1 - 7 of 7 Posts

#### Torzel

·
Joined
·
18 Posts
Discussion Starter · ·
Hi all;
Based on the following two graphs, I have 3 questions. Please let me know what you think.
Attached is an Impulse graph followed by an ETC graph of a full-range speaker. The questions are: 1) Whether the -ve peak in the Impulse graph around 2.5 ms, which corresponds to the +ve peak in the ETC graph means that the impact of this reflection is to create a dip in the frquency response?
2) Can then one use the combo of Impulse and ETC to predict which reflection cause dips and which cause peaks?
3) Finally, can one estimate the frequency that is affected based on the data on these two graphs?
Here are the two graphs:

Thank you for taking the time.

#### brucek

·
##### Registered
Joined
·
7,514 Posts
The impulse response contains all frequencies, so you can't be frequency specific, but the best plot to establish reflection points is the ETC (as you have shown in your second graph).

ETC shows the large direct pulse and then any reflections after that. Measuring the distance of the reflections from the main bang can be used to establish how much further the signal had to travel to reach the microphone. This knowledge allows you to determine where the reflection point might be. It could be a side wall, or a back wall, etc.

Use the REW measuring tool to tell you the distance. To use the tool simply place the cursor over the main pulse and hold down the Ctrl key and the right mouse button and drag to the pulse you want to measure. It will display the distance.

Here's an example of an ETC (showing the measure tool) where the extra distance the signal had to travel was 2.5 feet. This could be a back wall that was 1.25 feet away (2.5/2), or perhaps a side wall that required the signal to go an extra 2.5 feet to get to the mic. This would be a good spot to test treatment on to reduce the reflection.

ETC

brucek

#### Torzel

·
Joined
·
18 Posts
Discussion Starter · ·
Thank you Brucek for your reply. I understand measuring the ETC peaks to obtain distance with the technique you have described. My point was that at exactly the same time-point, there is a negative deflection (negative decibels) on the Impulse graph. Does this negative deflection suggest that the impact would be a dip in the frequency response?
Thanks again.

#### SturmMD

·
##### Banned
Joined
·
165 Posts
I suppose if you use the extreme simplification of considering only the direct wave and 1 reflected wave arriving later there will be cancellation at frequencies where time delay is equivalent to to 180 degrees of phase. But in your case (2.5ms), the first frequency where this would occur is about 200Hz.

#### stevekale

·
Joined
·
121 Posts
How do you get the measure tool? (OS-X)

The impulse response contains all frequencies, so you can't be frequency specific, but the best plot to establish reflection points is the ETC (as you have shown in your second graph).

ETC shows the large direct pulse and then any reflections after that. Measuring the distance of the reflections from the main bang can be used to establish how much further the signal had to travel to reach the microphone. This knowledge allows you to determine where the reflection point might be. It could be a side wall, or a back wall, etc.

Use the REW measuring tool to tell you the distance. To use the tool simply place the cursor over the main pulse and hold down the Ctrl key and the right mouse button and drag to the pulse you want to measure. It will display the distance.

Here's an example of an ETC (showing the measure tool) where the extra distance the signal had to travel was 2.5 feet. This could be a back wall that was 1.25 feet away (2.5/2), or perhaps a side wall that required the signal to go an extra 2.5 feet to get to the mic. This would be a good spot to test treatment on to reduce the reflection.

ETC
View attachment 13274

brucek

#### JohnM

·
Joined
·
6,709 Posts
How do you get the measure tool? (OS-X)
Same as Windows, Ctrl + right mouse button and drag.

·