[sdiy] Harmonic bandwidth

[Richard Wentk]

On 10 Jan 2008, at 13:29, Needham, Alan wrote:

> I don't agree with that - sure there are phase shifts inherent in the
> speaker design (and amp and preamp), but they will be a constant  
> for any
> one speaker

No. Well - they'll be consistent for amplitude and phase for one  
speaker/amp system, more or less, allowing for variations in  
temperature, air pressure, humidity and listening position.

> so that will result in, say, a 10 degree shift at the 10th
> harmonic of a hypothetical waveform.

You'll get different phase shifts at different frequencies. There's  
no such thing as a perfect phase linear speaker/amp combo. Most  
speakers have wildly non-linear phase responses.

> Does this mean nobody really knows
> what a square wave sounds like?

No, but it does mean that if we're trying to hear an effect, there's  
no guarantee that everyone's listening tools will produce the same  
sound from the same input waveform.

In fact it's almost guaranteed they won't. Try running a 20-20k sine  
sweep through any speaker and listen to how (un)even the response is.

Richard