In a previous thread, I asked the question of the likelihood of the STS stack surviving an early separation of the SRB using the SRB manual separation switch on panel C3:

Could the SRB's have separated successfully at 70 seconds into flight

enter image description here

Estimates ranged from the "Official" accepted answer of:

The Space Shuttle System was not designed to survive a failure of the Solid Rocket Boosters. There are no corrective actions that can be taken if the boosters do not operate properly after ignition, i.e., there is no ability to separate an Orbiter safely from thrusting boosters and no ability for the crew to escape the vehicle during first-stage ascent

To an estimate I received from a NASA engineer of 75% chance of successful separation. My educated guess, based upon herculean design efforts made to get maximum separation from the stack, is that a manual separation before booster burnout had a 50/50 chance of destroying the vehicle.

The question:

Suppose you are in the flight control room on January 28th, 1986. You are a support person for FIDO, perhaps abort support or range safety. (I don't want to label any specific party, so lets leave it general) You look up to John Aaron and have made a point to have lunch with him regularly at JSC. You have listened to him tell the story of how on the Apollo 12 launch he responded to an unexpected environmental condition (lighting strike). He used his extensive knowledge of how the systems work and more important how they fail. As a result of his extra effort, he was able to make a completely unrehearsed call up (SCE to AUX) that saved the mission.


In addition to a thousand other details, you have looked into the possible use of the SRB man separation switch in odd failure conditions. What if a booster suffered a case crack or insulation failure? What would that look like in the data? Would I be able to detect it? Could I respond to it? You have determined that the best way to detect a case breach would be visually. SRM engineers tell you that if it occurs it will propagate in seconds, so the response would need to be quick.

So your routine at launches has been to watch the data but to also watch the live feed tracking cameras to look for visual anomalies. ( like your mentor John Aaron, you do odd extra things because you see value, no matter what the "Official" position is). You don't know what you will do if you think you see an anomaly but you know the SRB manual separation would probably be the only chance, even if the likelihood of success is low. On 51L you are watching the following live feed.

At 16:39:01 you get the pit in your stomach moment (What the **** is that??). You know your John Aaron moment has arrived. Is that really a case breach?? By 16:39:04 you are sure. Months of thinking about it are upon you. Make the call, "FIDO, SRB MAN SEP NOW!!" "Look at the ROTI video!!! MAKE IT..."

Within 3 seconds, by 16:39:07, the trust between you and FIDO and FLIGHT has resulted in the call up to the flight crew "Challenger SRB MAN SEP, Dick, MAN SEP NOW!!"

The response takes another 3 seconds by 16:39:10, the SRB separation process has started.

In reality it was another 3 seconds before the RH aft ET attachment was destroyed and the RH Booster rotated into the ET and began the destruction of the vehicle.

Luckily the high winds aloft on that day assist in the separation. That and good design are on NASA's side and the SRB's separate and clear the accelerating stack.

Did 51L simply need a John Aaron mentor in Flight control that day??

If he had been, would the call up have been made??

Or was the official conclusion the correct one, ye who enter here abandon all hope ..at least till SRB burnout

  • 3
    $\begingroup$ My deepest respect and thanks, to Jim Kingsbury, head of MSFC engineering for some the ideas and details provided with regard to the Challenger accident. Jim was as close as anyone to solving the SRB problems in 1985. RIP Jim (1928-2012) $\endgroup$ Commented Jun 15, 2017 at 18:15
  • 1
    $\begingroup$ What do you think? If there was a method available to save the Challenger crew, NASA would have used it. @Challenger Truth $\endgroup$
    – Starship
    Commented May 20, 2022 at 22:42
  • $\begingroup$ Not necessarily true. In the STS-107, Columbia accident, the CAIB looked into the possibility that a rescue mission could have been launched and succeeded. It never happened for the same reason the Challenger was not saved. No one realized there was a problem until it was too late. This question is simply, what would have happened if you had pushed the SRB sep button prior to 74 seconds. The "why" it didn't happen is a much more complex and entirely different question. $\endgroup$ Commented May 22, 2022 at 22:17

1 Answer 1


My reading of the function of the SRB manual separation control is that it would not actually trigger a separation while the boosters were firing (bold mine):

SRB separation is normally performed automatically by the onboard GPCs; however, the flight crew can command separation through use of the SRB separation switches on panel C3. The SRB separation auto/man (manual) switch is positioned to man and the SRB sep push button depressed.

This manual function for SRB separation provides a backup for the automatic function; however, the manual function uses the same separation logic as the automatic. The automatic sequence is initiated by the software in the GPCs when the SRB chamber pressure is below 50 psi.

I haven't found a more detailed confirmation of this interpretation, however.

I suspect that even if a manual separation was available and would not have destroyed the spacecraft if attempted, it wouldn't have been possible for a controller to call for the separation and have it relayed quickly enough to make a difference.

