The Space Reviewin association with SpaceNews
 

 
Starliner
The Boeing CST-100 Starliner after landing in New Mexico, its test flight cut short by a timer problem. (credit: NASA/Bill Ingalls)

It’s all a matter of timing


Bookmark and Share

The timing problem that caused the partial failure of the Boeing Starliner’s first mission represents yet another space launch when bad timing—or at least bad synchronization of timing—occurred.

People were just putting down their coffee and getting ready to start taking pictures when the Thor suddenly lifted off, many seconds early compared to what they thought the count was.

On the first SM-75 Thor IRBM to be launched operationally at Vandenberg Air Force Base, someone forgot to cut the wire that secured the 35mm tape that served to provide electromechanical timing for the flight. As far as the booster was concerned, the plus count never got past T-0. The booster never entered its pitch and roll programs and climbed straight up until the Range Safety Officer at Vandenberg decided he’d better send the destruct signal.

On the Atlas 19F/NOAA-B mission, a partial engine failure resulted in a long delay in the booster attaining the proper separation conditions, resulting in the autonomous TIROS spacecraft attempting to separate while the Atlas booster was still thrusting (See “Launch failures: engine out”, The Space Review, December 31, 2012). Unable to separate from the booster and pitch to the proper attitude, the spacecraft entered an essentially useless elliptical orbit. In this case the spacecraft was not designed to receive a separation signal from the booster but just figure out the proper separation conditions based on its own accelerometer, and as a protection against a failure of that component a timer set for what seemed like a point well beyond what could ever be expected would initiate separation.

The Thor booster had a rather unique countdown, where after T-0 was reached, another 11 second countdown was started as soon as full LOX topping was indicated. So it was “3… 2… 1” and then people looked around each other, often a bit confused, until a big yellow “bug” light illuminated and the 11 second countdown started.

On the first Thor LV-2F DMSP Block 5D-1 mission to be launched from Vandenberg, the booster countdown got out of sequence with the countdown the Western Test Range was using. As a result, people were just putting down their coffee and getting ready to start taking pictures when the Thor suddenly lifted off, many seconds early compared to what they thought the count was.

The reported 11-hour mismatch between the actual and vehicle-perceived T-Plus count for the Starliner mission may set a new record, at least for the size of the error if not also for the impact on the mission.

On one of the Atlas Agena missions to be launched from SLC-3 at Vandenberg, the Atlas engines shut down due to a problem at T-0, but the Agena upper stage still started its own plus count and, after five minutes or so ,would fire its own rocket engine. This would have destroyed not on the Atlas and Agena but probably SLC-3 as well. Fortunately, the Agena was not only an upper stage but also a spacecraft and was equipped to receive commands from the Air Force Satellite Control Network. The Vandenberg ground station sent a reset command every five minutes, restoring the Agena to T-0 conditions, until the batteries in the spacecraft ran down and the vehicle could be rendered safe.

At Vandenberg in the 1970s, a BOMARC target vehicle lifted off early. Unable to determine if the launch was proper, the Missile Flight Control Officer sent the destruct signal and destroyed the vehicle.

At a Scout launch at Vandenberg in 1987, there was a problem with the booster’s range safety system and the countdown effectively entered a hold without anyone being told, including the Missile Flight Control Officer and the Range Control Officer. After firmly asserting that the launch window was only five minutes at the Launch Readiness Review, the US Navy Mission Director suddenly decided it could be extended to ten minutes. But as the final seconds of the extended window approached and a scrub seemed to be obvious, the Launch Controller suddenly announced that the count had been advanced from T-5 Min to T-10 seconds. The Range Control Officer reacted with a spasm of contacting range assets, and the launch occurred in the final seconds of the extended window. The situation was ripe for another MFCO intervention to destroy a loose bird, but they managed not to blow the vehicle. This led to some extensive discussions between the Air Force and NASA over the need to let the launch range know what was going on.

Having said all that, the reported 11-hour mismatch between the actual and vehicle-perceived T-Plus count for the Starliner mission may set a new record, at least for the size of the error if not also for the impact on the mission.


Note: we are temporarily moderating all comments submitted to deal with a surge in spam.

Home