My opinion is that much depends on the speakers you are going to use the xover with, the quality of the rest of the system also.
Assuming for the discussion that maximum performance is required, I'd look closely at doing the HP in passive entirely. Perhaps modifying the amplifier for the purpose as well.
My personal feeling is that in general high quality speaker level xover components sound better than most opamp implementations - especially above the usual sub & woofer frequencies (<200Hz. aprox).
In the LP section, you can clearly hear subjective differences in different opamps. The 5532 sounds very different there than does the OPA2604, for example. In my 65Hz. 4th order LP I used a combination of the two in order to get the "best compromise" in the subjective sound on my Quadripole Subs (tm BEAR Labs) ;_)...
If I had to do an opamp based HP xover, I'd look strongly at doing it as a "split passive" EQ design, rather than the usual Sallen-Key or similar feedback filter type of circuit.
The advantage to a descrete circuit is that you can run higher B+, which can be a benefit for headroom. Incautiously built opamp filter circuits run into a problem with headroom due to gain in the filter sections.
If ur system is just moderate in terms of ultimate maximum high-end quality, you might just go and buy a new or used commercial xover - they're plentiful and cheap these days. Swap the opamps if you want.
_-_-bear
Btw & PS... I'd be very surprised if that discrete circuit actually produces a proper L&R slope.... since there is no gain in either section... it's a 4th order of some slope, that it is... not my fav idea for such a circuit. Others may chime in and disagree...
Assuming for the discussion that maximum performance is required, I'd look closely at doing the HP in passive entirely. Perhaps modifying the amplifier for the purpose as well.
My personal feeling is that in general high quality speaker level xover components sound better than most opamp implementations - especially above the usual sub & woofer frequencies (<200Hz. aprox).
In the LP section, you can clearly hear subjective differences in different opamps. The 5532 sounds very different there than does the OPA2604, for example. In my 65Hz. 4th order LP I used a combination of the two in order to get the "best compromise" in the subjective sound on my Quadripole Subs (tm BEAR Labs) ;_)...
If I had to do an opamp based HP xover, I'd look strongly at doing it as a "split passive" EQ design, rather than the usual Sallen-Key or similar feedback filter type of circuit.
The advantage to a descrete circuit is that you can run higher B+, which can be a benefit for headroom. Incautiously built opamp filter circuits run into a problem with headroom due to gain in the filter sections.
If ur system is just moderate in terms of ultimate maximum high-end quality, you might just go and buy a new or used commercial xover - they're plentiful and cheap these days. Swap the opamps if you want.
_-_-bear
Btw & PS... I'd be very surprised if that discrete circuit actually produces a proper L&R slope.... since there is no gain in either section... it's a 4th order of some slope, that it is... not my fav idea for such a circuit. Others may chime in and disagree...
The LC Audio circuit is fine, you don't need gain to implement a Sallen-Key filter. There is the unity-gain version (this one) and the equal-components version (with gain).
Whilst the headroom comment is valid, I'd be surprised if anybody ran into a problem in real life.
I would go with the op-amp one purely because there is no DC offset with it. Also it works out cheaper, smaller and easier to build, as well as having better PSRR.
Whilst the headroom comment is valid, I'd be surprised if anybody ran into a problem in real life.
I would go with the op-amp one purely because there is no DC offset with it. Also it works out cheaper, smaller and easier to build, as well as having better PSRR.
If you are building your own power amps it can be attractive to add discrete crossover buffers/opamps to that PCB. You can use the same high voltages to get more headroom than a 12V opamp, and you may be able to reduce the total noise. Many power amps use JFET inputs for low noise with a high impedance input. If you put discrete JFET crossover opamps on the PCB you may be able to use higher gain bipolar input devices for the power amp driven by the low impedance discrete opamp output and get lower total noise.
Buy off the shelf product like Behringer DCX2496 or DBX driverack 260, the former is quite cheap, as if you build yourself the cost will be similar, the current price is about $250 each.
It has been used extensively by many DIYers here just do a search. I am getting one now. The unit is fully digital esp. the filtering part and then DAC and output to a max of 6 outputs. It is used mainly as X-over
cheers..
It has been used extensively by many DIYers here just do a search. I am getting one now. The unit is fully digital esp. the filtering part and then DAC and output to a max of 6 outputs. It is used mainly as X-over
cheers..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.