Loudspeakers - how?

[grandad]

I think the audio quality on this video is pretty good. The subtle differences come across. What's with the Fender LS I don't know!

 

[Bill Fitzmaurice]

The differences aren't subtle at all when what's being measured is the off-axis response at a distance. That's where size actually matters.

[Stub Mandrel]

These graphs could show the movement of a string, and electrical signal, the movement of a speaker coil or the movement of the air.

These frequencies are related, but they don't need to be. Just plot the signals and add them together and the resulting graph plots out what you get.

Sometimes what you get has some big, fast changes or really slow ones. These may bet 'filtered out' by the characteristics of the circuit or loudspeaker etc. and this can 'colour' the sound.

 

 

[leftybassman392]

Following up Stub's piece, perhaps you may care to see a couple of plots of real instruments. Ignore the text at the bottom: It got caught in the screen grab and I couldn't be bothered to take it out.

Notice how much more quickly the overtones die away for the flute. The very different plots for the same note are the reason the two instruments sound so different. As has been said by another poster, the really clever stuff goes on in your brain. As long as you've heard them before, it can identify which instrument is which. Not only that, but if you closed your eyes with two musicians in front of you playing these instruments at the same time, it can not only separate out the two sounds from the enormously complex waveform that's hitting your ear drums, but also allow you to locate each instrument in 3D space (not to mention a whole bunch of other stuff as well). All this from what arrives at your ears as a single pressure wave.

[Phil Starr]

9 hours ago, grandad said:

I think the audio quality on this video is pretty good. The subtle differences come across. What's with the Fender LS I don't know!

 

These things are fun, thanks for posting it.

The clue though is the difference between the two eight inch speakers. Not all eights sound the same so why should all the 10's, 12's or 15's? And they were all using different amps albeit all from Fender.

The 15 looks has a horn, if so did it also have a crossover modifying the signal to the bass driver? The Rumble 500 also has a horn but not the 12 and the crossover frequency may be different.

As I said these things are fun but not very objective and you can't really draw wider conclusions about other speakers in other cabs. Even with the Rumble series the sound a few metres away won't be the same as these recordings.

 

[Dan Dare]

1 hour ago, Phil Starr said:

the sound a few metres away won't be the same as these recordings.

 

18 hours ago, Bill Fitzmaurice said:

The differences aren't subtle at all when what's being measured is the off-axis response at a distance. That's where size actually matters.

Nuff said, really. Measuring or evaluating a drive unit at low levels by putting a mic on axis and about 6" from the cone tells you little about how it will perform in the real world.

Edited August 31, 2020 by Dan Dare

[Chienmortbb]

17 hours ago, leftybassman392 said:

Following up Stub's piece, perhaps you may care to see a couple of plots of real instruments. Ignore the text at the bottom: It got caught in the screen grab and I couldn't be bothered to take it out.

Notice how much more quickly the overtones die away for the flute. The very different plots for the same note are the reason the two instruments sound so different. As has been said by another poster, the really clever stuff goes on in your brain. As long as you've heard them before, it can identify which instrument is which. Not only that, but if you closed your eyes with two musicians in front of you playing these instruments at the same time, it can not only separate out the two sounds from the enormously complex waveform that's hitting your ear drums, but also allow you to locate each instrument in 3D space (not to mention a whole bunch of other stuff as well). All this from what arrives at your ears as a single pressure wave.

The flute's output is almost a  pure sinewave, unless played by Ian Anderson, and contains little in the way of harmonics so the top graph looks strange to me.

[Stub Mandrel]

39 minutes ago, Chienmortbb said:

The flute's output is almost a  pure sinewave, unless played by Ian Anderson, and contains little in the way of harmonics so the top graph looks strange to me.

They are fourier analyses of the sounds, no time domain in those graphs. Most of the flute overtones are inaudible

[leftybassman392]

1 hour ago, Chienmortbb said:

The flute's output is almost a  pure sinewave, unless played by Ian Anderson, and contains little in the way of harmonics so the top graph looks strange to me.

