I have added two subwoofers to the main speakers, running them in stereo mode and crossing them at 100Hz @ 24dB/oct. Since the electronic crossover used, ADS 642ix, has fixed crossover modules, not continuously variable, and there are 4 alternative modules: 65Hz, 100Hz, 130Hz, and 170Hz, all @ 24dB/oct. slope, has anyone ever tried setting the subwoofer frequency above 100 Hz? Can you describe the sound if using 130Hz or 170Hz settings?
At present, I could describe the sound character of the 100Hz setting as giving a lot of bass head punch, which sometimes causes irritation and fatiguing. I'm not sure would the problem solve if changing setting to 130Hz or 170Hz?
At present, I could describe the sound character of the 100Hz setting as giving a lot of bass head punch, which sometimes causes irritation and fatiguing. I'm not sure would the problem solve if changing setting to 130Hz or 170Hz?
If you have a left and right sub flanking the mains a higher crossover frequency will work fine.
I had a pair of subwoofers crossed at 300Hz to a pair of ribbon speakers. The subwoofers were placed in the corners of the room, the ribbons were a meter forward and two meters in from each side wall. The sound was very good and the subwoofer location never became obvious to the ear.
I had a pair of subwoofers crossed at 300Hz to a pair of ribbon speakers. The subwoofers were placed in the corners of the room, the ribbons were a meter forward and two meters in from each side wall. The sound was very good and the subwoofer location never became obvious to the ear.
I have set the sub frequency above 1000Hz in a specific configuration, but I'm not sure what your concern is in this case.
Are the subs next to the mains? In theory that alone shouldn't make any real difference provided they can deal with the range and if we're not talking about the room.
What is causing your difference, is it simply a bad cross or EQ?
Are the subs next to the mains? In theory that alone shouldn't make any real difference provided they can deal with the range and if we're not talking about the room.
What is causing your difference, is it simply a bad cross or EQ?
@AllenB yes, the subs are sitting next to the mains.
My purpose is to improve the tonal balance of the system. Now the bass seems to be played isolate from the music. They have a lot of punch at bass head, as mentioned above. The male voice seems a little bit thin. So I'm thinking to change the crossover of the subwoofer to higher frequency, in order to blend the bass to the whole music; warmer tone, less bass punch, and smoother bass.
My purpose is to improve the tonal balance of the system. Now the bass seems to be played isolate from the music. They have a lot of punch at bass head, as mentioned above. The male voice seems a little bit thin. So I'm thinking to change the crossover of the subwoofer to higher frequency, in order to blend the bass to the whole music; warmer tone, less bass punch, and smoother bass.
Do they specify what type it is? If it's a 4th order Butterworth, then it may be a bit resonant. The mains would have to be high-passed and positioned close to the subs for the relative phases to align and give the correct gain.
A softer slope like a Bessel or Linkwitz-Riley would be more forgiving, but with more high frequency leakage.
A softer slope like a Bessel or Linkwitz-Riley would be more forgiving, but with more high frequency leakage.
If the subs are symmetrical, set up like a stereo pair, you could go as high as 300Hz. (or even more)
It depends on the kind of subwoofee and construction if that's wise.
Think about port resonances, internal cabinet resonances (standing waves) as well as the performance of the woofers themselves (distortion, break-up modes etc etc)
It depends on the kind of subwoofee and construction if that's wise.
Think about port resonances, internal cabinet resonances (standing waves) as well as the performance of the woofers themselves (distortion, break-up modes etc etc)
From the above, is it correct to assume that a 100Hz high-pass 24dB/octave filter is also being applied to the main speakers?
One of the crucial parameters with low-pass and high-pass filters and summing their responses together is the filter topology.
For example, if the filters being applied were 4th-order (24dB/octave) Butterworth filters, then they would be –3dB at the crossover frequency, and they would sum to produce a 3-dB peak at that crossover frequency (when each filter has positive polarity). In your case, that's a broad 3-dB peak at 100 Hz, which may account for "a lot of bass head punch".
On the other hand, if the filters being applied were 4th-order (24dB/octave) Linkwitz–Riley filters, then they would be –6dB at the crossover frequency, and they would sum flat at that crossover frequency.
The user manual for the ADS 642ix states: "The factory installed filter modules are designed to bi-amplify satellite/subwoofer systems with 24dB/octave filters (low pass = 100Hz, high pass = 170Hz)." The fact that the high-pass filter has a higher cut-off frequency than its partnering low-pass filter seems to indicate to me that these are Butterworth filters, not Linkwitz–Riley ones. That's because separating the –3dB frequencies is an attempt to reduce the size of the 3-dB peak that would otherwise tend to occur.
Butterworth isn't worse, it's just different. In some cases it is a better choice. Of course none of this matters unless it is implemented correctly to begin with.
At woofer frequencies, the room acoustics (acoustic characteristics of the room and the placement of chair and speakers) play a large role in the sound that gets to your ear, as in the expression, "Man plans, then God laughs".
The solution today is to have adequate EQ adjustment in the system (that means DSP, which makes adjustment very simple and repeatable) and use REW. It is simply shooting in the dark to rely on theory and then work without measurement and iterative adjustment.
The solution today is to have adequate EQ adjustment in the system (that means DSP, which makes adjustment very simple and repeatable) and use REW. It is simply shooting in the dark to rely on theory and then work without measurement and iterative adjustment.
The following plot illustrates the response summation of 100Hz 24dB/octave Butterworth low-pass and high-pass filters. This is the on-axis response of a system comprised of two coincident drivers. The phase responses are not misaligned, as in order to get the +3dB peak at the crossover frequency it is necessary to have both drivers perfectly in-phase with each other. If they weren't, the –3dB responses would not sum to +3dB.
Below is the response plot when the –3dB point of the high-pass filter is shifted to 127Hz. The crossover frequency now occurs at approximately 113Hz, and the response there is approximately –5.6dB down, with about ±0.4dB ripple in the summed response through the crossover region.
Last edited:
The 4th order Linkwitz-Riley has a 3dB peak when equalised for flat power as well. Some of the talk of Butterworth crossovers was intended to be about odd orders. I failed to make that distinction clear.
However the flat power and flat response is able to be attained with any order if wanted.
However the flat power and flat response is able to be attained with any order if wanted.
Amplitude, probably. But I'd be wary of overhang and ringing. If the filter produces an impulse response with a decaying ripple on a linear graph, then consider what happens with hearing that is approximately logarithmic. So, there could be a bit boominess or droning.
In theory, each peak should perfectly coincide with a dip if the drivers have zero distance between them, but that's unlikely.
In theory, each peak should perfectly coincide with a dip if the drivers have zero distance between them, but that's unlikely.
I don't mean to be picky about your bit of humour, but I don't know which if any anechoic rooms are free of reflections in much of the bass region. Typically, the operators of such lab spaces have to use math corrections in the bass.
But back in the real-world of domestic rooms, the fine physics of FR and phase relations turns to garbage in the complexity of room acoustics and the way drivers are mounted relative to one another. Which may be one reason why many of us really like the sound of open baffles and panels: the sound gets a good stir before reaching your ears.