By 16:39:07 the ET is already compromised and the shuttle's fate is thereby sealed. From the point where the controller makes the call at 16:39:04, all of this has to happen in three seconds:

  • One or two people up the decision chain (FIDO and Flight Director in your scenario) have to hear, comprehend, and unconditionally accept the call
  • Capcom has to repeat the instruction to the spacecraft
  • Spacecraft commander has to hear and comprehend the instruction
  • Spacecraft commander has to locate the SRB sep switches
  • Spacecraft commander has to operate two switches while under 3G acceleration

I don't think this is a possible scenario; just repeating the call from supporting controller to FIDO to Flight to Capcom might take 8 seconds.

Compare with the timeline of the "SCE to AUX" incident on Apollo 12 (trimmed down to a few key points):

000:00:37 Gordon (onboard): What the hell was that?

000:00:43 Conrad (onboard): Roger. We had a whole bunch of buses drop out.

000:00:56 Conrad (onboard): I just lost the platform.

000:01:02 Conrad: Okay, we just lost the platform, gang. I don't know what happened here; we had everything in the world drop out.

000:01:08 Carr: Roger.

000:01:36 Carr: Apollo 12, Houston. Try SCE to auxiliary. Over.

000:01:39 Conrad: Try FCE to Auxiliary. What the hell is that?

000:01:41 Conrad: NCE to auxiliary...

000:01:42 Gordon (onboard): Fuel cell...

000:01:43 Carr: SCE, SCE to auxiliary. [Long pause.]

000:01:50 Conrad (onboard): SCE to Aux.

The crew's first report of a problem to the ground comes at 1:02, 25 seconds after Gordon sees a problem; Capcom Carr doesn't even "roger" to Conrad's call for six seconds. In mission control, they should already have seen, by this time, that the telemetry is bad. In particular, we expect that John Aaron is seeing the glitch by this point.

It's almost 30 seconds later at 1:36 that the first "SCE to Aux" call goes up to the ship; they have no idea what Carr is talking about and Carr has to repeat the call at 1:43. It's unclear if Conrad's "SCE to Aux" at 1:50 is him confirming he's flipped the switch, but assuming that it is, we're looking at 14 seconds from instruction to operation and probably at least 40 seconds since Aaron saw the problem.

In this case, also, "SCE to Aux" is a very low-stakes call, easy to make. If the problem isn't with the signal conditioning equipment, all Aaron's call does is waste a little time, and at this point the launcher itself seems to be doing fine. An emergency booster sep on the shuttle would be at best a mission failure, and depending on who you ask, anywhere from a 25% to 99% chance of loss of crew.

  • 8
    $\begingroup$ Try this: strap a ten pound sandbag to your arm, start a stopwatch, turn away from it, speak out loud the words "FIDO, SRB MAN SEP NOW", "Flight, SRB MAN SEP NOW", "Challenger, SRB MAN SEP NOW", and only then reach to turn it off. Takes me about 8 seconds without the sandbag. $\endgroup$ Commented Jun 15, 2017 at 19:38
  • 3
    $\begingroup$ I would disagree that the Shuttle's fate was sealed when the ET Hydrogen tank was breached at 16:39:07. The vehicle continued flight until the ET attachment failed and did not explode even with the Hydrogen streaming in the SRB plume. With the SRB's gone and a leak, I think it is possible that the stack could continue to execute a RTLS. Even in the accident, the large quantity of H2 and O2 fuel did not explosively burn. Hydrogen requires the right mix of oxygen to burn properly, it probably would not have existed at 50,000 feet from a leak out of the aft portion of the ET. $\endgroup$ Commented Jun 15, 2017 at 19:52
  • 5
    $\begingroup$ I strongly disagree. Uncontrolled venting from the H tank, even if it chooses not to explode, is likely to destabilize the orbiter+ET, and once it's turned more than a few degrees, it's cartwheeling and coming apart. Feel free to write your own answer, but my conclusion is that there's no way to save the crew. $\endgroup$ Commented Jun 21, 2017 at 14:07
  • 5
    $\begingroup$ I have received information from a member of the SAIL team in 1986, that both the SRB sep switch and the ET sep switch had minimal Flight Software checks prior to passing the signal on to the Master Events controller. In plain language, the button once pressed would trigger the event. With respect to the ET sep button the Astronaut office objected to any FSW that checked the 2% ET tank level during RTLS and that same objection was carried over to the SRB sep. button The pilot in command would make the sep decision not the computer. This means that successful or not this was an option $\endgroup$ Commented Jul 13, 2018 at 19:11
  • 3
    $\begingroup$ @ChallengerTruth, an RTLS abort was considered "an unnatural act of physics" by the astronauts, and even with an intact Shuttle and ET was a high-risk maneuver. With a leaking hydrogen tank, it's impossible. The ET needs to be nearly empty when you jettison it to keep it from colliding with the Shuttle; without hydrogen, there's no way to empty the oxygen tank (you can't burn it off, there's no dump valve, and running oxygen through the engines without hydrogen for it to burn will cause the engines to explode). $\endgroup$
    – Mark
    Commented Nov 2, 2018 at 23:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.