NASA had yet to score a victory in the new manned space race on April 25, 1961, a little over a week before they’d get Al Shepard briefly at the edge of space.
But suborbital launches just paid their space-race entry fee. NASA needed to get man into orbit to show relative equality to the Soviets. That required a lot of test flights to tame a very temperamental ballistic missile for safely and properly putting that man in orbit and not doing what rockets often did when they didn’t want to go to space that day.
So as NASA engineers finished weeding out the bugs for the two Mercury-Redstone flights, the agency made unmanned test flights of the Atlas rocket, each flying a genuine Mercury spacecraft with test equipment or dummy astronauts.
Mercury-Atlas 1, intended to be a suborbital test, crunched like a beer can on ascent, the balloon-tank design of the rocket failing to hold the spacecraft’s weight under acceleration. After strengthening the rocket, Mercury-Atlas 2 completed the suborbital test. So now, it was time to get a test vehicle in orbit.
Mercury-Atlas 3 rose high into the sky.that April day. And climbed. And climbed, which was a problem.
Rockets should follow a ballistic pattern if they aren’t to reach orbit. Else, they follow an elliptical trajectory. It’s why you commonly hear “roll” commands during launches of all kinds. Suborbital or orbital, rockets are always supposed to pitch a little bit eastwardish from the Cape, going over the ocean.
Mercury-Atlas 3 kept going straight up, which is bad.
Eventually the rocket simply would lose the fight against gravity and things would drop back from whence it came, which would be bad for the ground crew, pad, and perhaps small towns around the Cape.
After 43 seconds of that frightful nonsense, the range safety officer told MA-3 to explode. Back to the drawing board for everybody on the Atlas tamer team–except the launch escape design team, who might have gone off to have a beer.
As MA-3 broke up, the launch escape system detected the failure and pulled the test Mercury spacecraft to safety, landing it safely off the coast.
This was the very first Mercury test with a live escape tower and it worked flawlessly.
NASA had to take any little success they could back then. This was their second (unintentional) successful launch escape tower test. At least this tower left from a moving rocket.
Things would get better, of course. Mercury-Atlas 4 and 5 tests were very successful, clearing the path for Friendship 7 and John Glenn’s historic flight, a few months later in February 1962.