Viewing page 5 of 86

This transcription has been completed. Contact us with corrections.

Save NASA from its next mistake
by Gregg Easterbrook

Easterbrook, based in Washington, is a national correspondent for the Atlantic magazine.

THOUGH RECENT MONTHS have been unusually bad ones for NASA and the Air Force, the "rash" of rocket explosions is really a rash of coverage of rocket explosions. Rockets fail all the time, and have been failing for years.

When a Titan 34D blew itself to fragments over Northern California last August, for example, it barely merited a wire service blip, to say nothing of the holy-cow headline treatment given to the Delta failure on May 2.

Deltas, commonly called "reliable" in news accounts, have blown up or gone haywire 12 times in 178 launchings, for a 7 percent failure rate. Titan launchings have produced six failures in 136 attempts, a 4.5 percent rate. Twenty-six firings of the space shuttle led to one explosion, or 4 percent failure. Other rockets in the U.S. inventory, including strategic missiles such as Minuteman, blow up between 1 percent (Thor) and 10 percent of the time (Atlas).

The elevation of rocket failures to the status of media events has led to considerable speculation about whether NASA is slipping, contractors are snoozing and so forth.  Such charges may in fact be true, and if true merit serious action. But the millions of lines written and spoken about rocketry in the wake of Challenger seem to miss the point. Which is: Rockets cannot be made foolproof. Some will fail.

Looking back, it appears fortunate that a shuttle accident didn't happen sooner. Of the 24 missions flown before Challenger, plus one aborted on the pad, there were two main engine shutdowns; one near-burnthrough of a solid rocket booster nozzle, and one failure of the first of two SRB seals. This means four of 25 firings (16 percent) came within seconds of tragedy. Several other flights experienced several seal degradation or other conditions which, while not threatening to the mission, suggested that if the shuttles were repeatedly reused according to plan, failure would eventually occur.

Why should we expect otherwise? A percentage of conventional airplanes fail and crash — as a Boeing 747 did in Japan last year. These crashes occur despite the fact that airplanes operate under conditions only a fraction as stressful as spacecraft. Even military aircraft, where cost is no object, sometimes crash under routine peacetime circumstances.

Yet the United States built an incredibly costly space-shuttle program on the premise that the system would never have a fatal failure. Now the Reagan administration is on the verge of endorsing construction of a $3 billion replacement shuttle — which by the time it is ready to fly will represent a 20-year-old design — on the premise that there will never be another failure.

Whenever there occurs a national upset, Washington officialdom feels it must act tough by immediately spending a huge amount of money. Why not act smart instead? The unlikelihood of building a failure-proof rocket means that manned ventures into space should be limited to those situations when people are truly needed. There are such times; just not as many as the manned space lobby would like. Routine satellite launches are definitely not among them.

Most frightening, early plans show the X31 spaceplane, a NASA-Pentagon project, repeating the shuttle program's central flaw.

Spaceplanes may end up even bigger than the shuttle — plus costlier, more complex and again premised on the notion that they can be made immune to failure. This is a budget justifier's illusion for which there is no real world precedent.

Approximately $450 million was awarded to seven defense contractors last month for spaceplane research, the goal being a vehicle that can take off from a runway, rise into orbit in one piece and return to a runway. Spaceplanes would be propelled by a hypothetical engine called a scramjet, a 

see SAVE on Page 4

She defined 'woman'
Page 4.
[[image]]

Continued from Page 1.
Save NASA from its next mistake

cross between a jet and a rocket.  From an abstract standpoint this is an excellent goal, one that might, at last, deliver what the shuttle promised: low-cost access to space. But as now conceived the X31 may be a greater fiasco.

Spaceplane problems are:
• Unquestioning adherence to the biggest-is-best philosophy. Some early spaceplane design contenders are even bigger than the shuttle and likely to be the most expensive vessels ever built. This means the country will be able to afford only a few. So we are right back into the shuttle syndrome; gambling a vast amount of money, precious human lives and a large share of national prestige on even the most mundane mission.

• The bland assumption that scramjets, which have never powered a flying machine, will quickly become reliable. Scramjets have been subjected to only partial testing and then under laboratory conditions — about as similar to flight as hair styling is to brain surgery. For contractors to tacitly suggest they represent a known technology, as some are, represents something close to fraud. The shuttle program took large risks by pushing solid rockets and cryogenic hydrogen engines to unexplored performance levels, but at least there had been hundreds of firings of lesser engines using the same principles.

