I have a pair of BESL S5-MTM monitors crossed over to a pair of Rocket subs at 80 Hz. The crossover is a Marchand XM44, with 12dB BW slope for high pass and 24 dB L-R for low pass. I'm not unhappy with the XM44, but it isn't quite on the same level as my Meitner amps. Would changing out the XM44 for an XM46 passive be likely to improve the sound of my system?
I'm a little leery of passive devices after experimenting with a passive preamp. It was wonderfully transparent, but lacking in musical drive (pace and rhythm) compared to active. Maybe the XM46 would have a similar problem?
It would seem to me the XM46 would be sensitive to load. My amps have an input impedance of 22K and the sub amps are 15K. That would probably need to be incorporated in the design? Also would there be any problem with using 20-foot interconnects?
I'm a little leery of passive devices after experimenting with a passive preamp. It was wonderfully transparent, but lacking in musical drive (pace and rhythm) compared to active. Maybe the XM46 would have a similar problem?
It would seem to me the XM46 would be sensitive to load. My amps have an input impedance of 22K and the sub amps are 15K. That would probably need to be incorporated in the design? Also would there be any problem with using 20-foot interconnects?
I meant to say XM46 in the title, not XM44.
I want to maximize the quality of my crossover, because it's in the middle of my system. Any suggestions? opinions? experiences? rumours? anyone?
I've learned more about what counts in audio here than anywhere, by far. Don't let me down now.
I want to maximize the quality of my crossover, because it's in the middle of my system. Any suggestions? opinions? experiences? rumours? anyone?
I've learned more about what counts in audio here than anywhere, by far. Don't let me down now.
Just in case anyone cares, I did buy a Marchand XM46, but I bought the XM46SB, which just has two channels. The two channels are connected ahead of my main amp left and right channels. -3dB @ 80 Hz, second-order BW. The passive XM46SB replaced the active XM44 on the high pass only, I left the active crossover on the woofers, -3dB @ 80 Hz, third-order BW, which results in LR4 slopes for both. I wasn't sure about going all passive because of the long run of cable to my sub amps.
I love the XM46SB, it's as transparent and invisible as anyone could want. The XM44 was not invisible, it definitely left a thumbprint on the high pass, and was compromising my amps. I don't know how the XM46 would be as a low-pass, with the signal having to travel through the humungous inductor, but as a high-pass filter, it's as close to ideal as practically possible. And at a great price.... Highly recommended.
I love the XM46SB, it's as transparent and invisible as anyone could want. The XM44 was not invisible, it definitely left a thumbprint on the high pass, and was compromising my amps. I don't know how the XM46 would be as a low-pass, with the signal having to travel through the humungous inductor, but as a high-pass filter, it's as close to ideal as practically possible. And at a great price.... Highly recommended.
The passive marchand crossover is excellent. As audiophile as it gets. I complemented the low pass with Marchand´s BASSIS as an equalizer (it´s intended for sealed bass boxes) and am very satisfied. I recomend it. The bassis has op-amps and there is a trace of hiss that disappears by placing the low pass filter after the active circuit.
how so?
I had several marchands (of the active type), and run with behringer dcx now. I could not detect any "audio quality" difference between the marchand and the behringer except that there is more noise from the marchands, while the behringers are totally silent.
BTW - to me the terms "pace and rythm" are the product of audiophile scoundrels like those at stereophile. They are a quality of the music, not of the equipment.
The approbriate terms for equipment imho are: attack, sustain and decay, distortion and frequency response. The prat stuff is claptrap.
Ditto on "pace and rhythm". Marchand is not very far from me; I use his active crossovers and think his stuff is as good as you could ask for, not to mention priced very reasonably. You can put whatever opamps you want in the active units, and the passive ones should be fine save for interconnect length. IMO, a passive preamp or crossover system should be kept extremely compact. More than a few feet, and I'd go active.
I also own a XM44 (2 way, 150 hz 4 order LR). Was planning to experiment a bit with a PLLXO sometimes in the future (for the high pass part).
One way would be to use a second order PLLXO that coincides with the normal roll off of the mid sections (6.5" CQuenze in 12 l sealed). Probably around 110 hz). The low pass would remain on the XM44 (cant imagine those active sections do a lot of harm here, bass amp is a Crown K2).
Of course using this XM46SB might also be possible. Is it possible to change the frequenty, or is it only suitable for one?
A completely different direction is to go digital (eg with something like the Ground Sound, the new DCN28 if it is available) and also do some room corrections (probably only on the low channels).
One way would be to use a second order PLLXO that coincides with the normal roll off of the mid sections (6.5" CQuenze in 12 l sealed). Probably around 110 hz). The low pass would remain on the XM44 (cant imagine those active sections do a lot of harm here, bass amp is a Crown K2).
Of course using this XM46SB might also be possible. Is it possible to change the frequenty, or is it only suitable for one?
