Announcement

Collapse
No announcement yet.

Speaker evaluation using scales

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • guitar maestro
    replied
    Originally posted by 1100xxben View Post

    When you look at spectral plots of instrumental notes, you're essentially looking at the magnitude of a bunch of sinusoids that make up the note you're hearing. If all you heard was the fundamental, then everything would sound like a whistle. Applying other harmonics, or over-tones, with varying amplitudes and phase shifts is what gives instruments their unique sounds, even when playing the "same note". It's what makes a middle-A from a piano and the middle-A from a trumpet sound different. When you measure the frequency response of a speaker with sine sweeps, you're measuring the full band response of the speaker. It's the ability of the speaker to play these upper frequencies that allows the speaker to play "different waveform types". In other words, generically speaking, if a speaker has flat frequency response across the full band, then it will be able to play a full scale as well.
    Exactly what I was referring to: get the frequency response correct, and the scales will fall into place automatically, as mostly all harmonics will be "caught" within the limit of the high frequency unit, assuming it can reproduce out to 20k, unless there are very loud, very high order harmonics that lie above 20k.

    Leave a comment:


  • fpitas
    replied
    Originally posted by billfitzmaurice View Post

    OK, use this:
    "Does this subwoofer make me look fat?"

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by fpitas View Post
    I so wanted this thread to be about weighing speakers.
    OK, use this:

    Leave a comment:


  • 1100xxben
    replied
    Originally posted by Dave Bullet View Post
    Of course we have sine wave sweeps, but it occurred to me I probably need to include some scale playing across instruments and synthesized material (different waveform types) to assess any inconsistency or unnaturalness.

    Good instruments with wide bandwidth that spring to mind would be piano and saxophone.
    When you look at spectral plots of instrumental notes, you're essentially looking at the magnitude of a bunch of sinusoids that make up the note you're hearing. If all you heard was the fundamental, then everything would sound like a whistle. Applying other harmonics, or over-tones, with varying amplitudes and phase shifts is what gives instruments their unique sounds, even when playing the "same note". It's what makes a middle-A from a piano and the middle-A from a trumpet sound different. When you measure the frequency response of a speaker with sine sweeps, you're measuring the full band response of the speaker. It's the ability of the speaker to play these upper frequencies that allows the speaker to play "different waveform types". In other words, generically speaking, if a speaker has flat frequency response across the full band, then it will be able to play a full scale as well.

    Leave a comment:


  • fpitas
    replied
    I so wanted this thread to be about weighing speakers.

    Leave a comment:


  • guitar maestro
    replied
    Originally posted by billfitzmaurice View Post
    The same applies at A3, and all but the highest octave:



    That's complicated, because the strings that aren't being hammered still produce sound. But that's not the case with a horn. This unspecified horn playing an A3 clearly shows 2nd and 3rd harmonic content that exceeds the fundamental, while even the fourth harmonic comes close:
    Thanks BF, for providing evidence.

    Leave a comment:


  • Turtle
    replied
    Originally posted by philthien View Post
    It would be nice to have a collection of scales played on a variety of instruments.
    I agree. I'm sure there are some tracks available somewhere. As an alternative, I just use a good mix of tracks that include instruments like piano, mandolin, flutes, and so on. While it may not be the most "scientific" method, I prefer acoustic instruments for initial testing of speakers, then I move on to other genres to "push" the speakers.

    Leave a comment:


  • billfitzmaurice
    replied
    Originally posted by wavefunction View Post
    Hit the low key on a piano, it is about 20 Hz for the fundamental. What you hear is mostly harmonics.
    The same applies at A3, and all but the highest octave:



    That's complicated, because the strings that aren't being hammered still produce sound. But that's not the case with a horn. This unspecified horn playing an A3 clearly shows 2nd and 3rd harmonic content that exceeds the fundamental, while even the fourth harmonic comes close:

    Leave a comment:


  • guitar maestro
    replied
    Originally posted by craigk View Post
    no they are not. i have been playing the guitar for over 40 years and you are wrong. you are getting the effect of the instrument body amplifying the harmonics. if you found a way to create more energy than you put into the fundamental you are breaking the laws of physics. you are creating a perpetual energy machine.
    Well then agree to disagree, just like the others who also disagree with you. Nothing to do with breaking the laws of physics nor perpetual energy, that's for sure, lol.

