Audibility of allpass phase distortion (test)

[Davey]

If the electronics are "flat 20Hz-20KHz", the phase is about zero.

Each side of a crossover has phase shift. If both sides sum properly (not all do), the *combined* response should be little to no phase shift.

No, I'm afraid not.
Crossovers (with just a few exceptions) all exhibit non-zero phase distortion, even those that sum flat/properly. The audibility of the phase distortion associated with those crossover alignments is what Pavel is trying to evaluate.

Dave.

 

[PMA]

I think that PRR spoke about phase distortion in analog amplifiers. In case that the analog amplifier is properly designed, then the phase shift in 20Hz - 20kHz band is close to 0°, I agree. It should have -3dB bandwidth of some 2Hz - 200kHz to meet the requirement. However, as already said, possible amplifier contribution to added phase shift is usually negligible compared to speaker and headphone drivers. Exceptions exist, like tube amplifiers with output transformers.

 

[PMA]

I had a listen again today, this time via speakers and also with headphones as before. Again nothing worthy of note via Foobar and yet I still feel I am sensitive to changes in the sound from moment to moment. Very perplexing.

Karl, thanks for giving it a try. I have just tried the test also with speakers instead of headphones, 3 runs of Foobar ABX, and the result is quite positive, 7/10, 8/10 and 8/10. I have ABX protocols of these attempts.

It seems to me that, depending mostly on speakers and room acoustics, the allpass phase distortion may become audible with special test signals that are grossly affected in their time shape by allpass phase distortion, provided this lies in a frequency band where the ear may be phase sensitive.
I have not found, yet, a music sample where I would be able to get a positive ABX result with this allpass test.

 

[Davey]

I think that PRR spoke about phase distortion in analog amplifiers.

No, he specifically mentioned crossovers in the second part of his post.

Isn't identifying the possible audibility of an associated crossover in a speaker system your objective here?? I re-read your first post, and maybe I'm misunderstanding your premise.

Dave.

 

[PMA]

My goal is to investigate possible audibility of a general allpass system, i.e. the one with flat amplitude response but frequency dependent phase response (not meaning a simple time delay). Speaker crossover is one of the possible circuits.

I mentioned the amp for the reason that I thought it might have been blamed for phase distortion, the amp itself.

 

[PMA]

I have finally found the type of music sample where I can hear the sound difference between allpass under test and gain stage without any doubt. It is a drum kick followed by hi-hat. Try to concentrate on the first drum kick transient. One of my ABX attempts (15/15) attached.

 

[Mooly]

I listened to the drum clip but again, statistically I could not reliably differentiate. I scored 7/10 but followed that with a 2/10 and a 6/10.

The decay on these later clips sounds very 'grainy', almost like crossover distortion.

 

[PMA]

There is some background noise in the drum recording, but I wound not call it a "crossover distortion sound", unless the system used does not add its own crossover distortion. These transient signals are interesting.

 

[Mooly]

Earbuds on a laptop this time

 

[Destroyer OS]

Outofphase sounds less poppy to me on my laptop speakers.

 

[Destroyer OS]

On drums the amp one sounds metallic and sustains longer. It also pops a bit louder - more forward.