visualize flutter echo with the waterfall section

[Benoit Petit]

Hello everyone!

I am a freelancer PA technician who used to work on tuning capture and Smaart but I heard about REW and found it really useful !

I try to improve the acoustical comportement of a venue in Belgium.
There are a huge simple echo (EDIT : not a flutter as written in the title) at some places into the venue (due to the architectural form) . The venue is a kind of a circle... and to be able to reach the last row with the PA, we have to overshoot a little bit.
It result as a huge echo for some people.

I have some solutions to solve it but I would like to measure properly the effect of my solutions.
it is were REW seems to be the right tool but... I am not able to measure what I want :

On the impulse section I clearly see the second big peak of energy coming from the wall. (the first one is the oposite side of the PA)
But in the waterfall or spectrogram section I can not see it... It seems that the window parameters does not affect the waterfall and spectrograph parameters...
Do you confirm that?
What I understand is that REW take the biggest peak as a start for the spectrogram and the waterfall. So I can only see the reflection (because the reflection is louder than the speaker).
Is that correct?
So, does anyone has a solution to visualize the first AND the second peak in the waterfall?

Thank you!

 

[sm52]

If you increase the resolution on the horizontal axis in the 'impulse' tab, I think you will see all the reflections close to the peak of the impuls. On the Spectrogram tab, do the same for the vertical axis.

 

[Benoit Petit]

Hello! thank you for the quick reply!

Actualy, I used the term "flutter echo" but I should have written "simple echo" . (I just did the edit in the post)
The only echo I have, is the biggest peak on the impulse tab. So, this is the only peak I can visualize into the waterfall.
What I would like to visualize is the first peak and the second one into the waterfall.
Even in the spectrogram, with a longer "span before peak" I can not see the first peak of energy...

 

[sm52]

Then download the .mdat measurements file here. I'll take a look and others will take a look. The file contains a lot of information.

 

[Benoit Petit]

thank you again!
Here is the file.
I also let you a picture showing how I did the measurement .
The thiner arrow is the first peak. the thicker arrow is the reflexion on the wall and the bigger peak.

 

[John Mulcahy]

You could use the wavelet spectrogram to see the early peak.

 

[sm52]

As far as I understand, 10-13 m is added to the direct distance from the sound source to the microphone due to reflection. If between the microphone and the sound source is 35 m, then from the sound source to the place of reflection on the wall and then to the microphone is 45-48 m.

 

[Benoit Petit]

Thank you John! that was exactly what I wanted to see! I didn't know about the wavelet transform... my bad!

Sm52, you are right. I knew what generated my problem, but I wanted to be sure that I measure the proper reflexion so that I would be able to see clearly the difference aftertrying to fix de situation.

Here is the proper spectrogram before and after putting huge curtains on the wall. but since the problem is mostly around 700hz and 3k, I will try to diffuse it instead of absorbing it.

Thank you again for you help!

Rew is a great software and has a great community!

 

[John Mulcahy]

Just as a footnote: in a space any larger than you are currently measuring you should increase the length of the impulse response that is retained and use longer sweeps, the IR length to retain is in the Analysis preferences, "Truncate IR after...".

 

[sm52]

I don't see a problem at 700Hz and 3kHz. There is a reflection, from 200 Hz and below after 30-40 ms. Red spots.