A completely different direction is to go digital (eg with something like the Ground Sound, the new DCN28 if it is available) and also do some room corrections (probably only on the low channels).
I didn't realize anyone had posted to this thread. I thought because I started the thread I was automatically subscribed. Clearly not.
The Marchand XM44 (active) is not transparent enough in my system because I can hear slight degradation in the handling of sibilants and a bit of treble grain compared to the running the speakers full-range. I can hear the XM46SB (passive) too, but it's much less noticeable. I've ordered a single Auricap to replace the two caps in the XM46SB and I expect that will help. One thing for sure, the XM46SB allows for a more relaxed sound with the BESL MTM's.
I believe that Pace Rhythm and Dynamics was the term originally coined and defined by Martin Colloms, who is as credible and honourable as anyone in the audio field. Why is describing a musical quality not acceptable? What is the intended use of your system, playing impulses or music?
I fully understand what Colloms means in his description. I have heard these effects many, many times. There's nothing wrong with using commonly accepted audiophile terms if they are mutually understood. Audio is not pure engineering, it is also art. Are you doubtful that some components or systems play rhythm better than others? Or that some make the sound subjectively appear to drag or race? When I bring home a preamp to try, and it makes music a chore to listen to instead of grabbing my attention, I only care in a theoretical way whether it's due to poor attack and decay. It doesn't replay music the way I want to hear it, that's what I care about. I don't know of any distortion, FR or impulse test in a competently designed amp that will tell how it will play music. It's either there or it isn't.
http://www.stereophile.com/reference/23/
audio-kraut said:
The Marchand XM44 (active) is not transparent enough in my system because I can hear slight degradation in the handling of sibilants and a bit of treble grain compared to the running the speakers full-range. I can hear the XM46SB (passive) too, but it's much less noticeable. I've ordered a single Auricap to replace the two caps in the XM46SB and I expect that will help. One thing for sure, the XM46SB allows for a more relaxed sound with the BESL MTM's.
audio-kraut said:
I believe that Pace Rhythm and Dynamics was the term originally coined and defined by Martin Colloms, who is as credible and honourable as anyone in the audio field. Why is describing a musical quality not acceptable? What is the intended use of your system, playing impulses or music?
I fully understand what Colloms means in his description. I have heard these effects many, many times. There's nothing wrong with using commonly accepted audiophile terms if they are mutually understood. Audio is not pure engineering, it is also art. Are you doubtful that some components or systems play rhythm better than others? Or that some make the sound subjectively appear to drag or race? When I bring home a preamp to try, and it makes music a chore to listen to instead of grabbing my attention, I only care in a theoretical way whether it's due to poor attack and decay. It doesn't replay music the way I want to hear it, that's what I care about. I don't know of any distortion, FR or impulse test in a competently designed amp that will tell how it will play music. It's either there or it isn't.
http://www.stereophile.com/reference/23/
Conrad Hoffman said:
That was one reason I kept the active XM44 for the low pass. What do you think about running the preamp signal through a pair of inductors with values of several farads (yes, farad, not microfarad). Hard to believe that would be transparent.
I've always liked the concept behind the XM46. After my experience with the veil added by the XM9 and a few other XOs, i am very leary of putting active circutry in front of the HF amplifier.
It would be nice to find a source for the inductors being used in the XM46, so that one could diy the Passive CL XO if one couldn't get away with just a 1st order PLLXO.
dave
PS: starting a thread does automatically subscribe you but occassional database glitches will randomly unsubscribe you.
dave
It would be nice to find a source for the inductors being used in the XM46, so that one could diy the Passive CL XO if one couldn't get away with just a 1st order PLLXO.
dave
PS: starting a thread does automatically subscribe you but occassional database glitches will randomly unsubscribe you.
dave
Duck-Twacy said:
That's exactly what I'm doing with the XM46SB. My sealed BESL speakers are -3dB @ 70 Hz according to the FR graph that came with them. The PLLXO is set for second-order, -3dB @ 80 Hz (recommened on the BESL site). Which gives a nice LR4 acoustic rolloff. You don't need a full XM46 for that setup. The 46SB provides a second-order high-pass for both left and right channels. You just need level control on the sub(s), which you already have with the XM44.
The only way to change the crossover frequency on either the full XM46 or the XM46SB is with component changes. Some soldering required, but it's pretty straightforward. You can order it for whatever frequency you want, so you wouldn't need to change unless you change speakers.
I agree, the XM44 is fully up to handling the bass duties. I strongly suspect it would outperform a high-pass 4th-order PLLXO, especially in my system with the long interconnects.
