4 ms·

I used Audacity (and YouTube-dl) a few days ago to have an estimation of the top speed using the SSC information such as the gearbox ratios and the size of the

I used Audacity (and YouTube-dl) a few days ago to have an estimation of the top speed using the SSC information such as the gearbox ratios and the size of the tires, and the sound of the engine in the YouTube video. Because it's actually simple to listen how fast an engine rotates, I found that they went slower than what is advertised and that the engine didn't rotate as fast as they claim. I computed 7530rpm and they claim 8600rpm. I posted it on HN and I hope it's ok to post it again as a comment : https://fungiboletus.github.io/ssc-rpm/ https://fungiboletus.github.io/ssc-rpm/

SomeHacker44 6y ago
Just curious, did you take into account any possible doppler shift of the frequencies?
- speedgoose 6y ago
  No, should I have? The microphone was inside the car.
  - Waterluvian 6y ago
    If sound source and sensor were at a fixed distance from each other then no.
gameswithgo 6y ago
when i first saw the video the rate of acceleration in the 300mph domain made no sense to me, given the power of the car. it didn’t creep up to 316! it should have.
toxik 6y ago
Interesting analysis! I’ve seen the 2x multiplier myself, when I looked at the spectrogram of an AC induced whine in a speaker IIRC
tgtweak 6y ago
> in V8 there is a cylinder firing at every 90° rotation which means 2 cylinder fire in one crankshaft rotation. Your method and math checks out even without the tachometer observations - in a split-plane v8 (such as that in the scc tuatara) you have two cylinder fires per rotation of the engine. You have measured 251 cylinder fires per second (251hz) thus 125.5 rotations per second OR 7530/minute (RPM). Hard to refute that.
- m4rtink 6y ago
  Apparently (according to a YouTube video I watched :P) this is how they tell combustion engine powered torpedos apart via passive sonar. Analyzing the sound gives you the number of cylinders and number of propeller blades, which can identify the given torpedo type quite reliably.
  - bootcampwhere 6y ago
    Sonar tech here, this was true 20+ years ago but new torpedo and sub blade configurations are not only highly classified but any nuclear powered subs no longer even use blades to counter this exact problem. Only old russian and N. Korean subs can be identified this way anymore. Even when I was active duty 10 years ago china and us subs were basically silent and torpedoes surround themselves with air bubbles so the acustics dont really escape. You know something is coming at you fast but that it
    - m4rtink 6y ago
      Interesting! :)
    - trevorishere 6y ago
      Classified and visible on mapping services! https://i.stuff.co.nz/technology/47327/Maps-site-reveals-US-nuclear-sub-secrets https://i.stuff.co.nz/technology/47327/Maps-site-reveals-US-...
jml7c5 6y ago
One possible error in ground speed measurement: the tire likely grew in diameter as it spun faster. I don't know how large the effect is here, but it's worth checking or noting.
- speedgoose 6y ago
  True, it could explain a small difference. But here the difference is huge, especially in terms of RPM.
  - jml7c5 6y ago
    Yeah, it wouldn't be enough to make such a huge difference. Nor would it explain the different RPM. Just wanted to note it for the sake of completeness.

SomeHacker44 6y ago

Just curious, did you take into account any possible doppler shift of the frequencies?

speedgoose 6y ago
No, should I have? The microphone was inside the car.
- Waterluvian 6y ago
  If sound source and sensor were at a fixed distance from each other then no.

gameswithgo 6y ago

when i first saw the video the rate of acceleration in the 300mph domain made no sense to me, given the power of the car. it didn’t creep up to 316! it should have.

toxik 6y ago

Interesting analysis! I’ve seen the 2x multiplier myself, when I looked at the spectrogram of an AC induced whine in a speaker IIRC

tgtweak 6y ago

> in V8 there is a cylinder firing at every 90° rotation which means 2 cylinder fire in one crankshaft rotation. Your method and math checks out even without the tachometer observations - in a split-plane v8 (such as that in the scc tuatara) you have two cylinder fires per rotation of the engine. You have measured 251 cylinder fires per second (251hz) thus 125.5 rotations per second OR 7530/minute (RPM). Hard to refute that.

m4rtink 6y ago
Apparently (according to a YouTube video I watched :P) this is how they tell combustion engine powered torpedos apart via passive sonar. Analyzing the sound gives you the number of cylinders and number of propeller blades, which can identify the given torpedo type quite reliably.
- bootcampwhere 6y ago
  Sonar tech here, this was true 20+ years ago but new torpedo and sub blade configurations are not only highly classified but any nuclear powered subs no longer even use blades to counter this exact problem. Only old russian and N. Korean subs can be identified this way anymore. Even when I was active duty 10 years ago china and us subs were basically silent and torpedoes surround themselves with air bubbles so the acustics dont really escape. You know something is coming at you fast but that it
  - m4rtink 6y ago
    Interesting! :)
  - trevorishere 6y ago
    Classified and visible on mapping services! https://i.stuff.co.nz/technology/47327/Maps-site-reveals-US-nuclear-sub-secrets https://i.stuff.co.nz/technology/47327/Maps-site-reveals-US-...

jml7c5 6y ago

One possible error in ground speed measurement: the tire likely grew in diameter as it spun faster. I don't know how large the effect is here, but it's worth checking or noting.

speedgoose 6y ago
True, it could explain a small difference. But here the difference is huge, especially in terms of RPM.
- jml7c5 6y ago
  Yeah, it wouldn't be enough to make such a huge difference. Nor would it explain the different RPM. Just wanted to note it for the sake of completeness.