I just want to guage interest in active loudspeakers, and possible conversions of your passive loudspeakers to active.
<disclaimer> The following is a kind of advertorial, though I welcome all and any questions and crits...</disclaimer>
What is an active loudspeaker?
Beyond all the cable debates and debates about biwiring and biamping is the real deal: Active Crossovers. But first a bit about the other schemes (skip the next two paragraphs if you feel you know this).
Biwiring: the theory here is that if one separates the high frequency audio signal from the low frequency signal, then one can use wire optimized for the different bands, respectively. This is done by splitting the wires at the amplifier outputs and feeding the full-range signal to separate terminals for tweeter and woofer at the loudspeaker end. Loudspeakers that are biwirable have separate input terminals for the tweeter high frequency crossover filter and the woofer low frequency filter. The current drawn by the woofer and tweeter via their respective filter networks flows only in their respective dedicated cables.
Biamping: This takes the biwiring concept a step further upstream by splitting the full-range input signal before the amplifier and assigning dedicated amplifiers to the high frequency and low frequency circuits, respectively. This approach needs at least an extra pair of amplifiers and, of course, speaker wires. Although the full-range signal is voltage-amplified through the amplifier stage, the respective amps are only loaded by the high frequency and low frequency circuits, respectively. This means that, for instance, the amplifier dedicated to the woofer delivers current only to the woofer, etc. Biamping allows the implementer to use different-sounding amplifiers for the high and low band signals, to taste.
I will not go into all the claims made for and against the above two schemes right now, except to say that:
(1) the science behind biwiring is not widely accepted as being applicable within the audio band, and has been debated on these forums and elsewhere without any concluding consensus
(2) biamping does not yield the full benefits that a multi-amp setup can, though some have implemented a scheme whereby low-pass and high-pass filtering have been applied in order to extract a substantial amount of the insignificant portion of the respective signals (low frequencies in the high frequency amplifier and vice-versa) before amplification. This improves the dynamic headroom available in the respective amplifiers.
Electronic crossovers
In implementing an electronic crossover the setup is similar to that of biamping, except that there is no passive filter (the old crossover) between the amplifier and the drivers (tweeters, woofers, etc.). Instead, there is a line level electronic crossover that splits the signal into its respective frequency bands before entering the amplifiers.
Active Crossovers
This is simply the whole electronic-crossover-plus-amplifiers built into the loudspeaker, so that no high level signal is sent to the amplifier. The audio signal goes directly from the preamplifier to the loudspeaker, preferably in balanced form (as the low-level signal will now be more prone to noise ingress, due to the extra length of the line level interconnect). Activated loudspeakers each require their own mains supply to power the internal electronics. The cabling from the amplifiers to the drivers is now very short, such that their effect on the audio signal (as agonized over by cable-believers) is minimized. Also, in this scheme, the amplifier can exercise ?full control over the driver?s cone movement?, as agonized over by others. Most importantly, there is a major jump in system efficiency, as there is now no passive crossover to attenuate the high-level signal. The driver sees the full signal from the amplifier. This means that the amplifier does not have to work as hard, as in conventional loudspeakers, thus freeing dynamic headroom.
Crossovers do much more than simply filtering the signal into low, medium or high frequency signals. In the crossover the more efficient drivers are attenuated to bring them to the same acoustic level as the less efficient drivers. Response shaping is done so that the combined frequency response (i.e., the acoustic response of the driver, affected by the enclosure geometry and filtered by the crossover) matches the theoretical filter chosen by the designer (Butterworth, Linkwitz-Riley, second-order, fourth-order, etc.). Phase matching through the crossover region is another little-known function that is done by the crossover. All this can be done by a passive crossover.
There are further functions ? such as time delays and notch filtering - that can be done with passive crossovers, but you can see how complicated such a crossover network is starting to look. High quality passive components are not cheap, and often a compromise has to be struck between cost and complexity. This means that speaker manufacturers often opt for drivers that are ?easy? to cross over (that is, need few components in the crossover) ? typically having cones with highly-damped, smooth frequency responses. Although such a characteristic contributes to keeping the crossover component count (cost) down, my opinion is that drivers having highly damped cones are not able to resolve down to the level of detail that hard/stiff coned drivers do (all else being equal). The latter, conversely, need extra attention in the crossover because they generally suffer from harsh breakup-up modes (cone resonances) at the top end of their pass band (this phenomenon had earned metal diaphragm drivers a reputation of sounding harsh, before people learnt to build crossovers properly).
Extracting the best from metal, carbon fibre and other hard/stiff-coned high end drivers involves applying high Q notch filters to counteract the resonances, or else applying steep roll-off filters so that the breakup modes can be suppressed to a level that is benign, while still making maximum use of the flat portion of the driver?s frequency response.
It gets complicated and complex, trying to wring the best from a speaker and its main components.
Enter the digital crossover.
Imagine the ability to dial in any frequency response? any filter? equalize a wobbly curve? notch out a peak? compensate for room boundary effects? dial in time delays for true Linkwitz-Riley filters? all this is possible with digital crossovers...
This is just a primer. I?m working on a DSP-based active crossover that uses a 96kHz/24 bit DSP (digital signal processing) engine and ultra-low distortion analogue power amplifiers. What say you? interested?