The harmonic series for the flute is new to me as an area of study, so bear with me.

From my research there are some points that might be worth making:

1. There are many types of flute, all of which have slightly different harmonic spectra. The fact that an experienced ear can tell them apart means they can't be pure sine waves.

2. Like you, I've always thought of the flute as a very pure-sounding instrument, but analysis indicates that the actual sound spectrum you get from the instrument depends on a number of factors (note pitch and note intensity being two examples)

3. The dB scale used in the charts is marked at 5dB intervals. Examination of the plot for the flute indicates that the first harmonic is some 14 dB down on the fundamental, the second harmonic is around 5dB down, and subsequent harmonics drop away markedly thereafter. A drop of 3dB indicates a halving of sound intensity. From these numbers, the first harmonic has approx. 1/40th of the intensity of the fundamental, second harmonic a little over a quarter of the intensity, with subsequent harmonics contributing vanishingly small components to the sound. Bear in mind that this is a single Fourier plot at a single frequency and intensity from a single flute. One would expect the spectrum to be different under diferent conditions.

This link provides some examples of flute waveforms. The accompanying notes flesh out the details.

Edited August 31, 2020 by leftybassman392

[Chienmortbb]

1 hour ago, leftybassman392 said:

The harmonic series for the flute is new to me as an area of study, so bear with me.

From my research there are some points that might be worth making:

1. There are many types of flute, all of which have slightly different harmonic spectra. The fact that an experienced ear can tell them apart means they can't be pure sine waves.

2. Like you, I've always thought of the flute as a very pure-sounding instrument, but analysis indicates that the actual sound spectrum you get from the instrument depends on a number of factors (note pitch and note intensity being two examples)

3. The dB scale used in the charts is marked at 5dB intervals. Examination of the plot for the flute indicates that the first harmonic is some 14 dB down on the fundamental, the second harmonic is around 5dB down, and subsequent harmonics drop away markedly thereafter. A drop of 3dB indicates a halving of sound intensity. From these numbers, the first harmonic has approx. 1/40th of the intensity of the fundamental, second harmonic a little over a quarter of the intensity, with subsequent harmonics contributing vanishingly small components to the sound. Bear in mind that this is a single Fourier plot at a single frequency and intensity from a single flute. One would expect the spectrum to be different under diferent conditions.

This link provides some examples of flute waveforms. The accompanying notes flesh out the details.

The note beside the plot is telling. “Higher in the range of the flute, the energy is mostly concentrated in the fundamental and second harmonic. For very high notes, it is almost a sine wave.

Edited August 31, 2020 by Chienmortbb

[leftybassman392]

34 minutes ago, Chienmortbb said:

The note beside the plot is telling. “Higher in the range of the flute, the energy is mostly concentrated in the fundamental and second harmonic. For very high notes, it is almost a sine wave.

In what way is it telling? You seem to be ignoring all the other data for the sake of being (sort of) right for the notes at the very top of the instrument's range - a region where a real-world flautist would spend very little of his/her time. You've said nothing, for example, about the numerous harmonics picked up by the Fourier plot.

And while I'm here, I posted the link knowing that remark was there. If you're looking to be that picky about it, and given the wording you have chosen to highlight, it's also true to say that for most of it's range it's not almost a sine wave. Fundamental plus second harmonic doth not a sine wave make.

[Chienmortbb]

11 hours ago, leftybassman392 said:

In what way is it telling? You seem to be ignoring all the other data for the sake of being (sort of) right for the notes at the very top of the instrument's range - a region where a real-world flautist would spend very little of his/her time. You've said nothing, for example, about the numerous harmonics picked up by the Fourier plot.

