BMW M3/M4 s55 Engine RPM Redline Projections via Technical Sound Analysis

[bradleyland]

Originally Posted by Levi

You did a sound analysis quite easily, could do it to figure out if the engine is I6 or V6 (90°)?

That's easier to do by ear than it is through frequency analysis because of the difference in character between a cross-plane V8 and an I6. A cross-plan V8 has a "double" combustion event on each side as it works through the firing order. Scroll down to the "Chevy V8 Firing Order" graphic on this page to see what I mean:

http://craig.backfire.ca/pages/autos/v8-engines#firing

For a given RPM, a V8 will have a higher sound frequency, because it has more combustion events per revolution, so if you know the RPM, it's trivially easy to determine the number of cylinders through frequency analysis.

[NISFAN]

Originally Posted by bradleyland

It's possible, but you need a good clean upshift where the driver is WOT (or close) before and after the shift. Basically, a hard acceleration run at an easily estimated speed.

Watch this video and post any timestamps of regions where you find that:

If you do the footwork, I'm happy to do the analysis.

Thanks, 0:55 and 1:30 are two possible samples. It doesn't sound like max RPM upshifts to me, but that might be some audible illusion (???)

[bradleyland]

Originally Posted by NISFAN

Thanks, 0:55 and 1:30 are two possible samples. It doesn't sound like max RPM upshifts to me, but that might be some audible illusion (???)

Let me ask you a question. What insight do you want to gain? I assumed you were interested in the gear spacing, but I'm not sure that's correct. I can tell you the highest amplitude frequency before and after the shift, but that will only tell you the relative difference in RPM between gears; it won't provide any insight in to maximum RPM.

[bradleyland]

The "illusion" you refer to is called doppler shift. It can be corrected if you know the correct inputs, but we have to guess because we don't know the velocity of the car, nor the exact angle-of-travel relative to the recording source. If you're making a relative comparison, and the inputs are close enough, you needn't correct for it.

I used the 1:30 region for analysis.

Samples:

• Pre-shift: 01:28.2, 00:00.3 duration
• Post-shift: 01:29.2, 00:00.3 duration

Highest amplitude peaks (not corrected for doppler, Hanning window):

• Frequency pre-shift: 289 Hz
• Frequency post-shift: 232 Hz

Plots:

[BMW269]

Originally Posted by bradleyland

That's easier to do by ear than it is through frequency analysis because of the difference in character between a cross-plane V8 and an I6. A cross-plan V8 has a "double" combustion event on each side as it works through the firing order. Scroll down to the "Chevy V8 Firing Order" graphic on this page to see what I mean:

http://craig.backfire.ca/pages/autos/v8-engines#firing

For a given RPM, a V8 will have a higher sound frequency, because it has more combustion events per revolution, so if you know the RPM, it's trivially easy to determine the number of cylinders through frequency analysis.

I meant V6, not V8.

[NISFAN]

Originally Posted by bradleyland

Let me ask you a question. What insight do you want to gain? I assumed you were interested in the gear spacing, but I'm not sure that's correct. I can tell you the highest amplitude frequency before and after the shift, but that will only tell you the relative difference in RPM between gears; it won't provide any insight in to maximum RPM.

No you are correct, I wanted to understand the rev drop from one gear to the next. The problems relating that into anything tangible, is we don't know the speeds, or even the gear numbers (guesses aside).

Thanks for the frequency plot. If anyone can shed some light on speed approximation for the piece of track involved. We can at least guess a gear number.

[NISFAN]

Originally Posted by bradleyland

The "illusion" you refer to is called doppler shift. It can be corrected if you know the correct inputs, but we have to guess because we don't know the velocity of the car, nor the exact angle-of-travel relative to the recording source. If you're making a relative comparison, and the inputs are close enough, you needn't correct for it.

I used the 1:30 region for analysis.

Samples:

• Pre-shift: 01:28.2, 00:00.3 duration
• Post-shift: 01:29.2, 00:00.3 duration

Highest amplitude peaks (not corrected for doppler, Hanning window):

• Frequency pre-shift: 289 Hz
• Frequency post-shift: 232 Hz

Assuming negligible change in speed (cancelling doppler) between shifts, there is a 20% drop in engine revs, does that conclusion sound right?

I'm not going to guess on what the doppler effect is, as speed unknown. Assuming a gear change at say 7,000rpm, next gear is 5,600rpm. That is quite a long gear spacing if from 3-4th on a 7 speed box.

