Dealing with frequency specific channel imbalance.

[abdo123]

Now i cannot tell whether it's my Speakers (they're an identical pair) or my ears, but certain frequencies are dead on image center and some are off.

With Linkwitz's (RIP) Phantom Center track on his blog it is obvious to me that there is an imbalance at high frequencies, but i cannot pinpoint which ones exactly.

I looked it up and a lot of people do seem to experience frequency specific ear imbalance, playing a sine sweep test is like having an audio roller coaster (Right, then left, then right, then left, sometime center).


How do you deal with such things? As it is obvious that a measurement microphone won't measure my ear response.

 

[AnalogSteph]

For one thing, I'd suspect room interaction before your ears being at fault.

Try playback over just one speaker right in front of you, well away from potentially problematic surfaces.

 

[solderdude]

Try the following test:

Set both speakers next to each other.
Then sweep both at the same time or first left only then right only.
Or swap L and R speaker see if the same happens.

To test ears it is best to use a headphone.
When the drivers are not angled: first put the headphone on normally and then the wrong way around (so left cup on right ear)

That should leave you with conclusive evidence.

 

[dasdoing]

are the speakers equalized at LP?
that would be the fist thing to do.
also I have a theory that the phase response is a factor. if they differ a shift will also happen towards the side that comes earlier

 

[abdo123]

are the speakers equalized at LP?
that would be the fist thing to do.
next thing would be the phase response. if they differ a shift will also happen towards the side that comes earlier

The issue occurs with or without EQ.

As for the phase response, how do you measure and analyze it?

 

[dasdoing]

The issue occurs with or without EQ.

As for the phase response, how do you measure and analyze it?

actualy this is just a theory of mine that I have not confirmed yet. I edited my post above.
but the best way to look at this is to overlay the group delay plots.
there must a an effect if one side is delayed in comparison to the other at a certain frequency. I can't tell if this is realy causing the audible effect.
Also there is the question if this is fixable at all. I just compared my system before and after phase correction wit FIR filters and the after correction doesn't look more coerent.