And while I'm here, I posted the link knowing that remark was there. If you're looking to be that picky about it, and given the wording you have chosen to highlight, it's also true to say that for most of it's range it's not almost a sine wave. Fundamental plus second harmonic doth not a sine wave make.

You basically correct and I am not trying to pick a fight. One of the problems with any instruments is the way it is played and my referal to Ian Anderson was meant to convey that. His technique, in a rock setting, is agressive and will contain many harmonics, a softly played flute, far fewer. I suspect that a penny whistle would be closer to a pure sine wave as would a human whistle. It is also worth noting that every noise cosnsists of one or more sinewaves.

I should also apologise for my part in subverting the thread.

[leftybassman392]

TBH I'm not so sure we are subverting the thread. The OP was asking a question about loudspeakers producing different notes simultaneously. This part of the discussion is directly related to that, surely? Seriously, I wouldn't worry about it.

Whatever, I always find this type of discussion fascinating. I've studied psychoacoustics on and off for many years, and no matter how many times I return to it I never cease to be astonished at the quantity of information the human brain is able to extract from a single pressure wave.

[Steve Browning]

4 minutes ago, leftybassman392 said:

TBH I'm not so sure we are subverting the thread. The OP was asking a question about loudspeakers producing different notes simultaneously. This part of the discussion is directly related to that, surely? Seriously, I wouldn't worry about it.

Carry on, chaps. It's all new knowledge to me.

[leftybassman392]

9 minutes ago, Steve Browning said:

Carry on, chaps. It's all new knowledge to me.

Righto!

There is an acoustic phenomenon arising out of the overtone series that will be of particular interest to bass players: the missing fundamental

In simple terms, it's the ability of the brain to 'hear' sounds that aren't actually there. Compact audio speaker systems are for the most part physically unable to reproduce the lowest frequencies in the audio spectrum. However, the brain is able to use the overtone series to 'fill in' the missing fundamental. This is how well-designed small speaker systems are able to give the impression of extended bass that isn't really there.

As an added bonus, because the extended bass output is a psychoacoustic phenomenon that isn't being physically reproduced by the audio system, you're less likely to get complaints from your neighbours about thumping bass rattling their ornaments when you crank up the volume on your hifi. 👍

 

[Bill Fitzmaurice]

The missing fundamental effect is real, but the harmonics that create the effect are electronically generated. The only way the speaker enters into the equation is that since it doesn't have to play the fundamental it can be smaller.

[Beer of the Bass]

On the flute chat, are we talking about the sustained portion of the note, or the whole note from start to finish? I feel with flutes and whistles, a lot of the identifiable character is in the transient at the beginning of the note, so it could be possible to have a sustained tone close to a sinewave but still contain the complexity that makes the instrument recognisable.

From experience, it's surprisingly hard to make a convincing flute patch using a sinewave oscillator on a synth...

Edited September 1, 2020 by Beer of the Bass

[leftybassman392]

2 minutes ago, Bill Fitzmaurice said:

The missing fundamental effect is real, but the harmonics that create the effect are electronically generated. The only way the speaker enters into the equation is that since it doesn't have to play the fundamental it can be smaller.

Indeed it is common practice these days to intervene actively (as explained in the linked article), but compact speaker systems have been around a long time, and it is also known that the brain has the ability to fill in the blanks irrespective of any technologies that may be used to reinforce the effect.

While I'd agree that modern speaker systems are smaller because the need for size to reproduce deep bass is not there to the same degree, that hasn't always been the case.

[Stub Mandrel]

21 minutes ago, Beer of the Bass said:

On the flute chat, are we talking about the sustained portion of the note, or the whole note from start to finish? I feel with flutes and whistles, a lot of the identifiable character is in the transient at the beginning of the note, so it could be possible to have a sustained tone close to a sinewave but still contain the complexity that makes the instrument recognisable.

From experience, it's surprisingly hard to make a convincing flute patch using a sinewave oscillator on a synth...