    Leave a comment:


  • philthien
    replied
    Debate aside, I had been thinking scales as test material rather recently as well.

    It would be nice to have a collection of scales played on a variety of instruments.

    Leave a comment:


  • wavefunction
    replied
    Hit the low key on a piano, it is about 20 Hz for the fundamental. What you hear is mostly harmonics.
    As to guitars, I have programmed digital tuners for electric guitars. The level of the second harmonic often exceeds that of the fundamental, even with an unprocessed guitar. Add distortion as found in most guitar amps and the harmonics go crazy. High gain amps or preamps have shaping filters before the clipper stage to reduce the level of the fundamental even further, or else the tone will be too 'farty'.

    Leave a comment:


  • craigk
    replied
    no they are not. i have been playing the guitar for over 40 years and you are wrong. you are getting the effect of the instrument body amplifying the harmonics. if you found a way to create more energy than you put into the fundamental you are breaking the laws of physics. you are creating a perpetual energy machine.

    Leave a comment:


  • guitar maestro
    replied
    Originally posted by craigk View Post

    but harmonics are not louder than fundamental frequencies.
    Sorry but you're incorrect. When plucked a certain way/style, stringed instruments like guitar and bass can have harmonics louder than the fundamental. I don't know about other instruments, but I know guitars and basses for 27 years.









    As far as using scales to "detect anomalies", I think that test would be very subjective as to the experience of the listener, and even then you'd be missing some frequencies that a particular scale would miss. In essence, trying to detect instrument harmonics within the speaker's reproduction is equivalent to frequency response in a way. If the 5th harmonic lies at say 5kHz, but the speaker has a 10dB dip there, then it would be missing and would affect the reproduced instrument timbre. If you take care of the frequency response in the first place [before trying to resort to scales/instrument timbre reproduction], then you won't have to worry about that hypothetical 5th harmonic at 5kHz; it will be there because the frequency response dictates it will be reproduced at the specified level.

    Leave a comment:


  • craigk
    replied
    Originally posted by billfitzmaurice View Post
    It depends on a variety of factors. With low frequency instruments the 2nd and 3rd harmonics can exceed, sometimes by a wide margin, the fundamentals. Useful harmonic content on average extends to at least the eighth harmonic when the fundamental is 2kHz or lower. It's the harmonics that define timbre, and at the lower frequencies even pitch. At 40Hz and lower you can't tell the pitch without the harmonics. At all frequencies without the harmonics you wouldn't be able to tell one instrument from another.
    What instruments have harmonics that exceed the fundamental ? You might have an instrument that the fundamental causes the instrument body to create a resonant frequency that is louder, but harmonics are not louder than fundamental frequencies.

    Leave a comment:


  • Turtle
    replied
    Originally posted by billfitzmaurice View Post
    It depends on a variety of factors. With low frequency instruments the 2nd and 3rd harmonics can exceed, sometimes by a wide margin, the fundamentals. Useful harmonic content on average extends to at least the eighth harmonic when the fundamental is 2kHz or lower. It's the harmonics that define timbre, and at the lower frequencies even pitch. At 40Hz and lower you can't tell the pitch without the harmonics. At all frequencies without the harmonics you wouldn't be able to tell one instrument from another.
    That makes sense; Thanks for the info. So, if that many harmonics can be heard/measure, how useful would that be in evaluating a speaker? Is there anyway to reasonably assume what those harmonics should be measured at?

    Leave a comment:

Working...
X