REW measurements using Pink Noise v Sweep

[MediumRare]

I measured my system using the sweep method and then using pink noise. There were some pretty large differences, especially where the pink noise method showed a dramatic downslope from about 300 hz to 20k. There was also more LF nulls using the sweep method and a spike around 200hz using pink noise. The window and type of pink noise didn’t matter. Am I making a mistake somewhere in the settings? If not, how should I interpret these results?

 

[John Mulcahy]

If you were using the RTA to view the pink noise result did you have it set to an RTA mode? Pink noise will have a 10 dB/decade slope in spectrum mode.

 

[MediumRare]

If you were using the RTA to view the pink noise result did you have it set to an RTA mode? Pink noise will have a 10 dB/decade slope in spectrum mode.

I most likely did not - where is that setting? Also, would that affect just the slope, or otherwise, what would you make of the differences in nulls/spikes?

 

[John Mulcahy]

In the graph controls. It would be better to attach the mdat files for comment on the measurements. If you used random pink noise a lot of averaging would be required to get a useful low frequency response, Periodic pink noise is better and does not require averaging.

 

[MediumRare]

In the graph controls. It would be better to attach the mdat files for comment on the measurements. If you used random pink noise a lot of averaging would be required to get a useful low frequency response, Periodic pink noise is better and does not require averaging.

I tried both random and periodic with different sizes. With lower sizes I could hear pulsing, but didn’t seem to change the results much. Any comments there?

I’ll rerun the analysis and post files if there are still important differences.

 

[MediumRare]

Here we go after extensive tuning of the subwoofer phase (105 deg), crossover (75, 24 db slope) and internal eq. Also a little eq courtesy of my preamp, but not digitally.

Room is 17 W x 23 L x 9 H. Speakers are 104" apart nearly centered along the long wall (22" from wall) and both are 121" from the PLP. PLP is a little farther back than an equilateral triangle to avoid the null in the middle of the room.

There are some LF nulls I tried to correct but no matter what I did I could not fill them.

For those interested, I've included both a sweep method (orange) and moving microphone method (purple) so you can see the differences. Both are 1/12 smoothing.

Please take a look at as much as you can and let me know your thoughts. Thanks!

 

[DannyBadorine]

I suggest not trying to fill any low frequency nulls with EQ as your room is causing them. You can't fix them by adding EQ. The frequency response isn't too bad. I would EQ out some of the 600-700Hz. How does it sound?