The official audio Record of Apollo 11

[nomuse]

There was a nice reference to noise-canceling microphones I ran across while brushing up in my Yamaha Sound Reinforcement Handbook.  The technique is, essentially, a dual-ported capsule.  Sounds from far away reach both sides of the capsule at more-or-less the same time and being in phase, are canceled.  Sounds that are much closer to one port than the other create a pressure differential.  Obviously this is frequency dependent, and in practice you have to roll off the low end considerably, but otherwise is an elegant solution.

Basically, it's a humbucker for a human voice.  Can also do the trick with a pair of capsules wired out of phase.

[VincentMcConnell]

Yes. Listening to the Apollo 17 onboard voice recorders, you can hear a little bit of stress in the voices of the astronauts during Max-Q.

[nomuse]

Err, let me edit, I was describing phase and meaning pressure differential....anyhow, the principle of the dual ported/twin capsule is that sound propagates through air in the usual inverse-square.   So if the astronaut's mouth is two cm away, and the mic is two cm between the ports, the pressure of that wave has dropped to a quarter by the time it hits the rear port.  So you have a big difference in pressure and that difference is what you amplify.  The sound of the engine several meters away is pretty much identical in pressure on both ends of the mic, thus it gets canceled.

Of course phase WILL rear its ugly head here, as any wavelengths around the size of that distance between the ports are going to be amplified or attenuated based on their phase difference as much as by the pressure difference.  And in the real world, sounds aren't perfect point sources in still air -- the human voice emanates both from air moving out of the mouth, and the resonating surfaces of the sinuses (which is why we tape microphones to people's foreheads in live stage musicals).  But the general idea is still there!

[ka9q]

This method of passively adding wavefronts with different phases from different directions is also exactly how directional radio antennas work, albeit with electromagnetic waves rather than sound pressure waves. A noise-canceling microphone is nothing more than a highly directional microphone, with the main lobe pointed directly at the speaker's mouth.

[ka9q]

Here's the acceleration profile of a Shuttle launch (and landing):

http://www.russellwestbrook.com/Acceleration.htm

You can see how acceleration is (relatively) high at liftoff, 1.5-2g, much higher than the Saturn V (1.15g). For the pass through MaxQ at about 1 minute, it dips back to 1.5 g due to main engine throttleback (to about 65%) and the shape of the SRB propellant grain. Due to the rapidly decreasing atmospheric pressure as the shuttle gains altitude, aerodynamic pressure falls just off as rapidly as it peaks. So after MaxQ the engines are then throttled back up ("Go at throttleup"). MaxQ is a very obvious peak in wind noise in the cockpit video recordings.

Acceleration drops markedly at SRB burnout, as you'd expect.

I do not quite understand the constant 1-g region just after SRB separation. I'd expect acceleration to immediately increase exponentially from that point, not 80 seconds later. Perhaps the ET is still too full at that point for the main engines to gain much altitude without the help of the SRBs. Maybe a "lofted" trajectory is flown during the SRB phase to gain altitude relatively rapidly, allowing the shuttle to pitch down after staging to decrease gravity losses until the ET can lose enough weight to allow the shuttle to pitch back up and gain orbital altitude without excessive gravity losses.

After about 200 seconds the acceleration increases exponentially as the ET loses weight, followed by the throttleback to limit acceleration to 3g just before MECO.

And here it is for an Apollo/Saturn V launch, specifically Apollo 11:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19900066485_1990066485.pdf (pdf page 54, memo page 4-5).

Liftoff acceleration is barely more than 1g, and it takes a long time to clear the tower. Acceleration builds exponentially as propellants are burned off, so the inboard F-1 engine is cut off early to prevent exceeding 4g. Acceleration quickly builds up again on the remaining four until they also cut off at S-IC propellant depletion.

The S-II starts at less than 1g, again building exponentially as it depletes its propellants but never exceeding 2g. There are drops at center engine cutoff and again at EMR (mixture ratio shift). EMR decreases the flow of LOX to the engines relative to LH2, decreasing thrust but increasing Isp slightly. It is timed dynamically to result in simultaneous depletion of both propellants.

The first burn of the S-IVB is especially gentle, reaching only 3/4 g before MECO because most of the propellants are kept for the TLI burn.

[nomuse]

This method of passively adding wavefronts with different phases from different directions is also exactly how directional radio antennas work, albeit with electromagnetic waves rather than sound pressure waves. A noise-canceling microphone is nothing more than a highly directional microphone, with the main lobe pointed directly at the speaker's mouth.

Actually, that is what I thought at first.  And that confused me.  I work with cardiods and hypercardiods, and the off-axis rejection of those just didn't feel sufficient to be the whole story.

The cuteness of turning the microphone into a pressure-differential microphone is that it cares about proximity.  Sources with proximity have a greater difference in pressure.  Sounds that are not proximate are largely rejected.  Maybe the humbucker is a bad example -- because you can look at the humbucker just as you look at a balanced audio cable; it's all common-mode rejection.