A space shuttle main engine burns at 6,000 degrees F, but the outside of the nozzle remains cool to the touch. Prior to launch, sometimes it even frosts over.
The nozzle technology that allows a finger-width of ridged metal to contain and steer flames that would boil iron is just one of the scores of innovations designers came up with for the engines three decades ago.
Such advances were critical if NASA was going to realize its plans for a reusable space shuttle that, unlike the previous rockets, would not use its engines once and then drop them in the ocean.
Some of the others:
- A system that lets the engines be incrementally throttled up and down depending on the needs of the mission
- A hydrogen turbopump that spins 567 times a second with each 2” tall turbine blade generating 700 horsepower.
- A computer that runs 50 health checks on the engine every second using data from 200 sensors
- A system of pipes, or ducts, that withstand pressures as high as 7,000 pounds per square inch
- A main combustion chamber strong enough to contain the explosion of 970 pounds of oxygen and 162 pounds of hydrogen fuel every second, continuously for 8 1/2 minutes
- The only heavy-lift booster engine that continuously performs all the way from launch pad to orbit
- Engineering and materials that allow the engine to be reused multiple times
- A compact, efficient design that produces 8 times the thrust of a modern high performance jet engine per each pound of weight.
Added together, the innovations became a rocket engine that is more than 99.9 percent efficient, which means that almost all of its hydrogen and oxygen is used to create thrust. For comparison, an automobile engine is about a third as efficient, since most of its energy is created in the form of heat that does not turn the wheels.
"Everything in that engine is a whole science field," said Carlos Estrada, NASA's Main Propulsion Branch chief at Kennedy Space Center. "You look at the materials, you look at the components, you look at the way they designed that engine, how it's all designed for the different stages with the pump and pressures. I mean, every time you look at a component you have all these people with expertise in it."
Three main engines are used to launch a shuttle into orbit, along with help from a pair of solid-fueled boosters that separate two minutes after launch.
The advances did not come easily for designers who, working in the 1970s before computer-assisted design became commonplace, ran many of their calculations on slide rules and used judgments based on the experience they gained building massive engines for the Saturn V moon rocket.
Getting the start sequence correct alone took about a year of testing, fixing and more testing, said Dan Hausman, Pratt & Whitney Rocketdyne's site director at Kennedy. "We kept burning up the turbine blades, getting temperature spikes. Our analog models weren't that good with the start sequence. We had to figure out how to get it started because everything had an idiosyncrasy."
The idiosyncrasies he talks about are no small matter considering a single main engine creates more than four times the horsepower of the Hoover Dam.
When most people think of an engine, they usually picture a part of the engine called a bell or nozzle. It's the part that everyone sees at launch shooting flames and supersonic exhaust. Although a lot is happening inside the bell, it's one of the least active parts of the machine during launch. The real action is taking place in front of the engine bell in a maze of hidden machinery called the powerhead.
"The powerhead is the meat of the engine," said Stephen Prescott, a Pratt & Whitney Rocketdyne engineer specializing in the engine's turbopumps. "The nozzle is what's actually allowing us to gather the thrust, but the powerhead is what actually gives us the thrust."
The powerhead is home to four turbopumps, a robust computer controller and a network of ducts, wiring and valves designed to release 500,000 pounds of thrust without exploding. For as much power as it releases, the powerhead is not imposingly large. Standing above the nozzle in a workstand, the powerhead reaches about six feet from the floor. The high-pressure hydrogen turbopump, the strongest of the four, would fit on a desk.
"You run into some people who think it's easy," Hausman said. "Anybody who thinks it's easy doesn't understand it. Once you understand it, that's a marvel of engineering. It's a marvel that people can build it, and operate it and work it at the high reliability that we've done."
The first space shuttle main engine ignition took place well before Columbia lifted off on April 12, 1981, to inaugurate the space shuttle era. It happened in the mid-1970s at a concrete and steel test stand at NASA's Stennis Space Center in Mississippi where engineers and designers could put an engine through its paces without worrying about sacrificing a spacecraft and its payload if something went wrong.
And things went wrong, especially in the beginning. The liquid oxygen turbopump blew up. The hydrogen turbopump blades broke and exploded the whole thing. There was the occasional combustion instability, which is a polite way of saying the controlled exhaust thrust went out of control and, you guessed it, blew the engine up.
In fact, the first engine test Hausman saw in person at Stennis ended with a puff of black smoke and half the engine sitting at the bottom of the stand.
"There wasn't much left, it was all kind of a molten mass of dripping metal because when liquid oxygen eats metal, there's no evidence left because metal vaporizes," Hausman explained. "Twenty milliseconds, 40 milliseconds, 60 milliseconds, the engine's gone. Very fast."
It is that speed that keeps the shuttle engine's mechanics on their toes as they carefully evaluate every engine after a flight.
