Richard has it correct.
Technically, phase shift always involves a time delay - the angle of the
phase shift is mathematically dependent upon the delay AND the frequency
(and by inverse relationship, the period) of any component of the signal.
In the electrical realm, if we take a signal and delay it 1ms, the result
will be 360 degrees out of phase (delayed) compared to the original signal
ONLY at 1kHz (one full period). With the same delay (1ms), a 500 Hz signal
will be 180 degrees (1/2 period) out of phase (and will cancel when summed
with the original), 250 Hz will be 90 degrees out phase, and so on. Going
in the opposite direction, 1500 Hz will be 540 degrees out of phase
(360+180) and will also completely cancel when combined with the original,
2kH will be 720 degrees out of phase, 2.5kHz will be 900 degrees out of
phase (360+360+180) and will also completely cancel when combined with the
original, and on and on. This is how we get the comb-filtering we
associate with ³phasing,² whether accidentally or artistically created.
Phase shift always involves time delay, whether caused by azimuth
problems, digital latency, capacitors or inductors.
Polarity inversion (or polarity reversal, although I think that is a more
confusing term) is merely the result of changing the connections in a
circuit so that the voltage relationship between two wires is now inverted
- what used to go positive now goes negative. The simplest example is to
interchange the wires connected to pin 2 and pin 3 on a balanced analog
XLR connector - signals that used to make pin 2 go positive with reference
to pin 3 will now make pin 2 go negative with reference to pin 3. This is
signal polarity inversion. Notice that this does not involve any time
delay, and the result is complete inversion of the signal at every
component frequency - the whole complex waveform is ³upside down.² This
is not phase shift, and there really canıt be such a thing as ³phase
reversal² or ³phase inversion.² Describing polarity inversion as 180
degrees of anything is misleading, as 180 degrees of phase shift means a
specific time delay at a specific frequency.
Associate Professor of Music
Department of Recording Arts
IU Jacobs School of Music
>When measuring phase shift in degrees, it is relative to a given
>frequency. So, a 180 degree phase shift at 1 kHz would be the same time
>delay as a 360 degree phase shift at 2 kHz.
>A polarity reversal is independent of frequency/wavelength. When it used
>to go up, it now goes down and vice versa. It is also a binary function:
>in or out of polarity.
>I think 99% of the time when we say 180 ° phase shift, we mean polarity
>flip as we want it at all frequencies to be the same.
>On 2015-05-06 8:23 PM, Tom Fine wrote:
>> So is it correct to say that there can only be a polarity reversal
>> (180-degrees) whereas there can be any degree of phase shift?
>> -- Tom Fine
>Richard L. Hess email: [log in to unmask]
>Aurora, Ontario, Canada 647 479 2800
>Quality tape transfers -- even from hard-to-play tapes.