Chair Force Engineer

Wednesday, October 22, 2008

Morale Stops Here



Michael Griffin blames anonymous weblogs, critical of Project Constellation, for harming the morale of NASA engineers. Further, he claims that engineering disagreements between NASA insiders and the weblogs have escalated into personal vendettas. The message of the strawman argument is that bloggers should just shut their mouths and let the big boys get on with the serious business of Project Constellation.

I don't know if Mike Griffin has ever known how it feels to be completely demoralized, but I feel demoralized on a daily basis. I have also seen firsthand the ways that professional disagreements over engineering judgments have turned into passive-aggressive behavior in the workplace. There is no virtue in schadenfreude, and I certainly don't wish those things upon anybody. With that being said, none of my experience can be attributed to anything I've read in a weblog (although there's plenty of intelligent criticism to go around.)

In my years as a critic of Mike Griffin's NASA, I have never intended to criticize the engineers who have been tasked to make this plan a reality. They're doing the best they can with the plan they were handed by their management. The engineering staff at NASA-Marshall and the other NASA centers who are working Project Constellation are putting in long days and making great personal sacrifices in order to ensure the success of the Constellation program.

While my critics may disagree, I have no personal vendetta against Mike Griffin, either. I don't think he's done anything illegal, immoral, or scandalous. I'm convinced that he believes 100% in Project Constellation as he's implemented it. The problem is one of vision, and one of pride. Mike Griffin had his own vision for how to implement the Vision for Space Exploration, and his agency rushed a 60-day study based on multiple flawed assumptions. While the Griffin vision is now the law of the land, many of us are disappointed. We want to believe in things like afforability and sustainability. We haven't seen any evidence that "Apollo on Steroids" will be able to avoid the same fate as its namesake.

NASA's engineers are smart people, and they're plenty capable of making their own value judgements about the programs they work on. They don't need anybody's blog to help them make an educated decision. I want NASA’s engineers to be happy. If working on Ares makes them happy, they should keep plugging away. If they can’t stand the work they’re doing, they should try to get reassigned, or find a job outside of NASA that's more rewarding.

Along similar lines, I have made my own value judgments about the worthless nature of the work I have performed for the US Air Force. This miserable experience has soured me towards the engineering profession, the aerospace industry, and government bureaucracy. Next August I bid the Air Force "good riddance" and look for a non-engineering job that promises a rewarding experience of immediate benefit to humanity.

Over the last few months, I have kept a lid on criticism of the Ares I configuration, which should make Mike Griffin and his engineers happy. Fighting The Shaft is futile, as Ares I has progressed far enough where it will probably survive into the next administration. The argument should not be over whether it should be killed, but on ways to make it as safe as possible. But ending all criticism of the Griffin plan won’t fix the problem of NASA morale.

The morale of engineers is directly tied to the work they are given by their management. If you want to keep engineers happy, give them tasks that are worthy of their efforts. When management fails to do that, they have nobody to blame for poor morale but themselves.

Tuesday, October 14, 2008

Nova fizzles

I watched NOVA's "Space Shuttle Disaster" episode on PBS tonight, and I have to say that I was disappointed. It really didn't say anything that hasn't been said before, and it focused more on the cultural and managerial aspects of the disaster, rather than the technical side.

The problem with documentaries of this sort is that it's difficult to come up with something original to say when so many other documentaries have been made, and when the news media has covered so many facets of a highly-visible event like the Columbia disaster. I recall a program on the National Geographic channel from 2005 which covered the Columbia disaster in what I felt was a better treatment than the more recent one from Nova.

I will say that my initial reaction to the foamstrike theory, publicly announced later in the day when the disaster first occurred, was similar to that of many NASA managers. Debris strikes and over-pressures had knocked off tiles on previous shuttle missions, so why would a foamstrike be enough to doom Columbia? It turns out that this was the mother of all foamstrikes, targeting the orbiter's critical and brittle leading edges. NASA is lucky that they have better folks than I to make the life-and-death decisions that affect the crew. Even still, we are reminded that this is serious business with little room for error.

