Finding Peaks

[fouiz]

Hello,
seems that it is not possible to use this function over 500Hz (i intend to do measurements on the full range to set up a DEQ2496).
correct ?

[brucek]

seems that it is not possible to use this function over 500Hz (i intend to do measurements on the full range

The find peaks function is designed to create filters to cancel modal resonances. They simply don't exist over 500Hz.

There is nothing stopping you from entering your own manual filters into REW though. Be sure that any filters at those higher frequencies have a low Q or you'll be wasting your time.

brucek

[fouiz]

brucek wrote:

Be sure that any filters at those higher frequencies have a low Q or you'll be wasting your time.

brucek

OK for the modal resonances.
Now, can you be more precise about the Q setting ?

[brucek]

can you be more precise about the Q setting

When you take a full range measure you will see a very jagged looking response when the frequency gets above 500Hz. In fact it often looks so bad, many think they've made a mistake. They haven't. The jaggedness is a result of comb filtering reflections.

To get a better feel for the underlying trend of the signal (and to come a little closer to how we actually hear) you need to turn on smoothing feature. Use 1/6th or 1/3rd octave smoothing. (remember, smoothing is not for low frequency measurements as it would obscure the true shape of the response and so not allow accurate correction filters to be determined).

Now you can see the response much better when the smoothing is on. This also reveals that it would be pretty silly to start using narrow bandwidth (high Q) filtering. You really need to use the type of filter a typical equalizer would provide (i.e. low Q, 1/3 octave or greater).

brucek

[Wayne A. Pflughaupt]

fouiz wrote:

Hello,
seems that it is not possible to use this function over 500Hz (i intend to do measurements on the full range to set up a DEQ2496).
correct ?

Here's a thread on full-range EQ project that one of our members embarked on a few months ago - you mght find some good info in it.

http://www.hometheatershack.com/foru...ase-chime.html

Regards,
Wayne

[fouiz]

Thanks Wayne.
Don't you think it would be useful to split the forum in two parts : one for sub eq and another one for full range eq ?
Because it seems to me that REW can be usefull for both and not only subs ...

[Wayne A. Pflughaupt]

Absolutely REW is useful for both! We see very few people attempting full-range equalization, though - not enough to justify a separate Forum. Spridle was the last one I know of who opened a thread on the topic (the link above), and as you can see that was over five months ago.

Regards,
Wayne

[fouiz]

OK, so, i'll be next !
I'll try yo post every action since now, as i'm really starting to work on it now, and still discovering the soft.
I'll need help, though, as i'm not familiar with acoustics, and not that good in english too.
Hope you masters of the forum have a little time to light the trail !

I'm experimenting with equalizing my full-range speakers using the Inguz plugin for Slimserver (my source is a squeezebox 3). I'd been using the program drc to generate the correction filters, but I'm finding that the filters I make with REW give results that are at least as good, and it's much easier to adjust and check results with it.

Without knowing very well what I was doing I seem to have fallen into the good practices described in the thread linked to above - that is, I'm using pretty broad filters at high frequencies, since I quickly discovered the futility of trying to get rid of every peak. The improvement is greatest in the bass, but I've found the mid and treble are improved too (which I think is more a question of achieving the balance I like best than of equalizing to a flat response).

I'm curious to learn more about the differences between drc and REW. The algorithm drc uses to generate correction filters is much more complex than just adding some peak filters - it does fancy things with phase, for example. But in the end the results don't seem to be audibly better. Still, I wonder what a program combining the ease of use of REW and the horsepower of drc might be able to do.

At the moment things are stalled while I wait for a longer mic cable to arrive (as things are now the measurement process is a real pain, and the cable will make it much easier). I'll report back when I've made some more progress.