(Update: As a non-engineer, I missed an important point in the Titan II staging that made the original article appear a bit too finger-pointing, if not at least inaccurate. The corrected text will be shown in underscore. Thank you for one reader noting this. SB&G hates sensationalism and this can only be combatted with more facts. In this case, we simply looked both too closely at the event and missed an inherent design. The point of the article stands in the context of the safety of a manned spacecraft in proximity with two missions remaining without the fix used on a sister vehicle. I regret the error and appreciate that reader’s clarification. –Editor.)

The Titan II rocket family was a reliable series, with less, er, volatility than other rockets of the time. Of course, given the nature of rocketry, simply demonstrating that your rocket “explodes less” might not necessarily be a way to sell the vehicle.

The very first member of the family, the short-lived Titan I, used still-popular RP-1 and liquid oxygen fuels. But these didn’t work well in storage. And by “storage” we mean “in a missile silo, ready to launch in seconds in case of nuclear war.” Since you’d have to continually tap off cryogenic fuels like liquid oxygen, such missiles weren’t the best for rapid liftoff. The super-cold fuels would also cause structural problems over time.

Gemini_AC8
A diagram of Titan II for Gemini manned flight. (NASA)

In comes the Titan II. It used easily-stored hypergolic fuel and oxidizer, no refrigeration needed, which made engines simple to make since the fuel and oxidizer burned on contact. Upside: Very quick launch. Just pressurize and go. The downside is that such fuels require very, very careful handling since they were not only very explodey but so toxic that only a few parts per million would melt your lungs.

Titan II’s design essentially was a flying tank. The walls of the rocket body, like the Atlas, were also the fuel and oxidizer tank exteriors, with bulkheads inside to delineate the top and bottom. An all-aluminum design was shaved to the minimum thicknesses for strength while lightening the vehicle weight but without the drastic “balloon” tanking design that required Atlas SM-65 to be constantly pressurized, even when unfueled, to keep from simply collapsing under its own weight.

The Titan II was selected as the launch vehicle for the Gemini project.  Let’s zip to the Gemini 10 launch on July 18, 1966. Astronaut John Young commands on his second spaceflight, with Michael Collins enjoying his first mission.

The launch is perfect and Gemini 10 reaches orbit to rendezvous with their Agena target spacecraft.

In the mission report, NASA puts in a little addendum to the analysis of the Titan’ vehicle’s performance.

“Tracking films indicate that the Stage I oxidizer tank ruptured after the staging sequence was completed. The event had no detectable effect on the satisfactory operation of Stage II; however, further study is being conducted by the contractor and additional information will be provided in a supplemental report.”

Digging a little deeper, the agency noted:

“Post-staging event.- Motion picture tracking films indicate that an amber cloud appeared at approximately 1.2 seconds after BECO, followed by an unusual amount of debris. This evidence indicates that the Stage I oxidizer tank ruptured after a normal staging sequence. Telemetered data did not provide any evidence of the cause, and the event had no detectable effect on the satisfactory operation of Stage II.”

The first stage’s oxidizer tank rupture on Gemini 10 wasn’t the first or last time it happened in the program, nor in the Titan II family as a whole.

When NASA checked with the Air Force on their rocket history, they noted that their Titan ICBM counterparts demonstrated “several” occurrences of the rupture.

The timeline for Gemini 10, measured in seconds, shows booster engine cutoff at 158.38 seconds. Simultaneously, these events occurred:

  • Staging switches actuate
  • Signals from stage I rate gyro package to Flight Control System discontinued
  • Hydraulic switchover lockout
  • Telemetry ceases from stage I
  • Staging nuts detonate
  • Stage II engine ignition signal
  • Control system gain change

Stage separation begins at 152.44 seconds. Stage II engine ignition began at 152.64 seconds.

If we use BECO+1.2 seconds as the rupture time, the event occurred at 152.58 seconds, only .14 seconds after second stage began to separate. Second stage ignition occurred at the same time as BECO. Even counting the interstage, there was still very little distance separating the ruptured first stage and the departing second stage.