<disclaimer> The following is a kind of advertorial, though I welcome all and any questions and crits...</disclaimer>
What is an active loudspeaker?
Beyond all the cable debates and debates about biwiring and biamping is the real deal: Active Crossovers. But first a bit about the other schemes (skip the next two paragraphs if you feel you know this).
Biwiring: the theory here is that if one separates the high frequency audio signal from the low frequency signal, then one can use wire optimized for the different bands, respectively. This is done by splitting the wires at the amplifier outputs and feeding the full-range signal to separate terminals for tweeter and woofer at the loudspeaker end. Loudspeakers that are biwirable have separate input terminals for the tweeter high frequency crossover filter and the woofer low frequency filter. The current drawn by the woofer and tweeter via their respective filter networks flows only in their respective dedicated cables.
Biamping: This takes the biwiring concept a step further upstream by splitting the full-range input signal before the amplifier and assigning dedicated amplifiers to the high frequency and low frequency circuits, respectively. This approach needs at least an extra pair of amplifiers and, of course, speaker wires. Although the full-range signal is voltage-amplified through the amplifier stage, the respective amps are only loaded by the high frequency and low frequency circuits, respectively. This means that, for instance, the amplifier dedicated to the woofer delivers current only to the woofer, etc. Biamping allows the implementer to use different-sounding amplifiers for the high and low band signals, to taste.
I will not go into all the claims made for and against the above two schemes right now, except to say that:
(1) the science behind biwiring is not widely accepted as being applicable within the audio band, and has been debated on these forums and elsewhere without any concluding consensus
(2) biamping does not yield the full benefits that a multi-amp setup can, though some have implemented a scheme whereby low-pass and high-pass filtering have been applied in order to extract a substantial amount of the insignificant portion of the respective signals (low frequencies in the high frequency amplifier and vice-versa) before amplification. This improves the dynamic headroom available in the respective amplifiers.
Electronic crossovers
In implementing an electronic crossover the setup is similar to that of biamping, except that there is no passive filter (the old crossover) between the amplifier and the drivers (tweeters, woofers, etc.). Instead, there is a line level electronic crossover that splits the signal into its respective frequency bands before entering the amplifiers.
Active Crossovers
This is simply the whole electronic-crossover-plus-amplifiers built into the loudspeaker, so that no high level signal is sent to the amplifier. The audio signal goes directly from the preamplifier to the loudspeaker, preferably in balanced form (as the low-level signal will now be more prone to noise ingress, due to the extra length of the line level interconnect). Activated loudspeakers each require their own mains supply to power the internal electronics. The cabling from the amplifiers to the drivers is now very short, such that their effect on the audio signal (as agonized over by cable-believers) is minimized. Also, in this scheme, the amplifier can exercise ?full control over the driver?s cone movement?, as agonized over by others. Most importantly, there is a major jump in system efficiency, as there is now no passive crossover to attenuate the high-level signal. The driver sees the full signal from the amplifier. This means that the amplifier does not have to work as hard, as in conventional loudspeakers, thus freeing dynamic headroom.
Crossovers do much more than simply filtering the signal into low, medium or high frequency signals. In the crossover the more efficient drivers are attenuated to bring them to the same acoustic level as the less efficient drivers. Response shaping is done so that the combined frequency response (i.e., the acoustic response of the driver, affected by the enclosure geometry and filtered by the crossover) matches the theoretical filter chosen by the designer (Butterworth, Linkwitz-Riley, second-order, fourth-order, etc.). Phase matching through the crossover region is another little-known function that is done by the crossover. All this can be done by a passive crossover.
There are further functions ? such as time delays and notch filtering - that can be done with passive crossovers, but you can see how complicated such a crossover network is starting to look. High quality passive components are not cheap, and often a compromise has to be struck between cost and complexity. This means that speaker manufacturers often opt for drivers that are ?easy? to cross over (that is, need few components in the crossover) ? typically having cones with highly-damped, smooth frequency responses. Although such a characteristic contributes to keeping the crossover component count (cost) down, my opinion is that drivers having highly damped cones are not able to resolve down to the level of detail that hard/stiff coned drivers do (all else being equal). The latter, conversely, need extra attention in the crossover because they generally suffer from harsh breakup-up modes (cone resonances) at the top end of their pass band (this phenomenon had earned metal diaphragm drivers a reputation of sounding harsh, before people learnt to build crossovers properly).
Extracting the best from metal, carbon fibre and other hard/stiff-coned high end drivers involves applying high Q notch filters to counteract the resonances, or else applying steep roll-off filters so that the breakup modes can be suppressed to a level that is benign, while still making maximum use of the flat portion of the driver?s frequency response.
It gets complicated and complex, trying to wring the best from a speaker and its main components.
Enter the digital crossover.
Imagine the ability to dial in any frequency response? any filter? equalize a wobbly curve? notch out a peak? compensate for room boundary effects? dial in time delays for true Linkwitz-Riley filters? all this is possible with digital crossovers...
This is just a primer. I?m working on a DSP-based active crossover that uses a 96kHz/24 bit DSP (digital signal processing) engine and ultra-low distortion analogue power amplifiers. What say you? interested?
