Room mode, so bad bass frequency response (with or without dipole) ;(

[mbon]

I measured frequency response of my dipole subwoofer in room with Behringer ECM8000 and it's so bad, lots of peak and deep hole in sound...
Thinking it's my dipole sub eq that is no good, I try the loudspeaker without the sub and it's the same !
Room mode are so bad in my linving room...

I've tried different placement of the dipole sub (triphonic configuration using DCX2496 for crossover) and no good results.

I'm disappointed, dipole sub does not deal so good with room mode as I can read...

using fine eq to cancel all peaks is a solution but there's so much things to correct that I'm afraid one DSP won't be enough...

An externally hosted image should be here but it was not working when we last tested it.

 

[MBK]

Hi Mbon,

how did you measure FR? Any unsmoothed FR plot in a real room will look quite awful. Most of the peaks and dips will be fairly unnoticeable to the ear though, unless they are broader than 1/3 octave. For EQ you should only concentrate on the smoothed FR plots, and the broad peaks you see there.

If the peaks and dips occur at over 200 Hz they likely are not room modes btw - they just show the various absorption qualities of your room+furniture, plus speaker and driver anomalies.

For your H frame subwoofer, there will also be a resonance peak
unless you took that out already by EQ.

For the region below 200 Hz:

Dipoles reduce room mode influence , but they don't work miracles either. I have dipoles and still have some room modes. But not to worry too much about sharp peaks, just look at broad anomalies and EQ them out. In my case, the compounded resonance was at about 70-110 Hz, plus 6 dB here. That was audible, as a coloration / boominess in the mid bass. A single analog notch filter completely cleaned up the sound.

If you look at a 1/3 octave smoothed frequency plot and you see peaks and dips of no more than +- 3 dB you will have achieved a fairly good result.

 

[mbon]

Hi Mbon,

how did you measure FR? Any unsmoothed FR plot in a real room will look quite awful. Most of the peaks and dips will be fairly unnoticeable to the ear though, unless they are broader than 1/3 octave. For EQ you should only concentrate on the smoothed FR plots, and the broad peaks you see there.

If the peaks and dips occur at over 200 Hz they likely are not room modes btw - they just show the various absorption qualities of your room+furniture, plus speaker and driver anomalies.

For your H frame subwoofer, there will also be a resonance peak
unless you took that out already by EQ.

For the region below 200 Hz:

Dipoles reduce room mode influence , but they don't work miracles either. I have dipoles and still have some room modes. But not to worry too much about sharp peaks, just look at broad anomalies and EQ them out. In my case, the compounded resonance was at about 70-110 Hz, plus 6 dB here. That was audible, as a coloration / boominess in the mid bass. A single analog notch filter completely cleaned up the sound.

If you look at a 1/3 octave smoothed frequency plot and you see peaks and dips of no more than +- 3 dB you will have achieved a fairly good result.

That's true, I use Behringer ECM8000 mic and a PC (card calibrated with freqresplot software) and I'm talking of unsmoothed FR (step is 1 Hz between 20 & 200 Hz !!!!!!!!)

But to my ear, a glissando from 20 to 200 Hz has the same irregularities, it's true that when playing music, it's not a big issue but disapointing...