08-24-10: Mhedges writes
> There are commercial external crossovers for this exact purpose. I am using the now discontinued NHT X2 crossover to biamp my speakers. It is a good piece but would not be suitable for this application since it is really designed for biamping speakers with built in subs and only has low pass and high pass filters up to about 200 hz.
The problem there is that you need to duplicate the intended transfer functions of the current speaker level networks.
Most commercial solutions are not plug-and-play solutions for that.
Electrical cross-over slopes vary from the speaker's stated acoustic cross-over slopes which result from a combination of the electrical networks and driver's inherent band-pass functions, points and slopes are asymmetric to get proper summation with offset drivers, rising response and resonances are compensated for, there's usually a shelving low-pass for baffle step compensation, and the response may be intentionally non-flat to compensate for the decreased directivity above the tweeter cross-over point.
Simple cross-overs which only provide high-pass and low-pass won't do all that.
Where you have a programmable DSP unit (DCX) or plug-in cards providing the components surrounding an op-amp (the Marchand units) you need to measure the existing cross-over functions (using a program like ARTA) and duplicate the transfer function.
>One solution would be to use Harrison Labs PFMOD in line crossovers. These plug in between the amp and preamp and will provide the high and low pass filtering necessary.
That's unlikely to work well.
>The amp that you use on the tweeter needs to have no DC offset or turn on "thump" or you will kill the tweeter in a hurry since there is now no capacitor to protect it from DC
No. A small turn on transient won't run the tweeters out to their mechanical limits and a few milliwatts dissipated due to slight DC offsets won't hurt anything.
When I built my Linkwitz Orions I hung my oscilloscope off my intended tweeter amplifier that had a thump on power up, noted that the turn-on transient magnitude was less than the AA battery I used to check driver polarity, and didn't worry about it.
It's been seven years and well over a thousand power cycles since then without issue.
If you're really paranoid you can put a capacitor in series with the tweeter that are big enough to make a pole two octaves below the electrical cross-over point; or derive one of the cross-over poles from a smaller capacitor although good film capacitors cost more than replacement dome+coil assemblies for nice tweeters.
> There are commercial external crossovers for this exact purpose. I am using the now discontinued NHT X2 crossover to biamp my speakers. It is a good piece but would not be suitable for this application since it is really designed for biamping speakers with built in subs and only has low pass and high pass filters up to about 200 hz.
The problem there is that you need to duplicate the intended transfer functions of the current speaker level networks.
Most commercial solutions are not plug-and-play solutions for that.
Electrical cross-over slopes vary from the speaker's stated acoustic cross-over slopes which result from a combination of the electrical networks and driver's inherent band-pass functions, points and slopes are asymmetric to get proper summation with offset drivers, rising response and resonances are compensated for, there's usually a shelving low-pass for baffle step compensation, and the response may be intentionally non-flat to compensate for the decreased directivity above the tweeter cross-over point.
Simple cross-overs which only provide high-pass and low-pass won't do all that.
Where you have a programmable DSP unit (DCX) or plug-in cards providing the components surrounding an op-amp (the Marchand units) you need to measure the existing cross-over functions (using a program like ARTA) and duplicate the transfer function.
>One solution would be to use Harrison Labs PFMOD in line crossovers. These plug in between the amp and preamp and will provide the high and low pass filtering necessary.
That's unlikely to work well.
>The amp that you use on the tweeter needs to have no DC offset or turn on "thump" or you will kill the tweeter in a hurry since there is now no capacitor to protect it from DC
No. A small turn on transient won't run the tweeters out to their mechanical limits and a few milliwatts dissipated due to slight DC offsets won't hurt anything.
When I built my Linkwitz Orions I hung my oscilloscope off my intended tweeter amplifier that had a thump on power up, noted that the turn-on transient magnitude was less than the AA battery I used to check driver polarity, and didn't worry about it.
It's been seven years and well over a thousand power cycles since then without issue.
If you're really paranoid you can put a capacitor in series with the tweeter that are big enough to make a pole two octaves below the electrical cross-over point; or derive one of the cross-over poles from a smaller capacitor although good film capacitors cost more than replacement dome+coil assemblies for nice tweeters.