The more morbid parts of my mind focus on what the re-entry must have been like for the slowly-disintegrating orbiter, and for the crew. I wonder how much the accident investigators know about exactly what was going on within the wing structure as the plasma blowtorch intensified. I can only imagine the terror the crew must have felt during those final moments of radio silence, as the left wing disintegrated and the vehicle spun out of control before its ultimate disintegration. There can be no underestimating the bravery of men and women who knowingly choose to ride the Space Shuttle.

Sunday, October 12, 2008

Mars: From Surface Lab to Sample Return

Rocket Man has an interesting piece about the problems encountered by the Mars Science Laboratory program, which threaten to end the next Mars science mission before it even gets off the ground.

When I first read about the SkyCrane descent method, it seemed pretty needlessly complex in my view. How long would the "flying bedstead" need to sustain a hover in order to lower and free the science rover? How do you ensure a clean separation as the "flying bedstead" flies away from its rover? The concept could work, but it adds a lot of risk to the mission that I don't think is necessary.

Instead of pursuing MSL, Rocket Man proposes spending its budget on built-to-print copies of the Spirit and Opportunity rovers. Such a move would give NASA-JPL a reliable means of investigating multiple sites of scientific interest on the Red Planet.

But I would take the idea one step further, and tie the "multiple rovers" concept into the long-anticipated mission to robotically return soil samples from Mars. While a single Sample Return mission would have scientific value, the expensive Sample-Return mission could provide a much greater science return if multiple areas on the Martian surface could be sampled.

Conceptually, using several copies of the Spirit/Opportunity rover to grab soil samples for return to earth is an excellent idea. The problem is getting all of those samples back to earth. At the pace the rovers travel, they'd have to be landed fairly close to the rocket that will lift off from the Martian surface to bring the samples back to earth. This would defeat the purpose of having multiple rovers for a sample return mission.

Alternately, each rover could be delivered to the Martian surface with its own ascent vehicle. Bigger ascent vehicles could return to earth independently, increasing the chances that at least one Martian soil sample will make it back to earth. A more complex design would use smaller ascent vehicles, and make them rendezvous with a mother-ship in Mars orbit.

In closing, the successes of the Spirit and Opportunity rovers in exploring Mars should serve as the basis for future exploration efforts. Unfortunately, Mars Science Lab discards many of these proven concepts and systems in favor of a high-risk SkyCrane approach. If SkyCrane does not work out, JPL should seriously consider a return to the Spirit/Opportunity rover design, and perform science missions with an armada of simple, cheap rovers instead of a single, expensive rover. The "constellation of rovers" approach to science missions may be the best way to ensure the success of Mars Sample Return.

Wednesday, October 01, 2008

"Stages to Saturn" and Lessons from Saturn

I recently finished reading Stages to Saturn, Roger Bilstein's authoritative tome on the development of the Saturn rockets. The book was first written in 1979, with the perspective that Saturn-like rockets would soon be replaced with reusable launchers like the upcoming Shuttle. Perhaps Mr. Bilstein was surprised to see that, by the 2003 edition, his book could again be referenced by the designers of a new generation of launchers.

Philosopher George Santayana has been memorialized for his famous aphorism, "Those who fail to learn from history are doomed to repeat it." Naturally this begs the question, "How has today's NASA learned from history?" The major lessons from Apollo can be debatable, and NASA's implementation of those lessons is even more open to controversy.

When the Saturn A-I was still on the drawing boards in late 1959, there was debate as to what the upper stage would be. The cluster-tank first stage was already set, but various ICBM's were being eyed to form the second stage. While the Titan I first stage was preferred, it was rejected for numerous reasons (including structural limitations.) In Spring 1960, the Silverstein Committee convinced Wernher von Braun and his team to bite the fiscal bullet and develop an all-new, hydrogen-fueled upper stage.

During the Saturn program, NASA faced opportunities where existing hardware could have been used in a sub-optimal manner, but they instead invested in high-risk, high-payoff technologies that made Saturn a success (particularly the massive F-1 engine and the J-2 engine, which was a major impetus for the development of hydrogen as a rocket fuel.) The Ares rockets are shuttle-flavored, but not exactly shuttle-derived. The existing Shuttle SRB and Space Shuttle Main Engine were rejected because they couldn't meet mission requirements. But it's debatable whether a solid fuel first stage is necessary to begin with.

