First Round w/ REW -- Mixed Results

[Woochifer]

Now that I got the soundcard issues out of the way, I did a first go-round with REW and as mentioned in the title, got some mixed results. I need to more thoroughly go through the instructions, and get more used to the REW application. But, it seems like my previous settings using manual measurements produced better sounding results (keep in mind that my original BFD settings used 10 EQ filters).

After identifying the peaks and calculating the EQ filters, the REW gave me 4 EQ filters to apply. While they did indeed improve the sound over the unequalized subwoofer, the bass sounded peakier and more bloated than my original filter settings. It was also audibly louder overall, and I had to adjust the overall subwoofer level. And the graph still showed large peaks and dips. It seemed that the filters needed even more aggressive attenuation to get the bass response as flat as possible.

Ironically, I also ran the REW with my equalized settings, and it identified 6 EQ filters, some of which were very questionable. Basically, it identified EQ filters for the extreme lows and the frequencies above the crossover point. Most of these frequencies were outside of the frequency range for my subwoofer, so at least I knew that most of the operational range for the sub w/ my original BFD settings did not have large peaks.

One area of concern is that my graph using test tones and Sonnie's old Excel spreadsheet w/ the Radio Shack SPL meter correction values looks very different from what the REW frequency sweep measured. Did the correction values used on the "old analog SPL meter" REW file different from the ones on Sonnie's previous spreadsheet? My old graph w/ the equalized sub was relatively flat, but with the REW frequency sweep, it looks quite different.

I also had some trouble getting the input and output to calibrate properly. A couple of times, I got a warning that the noise level on the line was too high. Not sure if it's my laptop or the soundcard (as noted earlier, I use the SoundBlaster Live! 24-bit external USB card). I did have to unplug the AC adaptor to the laptop because it was causing noise on the soundcard's audio output (it also happens whenever I use the integrated audio outputs).

Sorry, I did not do any screen captures, so I'll tweak with the settings a bit more probably next weekend. Anyway, that's my first attempt!

[brucek]

And the graph still showed large peaks and dips

Sometimes you need several iterations.
Also if your response is not surrounding the target properly, REW has will have an unrealistic reference to suggest filters.

Basically, it identified EQ filters for the extreme lows and the frequencies above the crossover point. Most of these frequencies were outside of the frequency range for my subwoofer,

You have complete selectable control over the start and stop range to create filters. Did you set it?

looks very different from what the REW frequency sweep measured

REW sweep measurement is far more accurate. Use smoothing in REW to get a result that more closely matches a sixth octave measurement.

Did the correction values used on the "old analog SPL meter" REW file different from the ones on Sonnie's previous spreadsheet?

They were matched as soon as the three new calibration files were developed. Hopefully you don't have a really old excel file. Just check the calibration values in the file.

brucek