Not Exactly Rocket Science

A while ago, I talked about NASA’s invention of landing methods for the Mercury program – what to do when finding a solution for an entirely unknown problem. Tied into the question of landing methods for NASA’s first manned program was the design of the capsule. The basic constraints were laid out fairly early on in the program. Mercury would use a ballistic design proposed by Langley engineer Maxime Faget and splashdown in the ocean. This was the simplest method. In returning from space, NASA was content to let gravity do most of the work. (Pictured, Mercury model makers Richard Altimus and Arthur Lohse with model finisher John Wilson. 1960.)

With the basic capsule design set, there remained smaller design questions needing answers. What ballistic design would fare best against the heat of reentry? Throughout the descent stage, would one ballistic shape have better inherent stability than another or would the astronaut have to control the capsule’s attitude all the way down? Once the capsule was in the ocean, would it float? If the astronaut had to get out of the capsule, would it still float with a hatch open? In the 1950s, NASA sought answers to these questions in an age before computer programs could immediately generate answers. And so they did the next best thing. They tested model capsules, each shape designated by a letter, and picked the best design through trial and error.

The final phase of a Mercury mission, the reentry and splashdown, happens quite quickly. If everything goes smoothly, reentry and descent manoeuvres happen in sequence. Everything starts with the retrorocket pack, a cluster of three small rockets that is strapped over the heat shield during flight. Firing the retrorockets initiates reentry by slowing the capsule enough to allow gravity to take over. As the capsule starts to fall, it meets increasingly thick atmosphere. The friction during descent produces significant heat – temperatures up to 2000 degrees Fahrenheit – that would melt the capsule if it weren’t for the heat shield. The beryllium heat shield is ablative; it burns off as it’s heated, protecting the capsule and astronaut. To slow the capsule more, two parachutes deploy – a smaller drogue chute first provides additional stability followed by the main chute. By the time the capsule hits the water, the chutes have slowed its descent to a comfortable 28 miles per hour. Just prior to impact, a landing bag deploys to provide an additional measure of shock absorption for the astronaut. (Above left, Mercury capsule shape “A”. Above right, Mercury capsule shape “B”. 1958.)

But there was the very real possibility that reentry, descent, and splashdown wouldn’t go perfectly according to plan. As such engineers tried to anticipate and find solutions to as many problems as they could imagine.

Because the retrorockets covered the all-important heat shield, there was some question as to whether the capsule could safely reenter the atmosphere with the retrorockets in place. To test the capsule’s reaction to a reentry with retrorockets in place, engineers at the Ames Research Centre simulated the mission phase using two miniature-scale model Mercury capsules. Both models were made of aluminium with a mini heat shield made of phenolic resin with glass fiber reinforcement to simulate the ablative material. On one model, a tiny steel retropack was fitted over the mini heat shield. (Pictured, a full-scale Mercury model with the retropack covering the heat shield. The straps holding the former in place are clearly visible.)

Both models were placed in turn in a wind tunnel fast enough to simulate reentry speeds. The friction caused the mini heat shield to burn, giving the engineers a visual measure of the effects the retropack would have on its protective qualities. In both cases the mini heatshield burned away and melted evenly, as did the steel retropack. The engineers could determine that leaving the retropack in place would have no appreciable effect on the heat shield during reentry.

Anticipating and addressing this possibility of a reentry with a retropack proved to be serendipitous – this saved NASA’s first orbital mission. During his Friendship 7 flight, John Glenn had a warning light indicating his landing bag was deployed. Mission control confirmed the data. This was potentially disastrous. The landing bag was stowed between the capsule base and the heatshield and deployed only after the heat shield had done its job. If Glenn’s landing bag were really deployed, his heat shield would be loose and potentially unable to protect him from the heat of reentry. Keeping the retropack in place would hold the heat shield in place – the retropack’s straps would cover the heat shield, attaching it to the capsule body. The straps would burn away, but not before the heat shield had performed its function. Glenn did manage a safe return to Earth with his retropack attached. The warning light proved to be faulty.

Another problem during the descent stage was the capsule’s stability. If for any reason the automatic system failed and the astronaut was unable to control the capsule’s attitude, it would be in free fall. If the design was unstable and the capsule tumbled as it fell, the build up of g-forces could be hazardous to the astronaut. With the blunt capsule design chosen for Mercury, multiple slightly different designs were tested to determine which shape would fare best during the descent stage.

One stability test used a spin tunnel – a sort of upright wind tunnel – to compare the inherent stability of four blunt body designs. Each was placed in a wind stream. An engineer held each model in a desired orientation before releasing it. The model could then ‘float’, mimicking its performance during reentry. Some designs were more stable than others, and some righted themselves more readily than others. Some tests had engineers put the models in the wind stream blunt end up to see which designs would naturally fall blunt end down. The heat shield over the blunt end, after all, was the most important part of the reentry phase. (Right, a Mercury model with a parachute in the spin tunnel.)