The Saturn rockets became astounding successes because they had ample development time and budgets. The Saturn program had the luxury of investing in new engines like F-1 and J-2 and all the challenges that accompanied them. By contrast, Ares I is being developed on a shoestring budget until the Shuttle is retired in 2010. The conflicting demand is that Ares I be operational as soon as possible, to minimize the loss of space launch capability between shuttle retirement and Ares I first flight.

Another reason for Saturn's spectacular success was the inherent design conservatism of the team at Marshall who developed the moon rockets. Wernher von Braun was skeptical of the low mass estimates that were given for the Apollo spacecraft and lunar module at the start of the program. He discreetly had his engineers design to much higher performance margins. Sure enough, von Braun was vindicated as spacecraft mass grew, particularly in the lunar module. The structural margins in all of the stages were such that mass could be removed from the S-II stage in order to increase the Saturn V's payload.

In designing Ares I, NASA quickly forgot the guidance of von Braun. The mass budgets for Ares I are so tight that Lockheed Martin was forced to scale back Orion to a "zero base" vehicle and add redundancy back in as vehicle performance improved. While there's still some performance margin, the margin between vehicle performance and payload mass should be much higher at such an early stage in the program's life (currently past Preliminary Design Review, with delta-PDR's remaining on issues like thrust oscillation damping.)

One lesson from the Saturn staging process could have averted the failure of Falcon I Flight 3. During a Saturn launch, the vehicle was allowed to coast for a period of time following stage burnout. When the stages separated, a series of retrorockets would create additional spacing between the upper and lower stages before the upper stage ignited. My guess is that Falcon I's designers omitted the retrorockets to squeeze extra payload mass into the design; the consequence is that some reliability is sacrificed.

The use of common bulkheads was key to Saturn's mass savings on stages 2&3. In both cases, the design of the bulkheads gave the contractors significant manufacturing challenges. Common bulkheads were rejected for the S-IC because of the density difference between liquid oxygen and kerosene (although this didn't stop its use in Atlas.) Because of the temperature difference between hydrogen and oxygen propellants, the bulkhead consisted of two thin skins separated by a precisely-shaped layer of insulation. The segments of the skins had to be welded together from individual gores. NASA and Boeing will hopefully be able to apply the same techniques when manufacturing the common bulkhead for the Ares I upper stages.

A massive program like Saturn cannot survive without effective management. This became an issue during North American Aviation's manufacturing of the S-II stage, and it appeared that NAA was overextended between the S-II and Apollo spacecraft. This becomes a bigger problem for Project Constellation because there are fewer aerospace contractors left who can manage the production of a major component, like the spacecraft or a rocket stage. Will LockMart or Boeing be able to produce the Ares V core in addition to the Orion Spacecraft or Ares I upper stage? What about production of the Earth Departure Stage or Altair lander?

While Project Constellation is fairly young, logistics is a challenge that deserves serious consideration at this stage of the program. Apollo neglected logistics longer than it should have, but innovative solutions were found through the use of river barges and "Pregnant Guppies" for transport of large rocket stages. The logistics problem for Constellation takes on an added dimension because the Ares V core will be the largest rocket stage ever produced. I don't know if NASA has seriously looked into what it will take to produce the Ares V core and transport it from Michoud to the Cape. It's an open question of whether the existing facilities and vehicles are up to the challenge.

Most importantly, the Saturn rockets succeeded because of ample supplies of genius and luck. People like George Low, Werner von Braun, Sam Phillips, and George Mueller were instrumental to the program's success. One of von Braun's biggest virtues was not his own ideas, but his ability to support other people's ideas when they conflicted with his own (ideas like all-up testing and lunar orbit rendezvous.) Many of the leaders in the Apollo effort took large gambles; the biggest gamble of all was all-up testing for Saturn V. The fact that all three stages of Saturn V worked properly on the first launch (Apollo 4) is testament to the geniuses and methodical managers and meticulous technicians who got it correct on the first try. The same could be said about Apollo 8, launched around the moon prior to an earth-orbit test of the Apollo spacecraft aboard the Saturn V. Only time will tell if the same genius is currently at play with NASA.

Labels: , , , , ,