Tweet
Hello All,
I have been living with my dipole 3-way system for a while now and it does sound good to me. I did post pictures a while ago. Each channel of my system consist of an RD-75 in its own baffle (400 Hz and up), 4 units of Scan Speak 25W58565 in a narrow 14-inch baffle (70-400 Hz) for the mid-range and a pair of Phoenix W-frame woofers (four 12-inch Peerless drivers). The mid drivers ‘sit’ on top of the stacked woofers.
For EQ I am using a combination of two Behringer boxes, the DEQ2496 and DCX2496 along with a separate DIY analog processor which handles all dipole EQ. The DEQ2496 does the RTA duties and the DCX2496 does all the filtering (I am using L-R 4th order), and all the notch filters and the delays. I had to use a separate analog processor, because the DCX2496 simply does not have the headroom and I needed around 30 dB at 10 Hz to EQ the dipole woofers along. Since the DCX and the DEQ are connected digitally and my CD Player uses SPDI/F connection to the DEQ2496, I only have one D/A conversion. I have built a motorized volume control using a pair of motorized 4-gand Alps Blue potentiometers.
Here is the issue I am trying to address now, - I have an efficiency mismatch between the RD-75 and the midrange array. The 4 units in my midrange array of Scan Speak 25W58565 are wired in parallel-series fashion. That is two drivers are wired in parallel and then two parallel combinations are connected in series. It appears that the mids are about 10 dB louder then the RD-75. Now, I can correct this with DCX2496, but I do not want to loose the resolution as a result of 10 dB level reduction. I can also introduce an additional resistor-divider network just before the analog dipole EQ circuit. Well, as is all my dipole EQ circuits have about 25 dB attenuators (voltage dividers) built–in.
I need to do that in order to drop the output voltage of the DCX2496. It is commonly known that in order for the digital crossover to function at its best it needs the maximum full scale input voltage. For the DCX2496 it is 22 dB or about 12 Volts. Since the DCX is a unity gain device I have this 12-volts peaks appearing at the output. Therefore I had to use voltage dividers to bring the voltage down before the dipole EQ analog circuit, which is a common low-pass shelving filter.
Another solution, I suppose, is to rewire the midrange array into the series combination of drivers. I suppose this will give me a 32 Ohm load and the consequent decrease in efficiency. Well, I need an advice here, what to do, i.e. what is the best way to bring the efficiency of the midrange in line with the efficiency of the RD-75?
Thanks,
Victor
I have been living with my dipole 3-way system for a while now and it does sound good to me. I did post pictures a while ago. Each channel of my system consist of an RD-75 in its own baffle (400 Hz and up), 4 units of Scan Speak 25W58565 in a narrow 14-inch baffle (70-400 Hz) for the mid-range and a pair of Phoenix W-frame woofers (four 12-inch Peerless drivers). The mid drivers ‘sit’ on top of the stacked woofers.
For EQ I am using a combination of two Behringer boxes, the DEQ2496 and DCX2496 along with a separate DIY analog processor which handles all dipole EQ. The DEQ2496 does the RTA duties and the DCX2496 does all the filtering (I am using L-R 4th order), and all the notch filters and the delays. I had to use a separate analog processor, because the DCX2496 simply does not have the headroom and I needed around 30 dB at 10 Hz to EQ the dipole woofers along. Since the DCX and the DEQ are connected digitally and my CD Player uses SPDI/F connection to the DEQ2496, I only have one D/A conversion. I have built a motorized volume control using a pair of motorized 4-gand Alps Blue potentiometers.
Here is the issue I am trying to address now, - I have an efficiency mismatch between the RD-75 and the midrange array. The 4 units in my midrange array of Scan Speak 25W58565 are wired in parallel-series fashion. That is two drivers are wired in parallel and then two parallel combinations are connected in series. It appears that the mids are about 10 dB louder then the RD-75. Now, I can correct this with DCX2496, but I do not want to loose the resolution as a result of 10 dB level reduction. I can also introduce an additional resistor-divider network just before the analog dipole EQ circuit. Well, as is all my dipole EQ circuits have about 25 dB attenuators (voltage dividers) built–in.
I need to do that in order to drop the output voltage of the DCX2496. It is commonly known that in order for the digital crossover to function at its best it needs the maximum full scale input voltage. For the DCX2496 it is 22 dB or about 12 Volts. Since the DCX is a unity gain device I have this 12-volts peaks appearing at the output. Therefore I had to use voltage dividers to bring the voltage down before the dipole EQ analog circuit, which is a common low-pass shelving filter.
Another solution, I suppose, is to rewire the midrange array into the series combination of drivers. I suppose this will give me a 32 Ohm load and the consequent decrease in efficiency. Well, I need an advice here, what to do, i.e. what is the best way to bring the efficiency of the midrange in line with the efficiency of the RD-75?
Thanks,
Victor
Comment