Duck-Twacy said:
Yeah, digital is probably the best way to go with the bass channels because of its flexibility, but I already had the XM44. My subs have a single-point EQ, which is nice for controlling the room mode, but you could also use a notch filter in the Marchand. I don't have much interest in a digital XO for the high pass. I'm skeptical that digital could be as transparent as the passive line-level filter, and my turntable would never forgive me.
planet10 said:
The inductor in my XM46SB is 5.6 FARADS! I read the value about three times, then did the calculation to make sure it wasn't a misprint. 😕
I'm sure I could get away with first-order on the high-pass, but second-order has to be better.
audiobomber said:
Yes they have to be BIG
Maybe, maybe not... something about LR 4th order XOs i find annoying.
dave
That's an interesting idea. I changed the op amps in an XM1 one time and it was just a peel-n-stick operation. What do you propose would be an upgrade to the OPA2134's that are in there now?Conrad Hoffman said:
audiobomber said:
AD823 and LM4562 drop in, both upgrades to my ear, bypass V+ V- with .01 ceramic caps across pins 4 and 8, replace 10mf electrolytic cap with a 10mf tantalum, maybe bias into class A, this is with XM1 boards and yep it can sound great, also the russian Teflon caps found on ebay to replace the little poly caps can make a difference - but they're BIG
It seems the XM44 can be modified to complement the world's most expensive loudspeaker (Magico Ultimate, $229,000).
The active crossover is a Marchand XM44 solidstate
unit extensively modified by Mark Eckert with
Black Gate and REL caps, Caddock resistors, Bybee filters,
and modified Welborne Labs outboard power
supplies.
http://www.magico.net/MAGICO_Ultimate.pdf
I had superb results by adding Welborne Labs power supplies to my IsoMagic DAC. Maybe the traditional cap and power supply upgrade on the XM44 is the way to go.
The active crossover is a Marchand XM44 solidstate
unit extensively modified by Mark Eckert with
Black Gate and REL caps, Caddock resistors, Bybee filters,
and modified Welborne Labs outboard power
supplies.
http://www.magico.net/MAGICO_Ultimate.pdf
I had superb results by adding Welborne Labs power supplies to my IsoMagic DAC. Maybe the traditional cap and power supply upgrade on the XM44 is the way to go.
I have played with active (analogue) and passive XOs for a little while. IMHO, less components, either opamps or capacitors, etc, are better, as long as the correct filter response is obtained. Adding too many opamps or capacitors veils the sound. Less is more. Simpler is better.
I love the concept of PLLXO but I have not actually done it. I guess it is important to know the output impedance of the preamp precisely in order to make a PLLXO work. I was thinking about using the PLLXO for the HP section of my 3 way speakers, but my tube preamp could have an output impedance of anything from 2k to 4k and I have not tried to measure it. I used LTSpice to model the PLLXO - a little bit variation on the output impedance of the preamp will cause a variation on the filter response, and it is not small! So if I replace a preamp then the PLLXO has to be redesigned! Still not an issue if I know exactly the output impedance of the preamp, presumed it is flat all way from 20Hz to 20kHz. I imagined that I could use two output caps on the preamp (one for LP and one for HP), and one of them is used as part of the PLLXO for the LP so that from the preamp onward there are no more than 2 capacitors in the signal path from 130Hz!
So I design active XOs using the least possible number of opamps. For my current project, for a nearly perfect B3 or LR4 network, with dipole and BSD compensation, taking the driver's natural response into account, I was able to derive such XOs using 2 opamps including any required "buffer" opamp for the HP from 130Hz up. The result is verified by LTSpice. I guess it should work well.
Just my 2 cents.
Regards,
Bill
I love the concept of PLLXO but I have not actually done it. I guess it is important to know the output impedance of the preamp precisely in order to make a PLLXO work. I was thinking about using the PLLXO for the HP section of my 3 way speakers, but my tube preamp could have an output impedance of anything from 2k to 4k and I have not tried to measure it. I used LTSpice to model the PLLXO - a little bit variation on the output impedance of the preamp will cause a variation on the filter response, and it is not small! So if I replace a preamp then the PLLXO has to be redesigned! Still not an issue if I know exactly the output impedance of the preamp, presumed it is flat all way from 20Hz to 20kHz. I imagined that I could use two output caps on the preamp (one for LP and one for HP), and one of them is used as part of the PLLXO for the LP so that from the preamp onward there are no more than 2 capacitors in the signal path from 130Hz!
So I design active XOs using the least possible number of opamps. For my current project, for a nearly perfect B3 or LR4 network, with dipole and BSD compensation, taking the driver's natural response into account, I was able to derive such XOs using 2 opamps including any required "buffer" opamp for the HP from 130Hz up. The result is verified by LTSpice. I guess it should work well.
Just my 2 cents.
Regards,
Bill
HiFiNutNut said:
It's hard to imagine anything simpler than the XM46. There's only a small capacitance in series (0.7uf), with the 5.6 Henry inductor and a 2.55K resistor in parallel.
The preamp output impedance will affect the efficiency and current of the circuit. but it's the power amp input impedance that will affect the crossover frequency. Marchand used a clever trick in the XM46. He inserted a 2.55K resistor in parallel with the load, which makes the XO frequency less reactive to the amp's input impedance.HiFiNutNut said:
I meant 5.6 henries.