Back in BBC micro days one used a sine wave with a dash of white or pink noise at the beginning. 🙂

[leftybassman392]

48 minutes ago, Beer of the Bass said:

On the flute chat, are we talking about the sustained portion of the note, or the whole note from start to finish? I feel with flutes and whistles, a lot of the identifiable character is in the transient at the beginning of the note, so it could be possible to have a sustained tone close to a sinewave but still contain the complexity that makes the instrument recognisable.

From experience, it's surprisingly hard to make a convincing flute patch using a sinewave oscillator on a synth...

The plots I printed earlier are sustained note segments. Transients (which I agree is an important topic in it's own right) is not covered in the posts I, er, posted.

Giving away my age a bit here, but I'm old enough to have worked with the first generation of synthesisers such as the VCS 3 in the late '60s. IME a pure sine wave is actually a very boring, characterless sound. In order to give it enough character to make it interesting to listen to it was necessary to process it in a number of ways. Sawtooth waves were much more interesting to work with.

I can't recall ever hearing this tested, but I rather suspect that a sine wave generator playing the same sequence of notes as a flute would be markedly less engaging to listen to. Granted there would be clues in the sequence other than the character of the sound that a musician would pick up on, but even so...

As a footnote, it may also be worth reinforcing the point about the brain's remarkable ability to extract information from the audio content coming in through the ears, and in particular the ability of an experienced musician to spot absolutely tiny nuances in the timbre of different musical instruments of the same type. Hands up everybody who reckons they could spot the difference between a Jazz bass and a P bass playing the same sequence of notes in the hands of the same player...

Edited September 1, 2020 by leftybassman392

[Dan Dare]

6 hours ago, Bill Fitzmaurice said:

The missing fundamental effect is real, but the harmonics that create the effect are electronically generated. The only way the speaker enters into the equation is that since it doesn't have to play the fundamental it can be smaller.

That's interesting, Bill. I take it that means spec sheets for drivers aren't always the full story. Is there a rough rule of thumb that tells you how low a given size of driver can go? I appreciate that there is more to it than size, so may be a silly question.

[Beedster]

Great thread

Psychoacoustics is where the idea of audio objectivity disappears, as has been nicely out above. If a speaker was plugged into an amp in the woods, turned up to 11, but no-one was there to hear it, would it even have made a sound.......

[Bill Fitzmaurice]

1 hour ago, Dan Dare said:

Is there a rough rule of thumb that tells you how low a given size of driver can go?

I can show you eighteens that only go to 60Hz and fives that go to 30Hz, so it's no. As for the missing fundamental, that's useful tech to a point, that point being when for whatever reason you can't use a speaker capable of going low, so  at least it seems like it goes low. You don't see it used in pro-sound, because there's a lot more to deep lows at high SPL than just what it sounds like, namely what it feels like. The missing fundamental won't flap your pants in front of your rig, or pound chests out front.

[agedhorse]

Tha bass guitar in the lowest octave produces less than 50% fundamental, the rest of the sound is harmonic series of the fundamental. Position of the pickup in relation to the bridge changes the percentages of the various harmonics (and the percentage of the fundamental in the total), as does the type of string, their material, the position of the strings with respect to the pole pieces (or magnetic fields), it goes on and on. 

A bass that just played fundamentals would sound very much like a sine wave generator (which is what it would be).

[Steve Browning]

11 hours ago, Beedster said:

Great thread

Psychoacoustics is where the idea of audio objectivity disappears, as has been nicely out above. If a speaker was plugged into an amp in the woods, turned up to 11, but no-one was there to hear it, would it even have made a sound.......

It's fascinating isn't it?

Many moons ago I had an Altec loudspeaker that needed reconing. It was an insurance job and so I took it to the chap for the work. He asked if I wanted a generic cone or the proper Altec one. I made the schoolboy error of suggesting there would be no difference and then spent the next 90 minutes listening to him tell me why that wasn't so!!