advertisement


Measuring and interpreting the in-room phase response and time delay of loudspeakers

The Lockwood graphs are now here:

http://jcgl.orpheusweb.co.uk/temp/Lockwood.png

I'm puzzled because the results seem quite different for the Left and Right data sets. i.e. in one case the '9xDirac' seems to have no effect, but in the other it has a very large effect at LF. So far as I can tell, I've attributed the results correctly, but maybe I've muddled something. But it implies that the room positioning has quite different effects for the L and R cases.

Results all aligned to have a time=0 reference at 1kHz.
 
Bizarre. I'm very surprised there is no difference in the time delays for the three configurations of the right speaker and am as equally surprised that the left speaker's time alignment appears to have been made worse by the 'proper' Dirac calibration. I'll re-check my export data in case I got something mixed up.
 
Bear in mind that as a general rule 'efficient' power transfer means a better 'impedance match'. So as the claimed efficiency of a transformer rises that result depends more critically on the output impedance being as assumed by the transformer designer. (Using 'transformer' in a general sense here - in this case it is between mechanical impedances speaker/air as well as for the electronic side.)

Alternatively, it could be that the 'dirac' locations you chose couple very differently to the two speakers.

Tony's point may refer to the speakers themselves varying, though. That's not something I know about!
 
From my measurements of Tannoy DC drivers I can confirm that their frequency responses can vary significantly from driver to driver. IME however this mainly applies to the frequencies covered by the HF compression diaphragm and not those produced by the cone. The frequencies below 1kHz generally match up very well and the differences tend to be of a consistent level.

I tried to be as symmetrically exacting in my mic placements for the Dirac measurements as possible, and a measurement was taken for both speakers before the mic was moved to the next location, but yes it is still possible that each speaker interacts differently with particular mic placements. My room is rectangular, however there is a single bay window on the wall behind my listening seat which is closer to being opposite the left speaker than it is the right, and the door into the room is on the side wall adjacent to the right speaker. These two structural features are presumably responsible for the different low frequency responses shown in post #114 and all other farfield frequency responses I have uploaded over the years for this room.
 
Last edited:
You must log in or register to reply here.


advertisement


Back
Top