminars for management, “emphasizing a continuing change in attitude.”

The FAA chief acknowledged that at some facilities where staffing is low the workload can be a problem.

“Does that mean that it’s unsafe? No, sir.” he told the subcommittee on investigations and oversight. “Does that mean there’s more overtime work than they want or I want? Yes, sir.”

But Engen said that overtime has been dropping over the last two years, with workload and hours now “within safely manageable limits.”

Engen said he is making a concentrated effort to increase the number of full performance level controllers by using accelerated instruction schedules and offering incentives to controllers who move from some of the more coveted assignments (in places like Des Moines, where the workload and cost of living are relatively low) to control facilities that are harder to staff (places like New York, where the pace is hectic and the cost of living is high).

He said the present failure to meet a goal of 75% staffing with full performance level controllers does not mean that unqualified people are manning radar screens, it just means a loss of flexibility because some of the uncertified controllers can work only one position.

Engen said he expects to meet his goal of a total controller work force of 15,000 by the end of fiscal 1987. He said he is not considering proposals that he rehire some of the fired PATCO members.

While the General Accounting Office study of controller staffing did not include the FAA’s maintenance technicians--sometimes referred to as “service specialists”--union sources indicated that there are major staffing problems there, too.

“We are gravely concerned for the continued safety of the flying public and are frustrated that it has been jeopardized by FAA mismanagement,” PASS union president Johannssen said in noting that the FAA has used attrition to trim the ranks of maintenance technicians from about 11,000 in 1979 to about 6,000 today.

“There was a belief that a change to solid-state electronics would permit a drop in manning,” said Mark E. Schneider, executive vice president of PASS, which represents about a third of the technicians. “There were some savings, but nowhere near what was projected. . . . The result is that they’re critically understaffed today. . . .

A Matter of Maintenance

“We’ve gone from preventive maintenance done when it should be done to corrective maintenance done after a system fails,” he said.

Fellow union chief John F. Thornton, national coordinator of the NATCA, the organization seeking to represent the air traffic controllers, offered Johannssen support during testimony before the House subcommittee on aviation last June.

“In the past, the controller has had an adequate number of service specialists to call upon in case of equipment malfunctions,” Thornton said. “This is not the case today. . . . Signs of inadequate staffing and overwork are rampant . . . and morale is at rock bottom. Quite expectedly, many systems specialists are looking to leave government service.”

Palmdale controller Allec and L.A. TRACON controller Grundmann agreed with Beam that the maintenance cutbacks are a mistake.

“Sometimes, when things break, you have to call the guy at home,” Grundmann said. “If he’s not there, we have to do without.”

Engen, placing some of the blame on the policies of his predecessors and the constraints on spending, conceded that “we need to bring on more maintenance personnel.” He said that while solid-state electronics and automated monitoring of remote facilities “let us do a better job with fewer people, we are below the level that we should be. . . .

“My goal this fiscal year is to replace on a one-for-one basis the people who are leaving the system,” he said. “And then, in fiscal ‘88, to begin to build back to where we should be.”