As a further measure of stability, Mercury engineers sought to upset the models as they ‘fell’ blunt end down. For this test, they whacked each model with a stick while it floated in the air stream. Some models bobbed while others started turning cartwheels over themselves. The test proved that Faget’s simple model was most reliable for an optimal fall from space.

Four possible capsule shapes tested for stability in a spin tunnel. (Sketches my own.)

Another test was a simple drop test. With Faget’s model as the most stable from pervious tests, a scale capsule was put through drop tests. A capsule with a weighted escape tower was dropped from a prescribed height – the effect of the weighted tower was to make the capsule fall blunt end up – the opposite way it was intended to fall. At a certain point in the fall, the capsule separated from the weighted escape tower. The test was to see if by design the capsule would right itself to fall blunt end down. A series of tests determined that Faget’s proposed design would right itself, albeit with significant oscillations. At least NASA could feel confident that the capsule would naturally fall heat shield first.

Images depicting another important stability test: the abort test. Shown here the capsule shape "D" is tested. 1959.


After the fiery reentry and at the end of the descent stage comes splashdown. Once in the water, however, there were new problems for Mercury engineers to deal with.

The original design for the Mercury capsule gave the astronaut no way to get out; the hatch could only be opened from the outside. He was meant to wait in the capsule for the recovery helicopter to take him to the prime recovery carrier. Only when he was on the ship could he be let out of the capsule.  But the astronauts were not too thrilled at the prospect of waiting passively in the capsule after splashdown. While this was more a matter of preference and pride than a technical challenge, happy astronauts were necessary to a successful program. And so the Mercury engineers redesigned the capsule to the give the astronauts a way out.

The original exit point from the capsule was in its neck, through the cargo space that housed the parachutes. Mercury used two separate parachute systems, one for automatic deployment and the other for manual. Separating them was a matter of safety – if one failed, it was in no way connected to the other. Both systems were packed in separate half-circle bowls that fit into the neck of the capsule. From the inside, the astronaut could push the empty parahute cargo space out, opening the capsule to the air. He could then squirm his way out through the semi-circle opening. (Left, astronaut Scott Carpenter practices egressing through the Mercury capsule’s neck. 1962.)

It wasn’t the simplest egress, but it was better than waiting in the tiny capsule. Scott Carpenter made good use of this egress method on his flight, choosing to wriggle out of his Aurora 7 capsule rather than wait the three hours for recovery forces. He had overshot his splashdown point by 200 nautical miles.

Having the capsule open in the water, however, opened it quite literally to the possibility of flooding and sinking. An egress through the capsule’s neck was only good if the capsule floated in such a way that it was safe. The capsule was designed to be somewhat buoyant, but what if the astronaut’s weight pushed the neck underwater long enough to start the capsule sinking? (Right, Gordon Cooper during egress training sits on the neck of a Mercury capsule. 1963.)

Mercury engineers devised a simple solution to the problem. They installed four balloons around the capsules neck. Once inflated after splashdown, they would forcibly keep the capsule floating at an angle making it impossible for the neck to become submerged. To ensure the balloons would support the capsule even with the weight of a full-grown man on the neck, the engineers developed a simple test. A full-scale mockup capsule was placed in a pool. One engineer would sit inside the cabin while another would jump on the neck trying his hardest to get it under the water.

The tests proved the concept of balloons as an appropriate flotation aid, but the method was never used. Instead, NASA opted to make use of the Naval resources at its disposal. Rather than include flotation aids in the capsule, the Navy SEALS on hand for recovery operations attached a flotation collar around the capsule once they get into the water. This was a custom-fitted raft that attached around the base of the capsule, make it more stable and giving the astronaut and frogmen a better platform to work from. (Pictured, Navy SEALS or Frogmen attach a flotation collar to Gordon Cooper’s Faith 7 capsule after splashdown. 1963.)

Later in the program, a second exit point was added. The hatch was redesigned to be an easy way out. It was fitted with explosive bolts. When the time came for the astronaut to exit the capsule, he needed arm the bolts and blow the hatch. It would fly away, giving the astronaut a much simpler way out. A trade solution, however, that almost drowned Gus Grissom after his Liberty Bell 7 flight.

While whacking models with sticks or jumping on a capsule are unconventional approaches to rocket science, it is a testament to the “make it happen” attitude synonymous with NASA in its infancy.

Selected Sources/Suggested Reading

Carpenter, Cooper, Glenn, Grissom, Schirra, Shepard, Slayton. We seven: By the astronauts themselves: Simon and Schuster. 1962.

James Grimwood. Project Mercury: A Chronology. Washington: NASA. 1963.

I also have some suggested viewing. “Project Mercury: A New Frontier” compiles archival footage from the tests with interviews from Mercury engineers.


4 thoughts on “Not Exactly Rocket Science

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s