"We have a line from John Plowden, one our most senior engineers, that’s embedded in the DNA of everyone here: Never turn your back on a rocket engine," Prescott said. "Knowing what this engine can do to itself in a split second is what keeps us focused on knowing you can't just brush off something that you think is fine."
While spectacular malfunctions on the engines were a mark of the early part of the engine development, fixing them effectively and retesting over and over would become a hallmark of the main engine program.
"The key was test, test, test," Hausman said. "In the development program, the best learning we could ever do was have an engine blow up at Stennis, because we could find an issue and go fix it."
"Any part that flew here at Kennedy had a counterpart that operated twice as long at Stennis," Hausman said.
Stennis recorded 2,000 main engine test firings between 1975 and 1992. More firings, including flight certification tests for every engine used on a shuttle, took place until July 29, 2009, bringing the total to over 2,300 engine firings at that one facility.
Hausman credits the careful development work with setting up the engine to successfully cope with problems during a shuttle launch, though there were very few of those throughout the shuttle's 130-plus missions.
A shuttle mission has never failed because of the main engines, though there were a couple close calls. The first came in 1985, when one of Challenger's three main engines shut down during ascent, prompting the crew to fly to a lower orbit. The Spacelab mission still was successful and engineers traced the problem to one of the sensors on the engine that shut down.
A series of failures occurred during the launch of Columbia in 1999, when a pin broke loose inside the main combustion chamber and popped a couple tiny holes in three of the 1,080 hydrogen tubes in the nozzle. There also was a pair of short circuits in Columbia’s electrical system during ascent which resulted in a loss of electrical power on the primary channel to the engines. The redundant safety features designed into the engine allowed the controller to seamlessly transfer control to an alternate channel and continue on with the mission.
Eileen Collins commanded the flight and Columbia was able to reach orbit and deploy the groundbreaking Chandra X-ray observatory just as was planned.
Prescott was watching that launch and listening to the transmissions back and forth between controllers and the shuttle crew.
"We knew something had gone on, but we weren't quite sure just what had happened," Prescott said. "Eileen, that was the most perfect example of what kind of training those astronauts go through, because she was just so calm, cool and collected."
Engineers dove deeply into the engine after Columbia's return to find out what went wrong.
"That was pretty scary," Estrada said. "That was a big deal."
Another engine safety feature was demonstrated during Columbia’s third mission in 1982 when one of the orbiter’s three auxiliary power units shut down late into the launch, resulting in a loss of hydraulic power to one main engine. That engine’s backup electrical control system maintained control and performance until reaching orbit which was then followed by a fail-safe, pneumatically-actuated main engine cut-off.
Hausman said the redundant systems built into the engines paid off during those situations.
A great deal of effort and research went into developing the shuttle's main engines, but maintaining them and keeping them healthy during the shuttle's 30-year career has been equally advanced and careful.
"I came in around 1996, and to me it was amazing to see how much people needed to know to be able to manage such a piece of equipment," Estrada said.
The engines' overseers spend hours peering with one eye shut into a small borescope, basically a long, black, flexible fiber optic hose with a lens at one end and an eyepiece at the other. Doctors use them frequently to examine patients. The engineers are looking for anything amiss, whether it be a weld in one of the turbopump housings, a tiny hole in a pipe, unusual wear or erosion or something they've never seen before.
"Keeping the discipline of what you do and how you do it is critical," Estrada said.
It is repetitive and painstaking work that takes a full shift to complete on each major component of the engine. And that does not count all the other extensive inspections performed before an engine launches again.
"Pretty much everything on this engine is criticality one," Prescott said. "We can’t even so much as lose a fastener and not create a problem because we're pretty close to the limits on everything on this engine."
When the technicians find something amiss, no effort is too much to track it down and fix it.
"It's a three-dimensional puzzle that sometimes, like a Rubik's Cube, you don't even know you’re close to getting it together until all the sudden, the thing's solved in front of you," Prescott said. "We've been known to chase our tails trying to get just perfect alignment and before you know it, there it is, everything can go together."
Throughout the shuttle's operation, designers kept improving the machinery. The sensors were steadily improved to make them more robust, the powerhead was redesigned to reduce pressures inside the transfer tubes and smooth the fuel flow, and the main combustion chamber throat area was enlarged to de-rate the engine and add extra operating margin. The modified heat exchanger eliminated welds and was strengthened.
Perhaps one of the biggest changes came when additional robustness was designed into the high pressure turbo-machinery. Overall, these design changes resulted in an additional 700 pounds of weight, but increased safety by a factor of 3 over earlier configurations. A final engine upgrade was introduced in 2007 when the advanced health management system became active, providing an additional 23 percent safety improvement during ascent.
Why put so much effort into the engines? Hausman credits rocket pioneer and Saturn V developer Werner von Braun with detailing the argument:
"The gist of his discussion was, if you don't build the engine right, anything above it that you put your time and money in is a waste of your time because if you don't build this right, you're not getting into space," Hausman said.