5
$\begingroup$

I believe some important information could be recovered if the Launchpad audio of STS-51L still exists.

The leak in the RH aft field joint could have created a high pitched whistle between 0.678 seconds and 2.733 seconds.

Based upon the frequency of that sound, it could be determined much more accurately the size and dynamics of that leak (was it a continuous leak or a leak/seal/ leak process).

Roger's Commission found the following:

Eight more distinctive puffs of increasingly blacker smoke were recorded between .836 and 2.500 seconds. The smoke appeared to puff upwards from the joint. While each smoke puff was being left behind by the upward flight of the Shuttle, the next fresh puff could be seen near the level of the joint. The multiple smoke puffs in this sequence occurred at about four times per second, approximating the frequency of the structural load dynamics and resultant joint flexing. https://history.nasa.gov/rogersrep/v1ch3.htm

The question I am trying to answer is whether those puffs were part of a leak/seal/leak/seal process (implying the O-rings were still functional) or were the puffs a result of opening and narrowing of the gap (implying the joint was always leaking just at different rates)

Does anyone know of a source for shuttle Launchpad recordings?

| improve this question | |
$\endgroup$
  • 1
    $\begingroup$ I am asking if anyone knows whether this information exists and providing some context as to its use so that I can define the quality of the recording required. The cold O-ring theory is not written in stone. This data, if recovered, could just as easily confirm as disprove that theory, so I am not advancing either theory, just looking for data. $\endgroup$ – Challenger Truth Aug 21 '18 at 3:47
  • 2
    $\begingroup$ conjecture: "would have created a high pitched whistle", conjecture: "could be determined much more accurately the size and dynamics". Consider either sourcing this conjecture from a reliable, authoritative source, or labeling it as your own personal belief. Whistling requires a very specific geometry, otherwise a leak will just be a hiss. There is a huge amount of sound when the Space Shuttle takes off, and huge is an understatement. Microphones that were of low-enough sensitivity not to be completely saturated by this are less likely to pick up some high-pitched whistling. $\endgroup$ – uhoh Aug 21 '18 at 4:08
  • $\begingroup$ To my knowledge neither the NASA Accident Analysis team nor the Rogers Commission ever considered this possibility. It is strictly conjecture on my part that a whistle would occur. A typical referee whistle can generate 125 decibels from a 1/2 inch wide opening. link The NASA accident analysis team estimated the original leak size to be no greater than 3/4 of an inch. The human lungs can generate up to 3 psi. At 900 psi generated by the SRB at 678 ms when the leak started, it is entirely possible it was very loud $\endgroup$ – Challenger Truth Aug 21 '18 at 5:55
  • 3
    $\begingroup$ Does NASA record audio at the launch pad at all? In launch TV broadcasts, the microphone is usually miles away. $\endgroup$ – Hobbes Aug 21 '18 at 13:40
  • $\begingroup$ All I could find is the transcript, but this is from the cockpit... $\endgroup$ – Magic Octopus Urn Aug 21 '18 at 14:45
10
$\begingroup$

I don't know about 1986, but if the systems were like the ones that existed in the 2000s, pad audio recordings will not help your quest.

The microphone we put out on the pad, we only use that when the sparklers ignite. Once the engines kick on you have to get off that immediately, or you have nothing but overdriven clipping noise.

Loren Mathre, audio control technician, quoted here.

| improve this answer | |
$\endgroup$
  • $\begingroup$ If I understand the statement and the words, overdriven clipping occurs when an audio system attempts to recreate a sound and the amplifier "attempts to deliver an output voltage beyond the maximum capability of the amplifier". The focus is on the other end of the audio equation, is the recorder capable of recording the signal strength and frequency even if the system is incapable of recreating them using its amplifier and speakers. I am attempting to contact Dimitri Gerondidakis to see if he can shed light on the recording capability of the NASA Select systems. $\endgroup$ – Challenger Truth Aug 22 '18 at 3:13
  • $\begingroup$ @ChallengerTruth That "overdriven" has one definition or use does not preclude it from having others as well. One can overdrive a microphone with sound pressure, especially electret microphones or others that contain active electronics. These days microphones are supplied a small DC bias current because they contain FETs. See (just for one example) Electrical specification for minimum bias current available from 3.5mm cellphone microphone input? See also "hidden or unstated premise." $\endgroup$ – uhoh Aug 22 '18 at 4:00
  • 1
    $\begingroup$ @ChallengerTruth if the microphone clips electrically, or mechanically due to excessive sound pressure, further discussion is moot, at least in Space Exploration SE. As I said here you are trying to "advance your own theory or conjecture, and that's not what Stack Exchange is for." $\endgroup$ – uhoh Aug 22 '18 at 4:16
  • 7
    $\begingroup$ When a microphone clips, its membrane hits the structure around it. This adds lots of extra noise to the recording. This noise (= random fluctuations of the signal) has a broad spectrum, which makes it very difficult to filter out. In general, it's considered impossible to reconstruct a clean signal from a clipped input. In this case, you don't just have clipping, but the exhaust noise is also random and broad-spectrum, and much louder than any noise from the field joint leak. You'd be better off starting with clean audio from a microphone a few km away. $\endgroup$ – Hobbes Aug 22 '18 at 7:28
  • 3
    $\begingroup$ I actually took a natural speech and language processing course. There is a point at which all other audio recorded in a clip is completely lost given enough random, loud garble... lost meaning that no matter how much noise you remove, the noise is so much greater than the other sounds (E.G. someone talking) that you end up removing a lot of the sound you're looking for. The problem is further compounded by not knowing what you're trying to extract. If you wanted to prove someone said "hi" at the launchpad... Maybe? Finding a random noise among random noises? No. $\endgroup$ – Magic Octopus Urn Aug 22 '18 at 13:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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