Home Theater Forum and Systems banner
1 - 5 of 5 Posts

·
Registered
Joined
·
81 Posts
Discussion Starter · #1 · (Edited)
Hi, my first post here.

My goal is hifi stereo system for music alone. With a friend this afernoon we have experienced an unusuall response on the excess group delay plot. We think:blink:

I hope someone can clarify this one.

Two mainspeakers an one subwoofer are used, the Mic is at the listening position. We expected to locate the minimal and non minimal phase regions to practice later a room correction EQ.

Firts try, only the right main speaker was running (forgot to powered the left amplifier). The response is quite clear with a narrow and high peak between 60-70Hz, where there is a hudge hole in the frequency response. So we thought this might not be a minimal phase region and we better not try to EQ it.

We started over the whole thing again with right and left speakers on. This time the resulting excess group delay plot do not show anymore this 60-70 Hz range as a non minimal phase region, no more peak? Only little vague small bumps. We then thought that the overlapping waves neutralized the peak.

Unfortunatelly, stong distortion apears to be present exactly at this frequency range after EQ correction:rolleyesno:.

Questions.

1) Is the first excess GD plot, using only the right side speaker, was appropriate to locate non minimal phase region?
2) Is it possible that using the left and right spks could result in hiding non minimal phase regions? So triyng to EQ the frequency hole between 60-70Hz was not appropriate.

Any suggestions for newbie?

thank you all

jacques
 

·
Banned
Joined
·
516 Posts
If you have excess group delay using only one loudspeaker, the issue exists between the various energy arrival times of the various drivers, and/or, potentially, with a virtual diffractive energy source such as a baffle or cabinet or mounting edge.

Since this is only one loudspeaker, the answer is to employ an active crossover with precision microsecond delay, in conjunction with the ETC response, to determine the precise amount of signal offset in time, and to adjust the delay of the signal in the appropriate driver in order to effect the alignment of the signals relative to the measuring point. A passive crossover cannot affect this modification if the amount of delay is greater than 1 cycle at the crossover point.

If you desire a complete procedure specifically detailing the alignment procedure for two sources, in this case a full range speaker with a subwoofer (but equally applicable to any two sources), you can find the paper here and the PPT here.

If such a situation occurred between the acoustic origins of two or more loudspeakers, a delay line (either separate or integrated into a receiver, etc.) utilizing data gathered as per the procedure above with the ETC response, can be used to signal align the various signal sources.
 

·
Registered
Joined
·
81 Posts
Discussion Starter · #4 ·
Hi Wayne

We have worked at my friend house and we did not saved anaything. I'll hask him to make new ones, i left my mic over there for him to work on this week.

So I will post them when it's done.

PS: I was afraid that my explanations where not clear anough.

many thanks
jacques
 

·
Registered
Joined
·
81 Posts
Discussion Starter · #5 ·
If you have excess group delay using only one loudspeaker, the issue exists between the various energy arrival times of the various drivers, and/or, potentially, with a virtual diffractive energy source such as a baffle or cabinet or mounting edge.

Since this is only one loudspeaker, the answer is to employ an active crossover with precision microsecond delay, in conjunction with the ETC response, to determine the precise amount of signal offset in time, and to adjust the delay of the signal in the appropriate driver in order to effect the alignment of the signals relative to the measuring point. A passive crossover cannot affect this modification if the amount of delay is greater than 1 cycle at the crossover point.

If you desire a complete procedure specifically detailing the alignment procedure for two sources, in this case a full range speaker with a subwoofer (but equally applicable to any two sources), you can find the paper "" here.

If such a situation occurred between the acoustic origins of two or more loudspeakers, a delay line (either separate or integrated into a receiver, etc.) utilizing data gathered as per the procedure above with the ETC response, can be used to signal align the various signal sources.
Thank you SAC

I wil have to digest your answer and read the paper:D

Meanwhile, is it appropriate to use just one speaker to locate non minimal phase regions?

jacques
 
1 - 5 of 5 Posts
Top