Little movie I took in the M6, very similar. Forward to 27 seconds.

[bradleyland]

Originally Posted by NISFAN

Assuming negligible change in speed (cancelling doppler) between shifts, there is a 20% drop in engine revs, does that conclusion sound right?

I'm not going to guess on what the doppler effect is, as speed unknown. Assuming a gear change at say 7,000rpm, next gear is 5,600rpm. That is quite a long gear spacing if from 3-4th on a 7 speed box.

Little movie I took in the M6, very similar. Forward to 27 seconds.

Yep, that sounds about right.

[bradleyland]

Originally Posted by Levi

I meant V6, not V8.

Ah, in that case, I don't think waveform analysis is sufficient to identify the difference without some very specific circumstances. E.g., a V6 with dual exhaust, and a stereo mic placed on the centerline.

If you had a very detailed recording, you could examine the waveform from various examples of V6 engines versus I6 engines, but you'd need a high quality recording with many samples. Ultimately, I think there are too many confounding factors, like exhaust design and turbocharger configuration.

[NISFAN]

Originally Posted by bradleyland

Ah, in that case, I don't think waveform analysis is sufficient to identify the difference without some very specific circumstances. E.g., a V6 with dual exhaust, and a stereo mic placed on the centerline.

If you had a very detailed recording, you could examine the waveform from various examples of V6 engines versus I6 engines, but you'd need a high quality recording with many samples. Ultimately, I think there are too many confounding factors, like exhaust design and turbocharger configuration.

Agree. For example listen to this video. The cars are all Subaru's with a flat 4 (horizontally opposed) engine. All the cars except the Blue Hatchback (around 0:25, and the video front image) are the common Subaru exhaust headers Turbo one side of engine fed with unequal lenght headers from each bank), which give an uneven 'subaru burble'. The car at 0:25 is fitted with the 'Twin Scroll' equal length headers, which changes it's sound completely, it sounds like an inline 4 (which it should as firing timing is exactly the same as an inline 4). Ignore the first 'anti lag' rev pops for this one.

On a parallel twin turbo straight six, although not as dramatic as the Subaru, the front 3 cylinders typically have a longer exhaust path, which phases the sound out of sync with the rear 3. This could give the impression it is a V shaped engine sound. Certainly the M3/M4 doesn't sound like a conventional inline 6.

[Boss330]

Swamp, Bradleyland etc:

Does this clip contain any valuable sound info? Listen to the car passing at around 2:19 on the long straight at the 'Ring:

[swamp2]

Originally Posted by Boss330

Swamp, Bradleyland etc:

Does this clip contain any valuable sound info? Listen to the car passing at around 2:19 on the long straight at the 'Ring:

Nope, you need the sound right before a shift and a likely redline shift as well. As a side note I believe it is unlikely that the car will require redline shifts in the upper gears to extract maximum acceleration (unlike current car).

[Boss330]

Originally Posted by swamp2

Nope, you need the sound right before a shift and a likely redline shift as well. As a side note I believe it is unlikely that the car will require redline shifts in the upper gears to extract maximum acceleration (unlike current car).

My bad, I thought there was a shift on that straight, but re-listening I see that I was mistaken...

[Boss330]

Is this whistling noise on the MB CLA there just to disguise engine RPM's?

[bradleyland]

Originally Posted by Boss330

Is this whistling noise on the MB CLA there just to disguise engine RPM's?

I don't know what the heck that whistling noise is, but it won't prevent a Fast Fourier Transform analysis of the engine RPM. To do that, you'd need to produce a tone similar in frequency to the primary combustion impulse. That whistle is way too high-pitched. Maybe it's a harmonic, but each multiple you move from the fundamental, the benefit of confusing the base frequency would be diminished. That whistle would have to be a shriek.

[kyleb350]

Originally Posted by Boss330

Is this whistling noise on the MB CLA there just to disguise engine RPM's?

It's got those whistle tips. woo WOO!

[swamp2]

Originally Posted by Boss330

Is this whistling noise on the MB CLA there just to disguise engine RPM's?

I strongly suspect it is just pressurized air, air/fuel or even exhaust leaking in the turbo system somewhere due to a poorly fitting part or an improperly torqued bolt(s). This is a known issue. As noted above this would not disguise the fundamental combustion frequency in a fourier transform.

[Dr. Sound]

Please have a look: http://f80.bimmerpost.com/forums/sho...7&postcount=24