Large Null around 90Hz - Page 2 - Home Theater Forum and Systems - HomeTheaterShack.com

Old 10-30-10, 10:07 AM Thread Starter
Shackster
Stylus

Join Date: Aug 2009
Posts: 25
Re: Large Null around 90Hz

Here you go:

LEFT

RIGHT
Stylus is offline

Old 10-30-10, 12:22 PM
Senior Shackster
Bill

Join Date: Sep 2009
Location: Austin, TX
Posts: 676
Re: Large Null around 90Hz

Isn't that fascinating! Neither the left nor the right speakers alone shows the sharp null near 90Hz. Together they do.

The obvious hypothesis, then, is that you are getting negative reflections from one that are subtracting from the other. In particular, when you calculate the direct reflections from the opposite walls, can you come up with a case where the reflected distance from speaker #2 to the opposite wall and back to the listener is ~6' longer than the direct distance from speaker #1. (I know, it's hard to see how this happens if the speakers are symmetrically placed in a symmetric room, and the listener is on midline, but as the left and right curves are not identical, the room appears not to be symmetric.)

Do you have your four acoustic panels covering the first reflection points?

Here's another idea. Using REW v5, if you loopback the left channel on the soundcard, you can use it to establish an absolute reference to time zero. The phase graphs from the two speakers, measured separately, can then be compared to verify that the overall signal from one is out-of-phase with the other. This might even give a definitive answer to which signal is arriving later, and so is a longer effective distance.

Bill
laser188139 is offline
Old 11-05-10, 06:14 PM Thread Starter
Shackster
Stylus

Join Date: Aug 2009
Posts: 25
Re: Large Null around 90Hz

Quote:
laser188139 wrote: View Post
Isn't that fascinating! Neither the left nor the right speakers alone shows the sharp null near 90Hz. Together they do.

The obvious hypothesis, then, is that you are getting negative reflections from one that are subtracting from the other. In particular, when you calculate the direct reflections from the opposite walls, can you come up with a case where the reflected distance from speaker #2 to the opposite wall and back to the listener is ~6' longer than the direct distance from speaker #1. (I know, it's hard to see how this happens if the speakers are symmetrically placed in a symmetric room, and the listener is on midline, but as the left and right curves are not identical, the room appears not to be symmetric.)

Do you have your four acoustic panels covering the first reflection points?

Here's another idea. Using REW v5, if you loopback the left channel on the soundcard, you can use it to establish an absolute reference to time zero. The phase graphs from the two speakers, measured separately, can then be compared to verify that the overall signal from one is out-of-phase with the other. This might even give a definitive answer to which signal is arriving later, and so is a longer effective distance.

Bill
I'm not sure, here is a rough estimate of the layout. You are right that it is not symmetrical. The left speaker is ~5' from the left wall, the right is ~3' from the right wall. The speakers are ~8' apart and the listening position is between the speakers ~9' away. The left wall also notches in where the listening position is about 2'. So the listening position is roughly centered between the 2 walls ~6.5' on each side.

I do not have the first reflections covered, I have one panel on the front wall between the speakers and 3 on the back wall. I also forgot about 2 2'x2'x4" panels that I have on the floor behind each speaker and a cube thing I made for an alcove above a built in cabinet.

This is terrible, but here is a quick layout of the room.

Can you explain the loopback thing more, not sure I understand.
Stylus is offline
Old 11-06-10, 02:38 PM
Senior Shackster
Bill

Join Date: Sep 2009
Location: Austin, TX
Posts: 676
Re: Large Null around 90Hz

Quote:
Stylus wrote: View Post
... Can you explain the loopback thing more, not sure I understand.
In its default mode, à la REW v4, REW has no absolute reference for what is time zero. So it starts time zero in its analysis of the impulse curve with the arrival of the first sound at the mic/computer. It doesn't matter if speakers are at difference distances, each speaker is analyzed starting at its own start point. So the phase graph shows changes across the speaker's response, relative to itself, but the phase graph gives you no information to compare one speaker to another.

If you wire the left channel from the computer output back to its input, similar to the soundcard calibration wiring but this time on the left channel instead of the right, you can go into the REW Preferences -> Analysis and tell it to use the left channel as the reference for time zero. Now, REW evaluates the impulse curve relative to the looped-back impulse on the left channel, so it can give you absolute time measures to each speaker (via the group delay graph) and absolute phase measures that include the distance to the speaker. So you can see whether the phase of the signal from one front speaker at a specific frequency matches the phase of the signal from the other front speaker at the at the same frequency. (I used this with REW v5 to determine if my sub distance was correct. I described the Preferences dialog here.)

If you measure each front speaker this way, and you see that the phase difference between the two is near 180°, you can be satisfied that this is the explanation of your dip. I think it would be a reasonable inference that whichever speaker is giving you the later time in the group delay at this frequency is the one whose reflection is subtracting from the other. (This inference might be a stretch; I've not tried this myself.) The phase graph for each might also be interesting on this question: if one speaker's phase curve appears smooth around 90Hz and the other changes behavior near 90Hz, the one with the anomalous behavior should be the one whose reflection is causing the dip.

Bill
laser188139 is offline

 Bookmarks

 Tags 90hz , large , null

Message:
Options

## Register Now

Random Question
Random Question #2

User Name:
 PLEASE READ BELOW PRIOR TO ENTERING AN EMAIL ADDRESS! ATTENTION! YOU MUST ACTIVATE YOUR ACCOUNT!Activation requires you reply to an email we will send you after you register... if you do not reply to this email, you will not be able to view certain areas of the forum or certain images... nor will you be able download software. AN INVALID EMAIL ADDRESS WILL CAUSE YOUR ACCOUNT TO BE DELETED! See our banned email list here: Banned Email List We DO NOT respond to spamcop, boxtrapper and spamblocker emails... please add @hometheatershack DOT com to your whitelist prior to registering or you will get nowhere on your registration. Email Address:
OR

## Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.