• There's little if any logical answer to what spaceplanes would do. Military planners know they make no sense as bombers or fighters — missiles can deliver warheads at a fraction of the expense; once a spaceplane re-enters the atmosphere it will be a lumbering sitting duck that a World War II fighter could destroy. Spaceplanes, planners are therefore saying, would be good for reconnaissance.

In aerospace procurement, reconnaissance is the last refuge of scoundrels. During the 1960s, the Air Force tried to revive its canceled XB70 supersonic bomber as a reconnaissance vehicle; its current SR71 spy plane began life as the YF12 superfighter, which was supposed to revolutionize air-to-air combat but turned out to be too expensive and fragile for fighting. That spaceplane proponents are already resorting to the reconnaissance rationalization indicates the paucity of realistic justifications for a big aerospace machine.

Behind spaceplane planning is the same mutually corrupting trio of interests that distorted shuttle planning 15 years ago:

• Jobs for defense contractors — spaceplanes as now conceived will be far more expensive than possible alternatives.

• Empire-building for NASA and the Air Force.

• Glamour for everybody else in the self-serving Washington swirl — the president makes dramatic announcements about projects paid for by somebody else's administration; congressmen steer hefty contracts, the media get great footage and features on astronauts.

The only losers are taxpayers and common sense, neither of which has influential Washington lobbies. Those, and ultimately the valid scientific and military uses of space.

The corruption here is not in the money sense. It's something worse: institutional self-deceit.

Direct cash corruption is rare in U.S. government, considering the vast sums of money involved. Corruption through institutional self-deceit is, on the other hand, rampant. When government agencies, corporations and other large institutions start believing their own sales pitches, results can be costly.

Direct corruption usually involves only a few individuals who can be rooted out. Institutional self-deceit often involves thousands of people, no one of whom has done anything legally wrong but all of whom have a considerable stake in averting their eyes. Finally, where direct corruption is found, the guilty can be punished by law. Institutional corruption at the national level can be stopped only be a president or congressmen willing to take on interest groups — not exactly the most common personality types in Washington.

How else can we explain the non-reaction to Ronald Reagan's renomination of James C. Fletcher to head NASA? Fletcher headed the agency from 1971 to 1977, when the shuttle was designed and sold. More than any other person, he is responsible for its conceptual flaws and engineering defects.

Fletcher promised Congress each shuttle launch would cost $10.4 million. They're $151 million, or six times as much adjusted for inflation. Fletcher promised Congress shuttles would fly 60 times a year; Richard H. Truly, the new program director, now says the optimistic goal is 15. Anyone can make an honest mistake, but Fletcher's promises about two key facts of shuttle use turned out to be not even in the general vicinity.

At his confirmation hearings Fletcher almost taunted senators who tried to question him about the estimates — saying he had forgotten how they were arrived at, then adding with a smug smile, "Something must have happened on the way to the bank."

What was amazing about the testimony was not so much Fletcher's evasions — standard operating procedure — but how the Senate rolled over and played dead in the face of them. The Commerce Committee, which has jurisdiction over NASA, recommended the Fletcher nomination to the floor with just one dissenting vote, from Albert Gore Jr. D-Tenn.

That the same senators who two months before had been screaming bloody murder over the condition of the shuttle program voted blithely for the return of its mastermind shows Congress, too, is deep in institutional self-deceit. The essence of the exercise is to keep the money flowing but the blame fixed elsewhere. Now, if something else goes wrong, then the White House and the bureaucrats are to blame.

It may be too late to alter the shuttle in any fundamental way: From a policy standpoint the best approach may be to wind the program down. There is time, however, to make sense of the spaceplane.

Safer, efficient access to space may be obtainable using a small, sensible spaceplane designed to transport people on missions when they are actually needed. U.S. technology buffs will chaff at the idea of building something small when we could be sinking half the federal budget into an interplanetary office building.

But a small, sensible spaceplane may eventually provide enough flight experience with scramjets to build a cargo spaceplane in the 21st century. In the meantime, it should be coupled with a new generation of expendable rockets for launching satellites, cargo and hardware that crews aboard the spaceplane could use. New expendable rockets should be built along "minimum cost" engineering principles, rather than minimum weight. This promising design approach was thrown out some 15 years ago with the decision to sink all available funds into the shuttle.

If congressmen and journalists repeat their shuttle performance on the spaceplane project — standing on the sidelines mumbling "golly gosh," then after the fact crying, "Why didn't somebody do something about this years ago?" — the next form of space transportation may become as badly flawed as the current system. And that, while serving many institutional interests in the short run, in the long run serves no one.