The reason behind the close proximity of the first and second stage? It’s by design. If you note this short video of a Titan II, you’ll see a stage separation. The second stage engine practically ignites at the same moment of separation. Vent holes about the interstage reduce, but doesn’t eliminate the kinetic energy of the event. The interstage’s thin aluminum goes scattering about as the stages part.

A reader noted that the simultaneous ignition with BECO is to keep the second stage propellants ready for use, avoiding the need for ullage rockets that settle the fuel in later rockets such as the Saturns.

While NASA would’ve surely known of the forces of separation and the interstage, something is more amiss in safety concerns.

While the interstage’s destruction was probably far away enough for safety and directional force, the energy that occasionally blew the oxidizer dome was still statistically risky. Changes would be added in the Titan III vehicle’s dome (particularly with the context of man-rating the vehicle for flying the Dyna-Soar or Manned Orbiting Laboratory) but none were done for the Titan II for two remaining missions.

NASA treated the overall condition as something they and the astronauts could live with, since the ruptures did not apparently affect the second stage.

Gemini 12‘s launch vehicle also showed a first stage oxidizer tank rupture after staging.

No matter. The Gemini program was over with Gemini 12 and NASA focused in earnest on Apollo.

As with Gemini, the new Command Module would use a pure-oxygen atmosphere. Worried that the smaller hatch would not be as secure, an inward-opening hatch was developed.

You might figure that we’re pointing out how NASA lacked a serious need to be proactive to the worst possible complications of vehicle and spacecraft as early as Gemini. As little as one-quarter of a second separated Gemini 10 from the potential of a very bad day.

What would have happened to a mission where the first stage tank rupture threw debris at or into the second stage engine or structure? Nothing good, obviously.

In the worst case, debris from the first stage oxidizer could have reached the second stage, damaging its fuel tank or causing engine damage. In either scenario, the Gemini spacecraft would have to initiate abort mode II: Shutdown the second stage (if possible), separate from the equipment module and fire the retrorockets in the reentry module to push away. It’s not clear how fast that Gemini could escape in this scenario.

By the time of staging the spacecraft was far too high to use their ejection seats: Their maximum altitude was 45,000 ft (14 km). First stage separation occurred at 42 miles (68 km). Abort mode II is basically a matter of “riding out” the situation until retrorockets can or must be used.

Even in more ideal conditions, the ejection seat was questioned by astronauts. While the technology was already in use in fighter jets, NASA engineers did conduct extensive testing for launching them sideways, including the stabilizing ballute to stabilize the seat until reaching a safe parachute deploy altitude.

Want to read the report on the testing regimen? Here it is. It took eleven simulated pad tests to ensure that the dummies didn’t “die” and that the system worked satisfactorily. There were many, many more tests required to ensure that the system worked–especially opening the hatch sufficiently fast before ejection occurred.

But all ejection tests were made in boilerplate spacecraft (or no spacecraft) that were not using pressurized 100% oxygen cabins as used in all spacecraft to that date. Later, Apollo 1’s loss would emphasize how a fire in a pure-oxygen cabin would propagate, and how everything inside, including space suit Nylon, would be saturated with the gas.

Tom Stafford and Wally Schirra, in the second of three launch attempts of Gemini 6, could’ve been badly burned if they ejected when their Titan had a engine failure during an attempted liftoff. “Roman candles” are the words Stafford used himself  in retrospect of the possibilities in that scrubbed launch.

NASA dodged a bullet (or fragments similar to them) with two in-flight explosions on manned Gemini flights and not making any quick changes to reduce the danger to the final two flights. With similar lack of precaution or foresight that would forbid or reduce the quantity of any components with explosive or flammable potential to exist later in Apollo without failsafe, redundancy or safeguard, NASA would ultimately place one crew in a ground test and one on their way to the moon in great misfortune.

Sources: