Hello Everyone from Italy,
I am planning to build passive crossovers for my speakers which I built a LONG time ago out of pro, high efficiency, RCF louspeakers 18" woofer L18.551 and mid/high N580+6422 horns. Efficiency for the woofer is of 96db and the 102 for the hornloaded mid/high, and nominal impedance of the drivers is 8 Ohm. These speakers were built upon the original RCF design that was provided for this particular combination.
So far I biamped these speakers, with two 70watts solid state amps, but I recently purchased a tube amp (100+100 w) and since I am not happy about hybrid biamping (SS+Tube) I decided to equip these speakers with passive crossovers so I can enjoy the tube amp on the whole range.
Although those drivers are long since discontinued I have kept the tech specs, and the original RCF project for the crossover (800hz) which I am posting here below.
RCF also provided the recommended design of the inductance for this particular combination
The inductors are pretty large with a thick wire (1.6mm), and those that I have found online that fulfills these requirements are expensive. I assume the thick wire gauge is possibly due the necessity to match the power handling of these speakers of 300 watts continuous. Since I am using these speakers as hifi speakers (and I am so happy how they sound) in a 20x17ft listening room, I seldom if ever drive them with more than 15-20 watts, also due to the 100db/w/m average efficiency, and even if and when I would decide to go deaf, the amplifier which is going to drive these speakers has an output of 100w per channel.
For all these reasons I would consider a total waste of money investing in an inductor that can handle 300 watts.
However, since I am aware of the impedance etc, related drawbacks when using too thin wire, and the different types of inductors (of which I don't know nothing and have no idea which type would best fit the project) I need advice about the inductors that would fit this speakers/amplifier combo.
I guess its important to mention that the minimum impedance recommended for the amplifier is of 8 Ohm, and that like most tube amps it might be not so linear on a "real" load.
It is also worth to mention that RCF designed these horns so that the mid/high drivers are exactly one full wave lenght (@800hz) back from the woofer center of emission, so there should be no phase rotation at the recommended crossover frequency.
Any help in sorting out this matter would be greatly appreciated !
I am planning to build passive crossovers for my speakers which I built a LONG time ago out of pro, high efficiency, RCF louspeakers 18" woofer L18.551 and mid/high N580+6422 horns. Efficiency for the woofer is of 96db and the 102 for the hornloaded mid/high, and nominal impedance of the drivers is 8 Ohm. These speakers were built upon the original RCF design that was provided for this particular combination.
So far I biamped these speakers, with two 70watts solid state amps, but I recently purchased a tube amp (100+100 w) and since I am not happy about hybrid biamping (SS+Tube) I decided to equip these speakers with passive crossovers so I can enjoy the tube amp on the whole range.
Although those drivers are long since discontinued I have kept the tech specs, and the original RCF project for the crossover (800hz) which I am posting here below.
RCF also provided the recommended design of the inductance for this particular combination
The inductors are pretty large with a thick wire (1.6mm), and those that I have found online that fulfills these requirements are expensive. I assume the thick wire gauge is possibly due the necessity to match the power handling of these speakers of 300 watts continuous. Since I am using these speakers as hifi speakers (and I am so happy how they sound) in a 20x17ft listening room, I seldom if ever drive them with more than 15-20 watts, also due to the 100db/w/m average efficiency, and even if and when I would decide to go deaf, the amplifier which is going to drive these speakers has an output of 100w per channel.
For all these reasons I would consider a total waste of money investing in an inductor that can handle 300 watts.
However, since I am aware of the impedance etc, related drawbacks when using too thin wire, and the different types of inductors (of which I don't know nothing and have no idea which type would best fit the project) I need advice about the inductors that would fit this speakers/amplifier combo.
I guess its important to mention that the minimum impedance recommended for the amplifier is of 8 Ohm, and that like most tube amps it might be not so linear on a "real" load.
It is also worth to mention that RCF designed these horns so that the mid/high drivers are exactly one full wave lenght (@800hz) back from the woofer center of emission, so there should be no phase rotation at the recommended crossover frequency.
Any help in sorting out this matter would be greatly appreciated !
The thicker the inductor wire, the lower the dc resistance of the inductor.
Low dcr will ensure minimal power loss in the inductor, as well as rapid damping of unwanted driver cone movement.
Low dcr will ensure minimal power loss in the inductor, as well as rapid damping of unwanted driver cone movement.
This will be fine.
https://www.parts-express.com/Jantzen-1065-2.2mH-18-AWG-Air-Core-Inductor-255-268
Don't be too worried about the phase response, since that crossover topology can't have perfect phase.
https://testhifi.com/2019/07/05/loudspeaker-phase-response/
https://www.parts-express.com/Jantzen-1065-2.2mH-18-AWG-Air-Core-Inductor-255-268
Don't be too worried about the phase response, since that crossover topology can't have perfect phase.
https://testhifi.com/2019/07/05/loudspeaker-phase-response/
Last edited:
No. All the impedances are complex. Add excess resistance and the response shapes are altered. Thin wire in the woofer coil will make woofer down overall except at resonance (more "boom"). Thin wire in the tweeter makes no difference in the passband but spoils the 12dB/Oct slope down in the stop band.
Perhaps you could go skinny in the tweeter coil since you play not-loud and the residual bass won't really blow a PA-grade driver (will raise IM a bit).
The coils that Rayma links are probably as "cheap" as is worth buying.
Attachments
Thank you for the help. Those coils are really inexpensive ! I hope I can find then in some EU shop as well.
I browsed the parts express website and I found this one with a thicker wire which is also within my budget.
https://www.parts-express.com/Jantzen-5010-2.2mH-15-AWG-P-Core-Inductor-255-108
But I see its a different type. Would it be worth choosing this one instead ?
Rayma, with regards to phase, do you think the phase shift of the crossover would worsen the overall performance of the speakers ? Bi-Amped they reproduce beautiful broad and deep soundstage with nicely focused instruments, and the tube amp has improved this even more.
I wouldn't want to lose any of this due to the phase shift introduced by the crossover.
I browsed the parts express website and I found this one with a thicker wire which is also within my budget.
https://www.parts-express.com/Jantzen-5010-2.2mH-15-AWG-P-Core-Inductor-255-108
But I see its a different type. Would it be worth choosing this one instead ?
Rayma, with regards to phase, do you think the phase shift of the crossover would worsen the overall performance of the speakers ? Bi-Amped they reproduce beautiful broad and deep soundstage with nicely focused instruments, and the tube amp has improved this even more.
I wouldn't want to lose any of this due to the phase shift introduced by the crossover.
A crossover actually corrects phase shift. The point which has been made, is that the crossover should work together with what you have acoustically and this includes the way the driver responses aren't flat to begin with. This is the same whether you biamp or use any kind of crossover.
What kind of filter do you use in the biamped version? If it's also a second-order analogue filter, then its phase response will be very similar to that of the passive variant.
You were looking for a shop in the EU. My favourite loudspeaker DIY shop (only a webshop now, they used to have a real shop here in Haarlem, the Netherlands, until April 2020) has several 2.2 mH inductors, not all in stock though:
Air core inductors:
https://shop.speakerenco.nl/jantzen...ize=12&viewMode=grid&orderBy=15&pageNumber=81
This is probably not a real air core inductor, Intertechnik uses a small ferrite core from 1 mH onwards if I remember well:
https://shop.speakerenco.nl/220-mh-spoel-rond-140-mm-lu7830-1
Ferrite core inductor:
https://shop.speakerenco.nl/220-mh-095-mm-hq4030
They also have capacitors.
Air core inductors:
https://shop.speakerenco.nl/jantzen...ize=12&viewMode=grid&orderBy=15&pageNumber=81
This is probably not a real air core inductor, Intertechnik uses a small ferrite core from 1 mH onwards if I remember well:
https://shop.speakerenco.nl/220-mh-spoel-rond-140-mm-lu7830-1
Ferrite core inductor:
https://shop.speakerenco.nl/220-mh-095-mm-hq4030
They also have capacitors.
It's a Cabre AS45, an active crossover that was made in Italy, from the same geological era as the speakers (LOL). The crossover frequencies can be selected through microswitches and the slope is 12db octave. Apparently it is still appreciated nowadays. I honestly don't feel the need to replace it, other than the desire to let the my new tube amp drive the speakers throughout the whole range.
https://www.diyaudio.com/community/threads/cabre-as45-active-crossover.296046/
Browsing the web I have come across this EU website which apparently has the whole range of the Jantzen inductors in stock.
What are the differences with the different kinds of cores ? Ferrite, air, p-core etc ? The original design by RCF has a brass core. Wonder how the core type impacts on the crossover behavior, sound quality etc...
Air core inductors have practically no distortion, while ferrite core inductors have a small but measurable distortion. Ferrite core inductors have less DC resistance for a given wire thickness, so they have less effect on the damping of the fundamental resonance. The differences are quite subtle. No idea what the advantage of a brass core is.
As the active and passive filters are both second order (12 dB/octave), assuming equal cross-over frequencies, the phase aberrations should be very close to what you have now.
As the active and passive filters are both second order (12 dB/octave), assuming equal cross-over frequencies, the phase aberrations should be very close to what you have now.
Many tube amps are lacking in bass quality (and I have used many tube amps), but you can try it and see.
Swap the tube amp to the bass, and the ss amp to the horn, and see what happens.
My guess is that you'll prefer ss bass and tube horn. And the speaker level passive crossover will just muck things up.
Which tube amp do you have?
.
Last edited:
Well, I know I am going against the tide, but currently I am driving the woofers with the tube amp. At first I put the tubes on the horns, but I realized a good deal of the "warmth" and soundstage of the tube amp was lost that way, possibly due to the rather high crossover frequency of 900hz. I agree bass have a little less control with the tube amp vs SS, but not much, possibly due to the woofer suspension being rather stiff, or perhaps due to the resonance being as low as 20hz. Anyway it's a trade-off that I am willing to take, but I guess if the cross was lower I'd probably go for SS on the bass.
The amp is a Marshall 9200. It's a dual block, 100watt per block guitar and PA power amp, and I know it's not born as an hifi amp.
However, it has as a very flat frequency response, well beyond the audible range, and if bias AND BALANCE are properly adjusted the distortion is perfectly fine for a tube amp. The power tubes are 6L6 which can be replaced with EL34, but I prefer the warmer smoother sound of the 6L6's which I find to perfect match for the speakers which may use some taming since like most high efficiency units can sometimes be a little harsh/edgy. I'm really looking forward to drive the whole range with the tube amp.
I agree the passive crossover may muck things up, that's my worry and reason why I am seeking help about it. Viable alternative would be another tube amps for the horns, possibly a small single ended...
I started wondering about this possibility when searching the inductors what I found was about 150 each...
Last edited:
You can increase damping if you need to in a number of ways. The first you can try is the 4 ohm tap.
Sometimes a compression driver is more of a challenge, but when you get the response right then it will be the same.
As I said in the opening post minimum impedance is 8 Ohm on this amp, so no 4 Ohm tap. It's got 8 Ohm and 16 Ohm plugs and a switch to select them. Two plugs for each, so maybe it can stand a 4 Ohm load but I'd rather not mess with stuff which may damage the amp.
With regards to crossovers and compression drivers, I am a total newb with passive crossovers. The original design by RCF seems pretty basic to me. It could be it is so because they assumed the combination was supposed to be aimed at PA, or maybe because it's all it takes.
Back in the days RCF was very accurate about their products (and still is to some extent). Only producers that I know taking into account the length of the horns for correct phasing are JBL and RCF, as I find it hard to believe that the crossover frequency matching the placement of the horn drivers within millimiters (when woofer and horn are installed on the same plane) is not by chance. While this shows an unusual attention to details, it doesn't mean the original crossover design can't be improved because one thing is phase and another thing is frequency response.
What would you suggest in order to design a crossover which doesn't sound worse than my current active setup ?
I don't know how good your current crossover is so I would consider measuring. For example you could start by doing each way separately. If you can simulate then put that data in and look for issues with each way, and with how they combine. Otherwise measure them together and do the same.
By measuring you mean checking the frequency response of each driver separately ? How far from the speakers and into the listening room ? Of course I don't have an anechoic chamber.
I have taken measurements many times previously, for room correction purposes, with a (Rotel) third octave real time spectrum analyzer with pink noise generator and the microphone it came with.
However since after loaning the device it came back basically broken I have ordered a Behringer ECM8000 microphone to use with the Friture software spectrum analyzer.
Not sure if I can use the ECM8000 with REW because REW requires a calibration file for the microphone, which Behringer do not provide. The ECM8000 is good enough for room correction because it seems to be pretty flat as is without any calibration file.
With regards to the drivers frequency response, from previous measurements I took in the listening room, the response is flat at and around the crossover frequency. The horn drivers start showing a dip at about 5 k, and the response becomes questionable at 16k.
For this reason I am also planning to add a horn loaded super tweeter (RCF 1411-M mated to JBL 2404 "baby buttcheek" horn) and turn the speakers into a three way system.
But I assume this better be the subject of a specific thread, unless the mid/high crossover requires being designed also based on the low/mid crossover.
I have taken measurements many times previously, for room correction purposes, with a (Rotel) third octave real time spectrum analyzer with pink noise generator and the microphone it came with.
However since after loaning the device it came back basically broken I have ordered a Behringer ECM8000 microphone to use with the Friture software spectrum analyzer.
Not sure if I can use the ECM8000 with REW because REW requires a calibration file for the microphone, which Behringer do not provide. The ECM8000 is good enough for room correction because it seems to be pretty flat as is without any calibration file.
With regards to the drivers frequency response, from previous measurements I took in the listening room, the response is flat at and around the crossover frequency. The horn drivers start showing a dip at about 5 k, and the response becomes questionable at 16k.
For this reason I am also planning to add a horn loaded super tweeter (RCF 1411-M mated to JBL 2404 "baby buttcheek" horn) and turn the speakers into a three way system.
But I assume this better be the subject of a specific thread, unless the mid/high crossover requires being designed also based on the low/mid crossover.
I use to make coils according to a calculator similat to this one : http://www.nessengr.com/technical-data/brooks-coil-and-calculator/
an i simply glue several plywood circle one over the other for the core, then dip in varnish and bake in a oven.
an i simply glue several plywood circle one over the other for the core, then dip in varnish and bake in a oven.
Try measuring at 1m. Since it's the speakers you want to measure, this is different to the 1/3 octave that include the room contribution.
Ok, I will take the measurements when the microphone is delivered.
May I ask why compression drivers are more of a challenge ?