Bi-amping Maggies

[DonH50]

As some may have noticed I am looking at crossovers to roll off the low end to my Maggies where my subs take over. Now, I am thinking of bi-amping again, but my old crossover system is dead and inexpensive (and most expensive) crossovers these days all seem to be Linkwitz-Riley 4th-order and lack the ability to provide independent asymmetric upper and lower slopes as specified by Magnepan (for mine, 300 Hz, 3rd-order LPF and 800 Hz, 1st-order HPF).

Has anybody used a "standard" crossover (e.g. two-way at a single frequency and equal hi/low slopes) on their Maggies and if so how did it sound?

Curious, thanks - Don

 

[Gregadd]

Don-I was just thinking you were the perfect person to answer that question.

 

[DonH50]

Yeah, well, I may be! Not perfect, just answering the question... Mine spent most of their previous life bi-amped, using a custom 3rd-order active LPF built into my sub's servo circuit for the bass panels (the box also included the sub's filters) plus a simple passive 1st-order circuit (capacitor) that I built into the preamp (ARC SP3a1) for the HPF. I'll probably call Magnepan next week and see what they say. I am wondering about the relative acoustic output of the bass and midrange panels since their crossover points are more than an octave apart. They must match outside the crossover area, however. I may pick up an inexpensive active crossover (dbx324s) and see how things sound, with a second Emotiva XPA-2 to drive the second speaker.

Still pondering the miniDSP, but for starters am leaning toward analog, and passive can be a pain especially when I want to play around with crossover frequencies to see what measures and sounds best first. In the primordial past I used my ARC tube crossover that I had tweaked to dial in the frequencies, then built my custom rig and later sold the ARC (bad, bad move!) The dbx unit may not be the best, but it is fairly cheap, has reasonable specs, and does what I want. Most of the other cheapie crossovers I saw did not go low enough for the sub crossover, and I am not ready to drop a grand on the xover just yet.

I am also thinking of modding the Emo's with more power supply capacitance... I need to stay away from Tony, he's a bad influence, making me remember all the fun I had in my DIY days!

 

[mep]

Don-I have a DBX 223 that I bought brand new that I will give you if you want to play with it. I bought it to crossover a pair of subs I used to own (transmission line subs that I designed and built). I still have the original box for it to.

Mark

 

[DonH50]

Hmmm, that's tempting, Mark! Standard or XL (I do not need XLR)? It would be an interesting trial to see if getting the three-way 234 is worth it. It would work for my subs now, and I could use it to piddle with the Maggie's crossover as well just to see how it works.

I actually found my old sub, and the Hafler amp I had on it, but the control box is missing. Probably in the box with my missing manuals, including the one I owe you!

 

[mep]

Don-It has the XLR inputs and outputs on 1/4" jacks. I just bought a bunch of 1/4" to RCA adapaters and used it that way. It will get your subs up and running. It just doesn't have the greatest flexibility. You can set crossover points, levels, and invert phase if you need to. The best part is that it's free to you if you want it.

Mark

 

[DonH50]

Well, shoot, the price is right, ship it to me! Let me know the cost and I will drop you a check for the shipping. My plan is to use it as a HPF for the amps driving my Magnepans, getting the very LF stuff out of the panels (they are full-range now, I am using the subs' built-in LPFs but not touching the Maggies). They are a little happier that way, at least measurably. I am still not sure the difference I heard was real or wishful thinking, but I figure what the heck. I already have 1/4" TRS to RCA, just need to pick up XLR to RCA. I may even have some of those in my bag, though I got rid of most of my patch cables when I (pretty much) quit doing live sound.

As a two-way, that also prevents me from running out and buying another amp to bi-amp the Maggies immediately; a good thing the way college costs have shot up! This way I can try out the unit and see how it sounds to me before deciding the next step.

Thanks! - Don

 

[garylkoh]

Don,

I guess that my question to anyone who wants to use an external crossover for a loudspeaker that is not designed for an external crossover is "Why???" As you've already noted, the crossover points on the bass and midrange is more than an octave apart. Actually this is incorrect as the "crossover" point is where the bass crossovers to the midrange. I am guessing that what you are saying is that the -3dB point on the filters is an octave apart.

Unless the acoustic output of the drivers that you are integrating are absolutely flat across the passband, an active crossover that is absolutely flat plus a real-world driver would not result in a flat output frequency response. Then you have edge effects, phase effects because the drivers are not coincident, etc. etc.

Just a week ago, I had the dubious pleasure of hearing one of the most absolutely horrible loudspeakers I have ever heard in my life - a Genesis 1 because the owner tri-amped it with external crossovers, added supertweeters, and removing the servo-feedback on the woofers. Just because the crossover is "perfect" and the amplifiers is "perfect" (over $1,000,000 in electronics) does not mean that the resulting sonic output would be perfect. A loudspeaker is a transducer after all.

Your Maggies would be the same. Even if the electronic crossover summed up flat, or you have gain controls on each passband, it's a LOT of work to get it sounding right.

 

[fas42]

Just a week ago, I had the dubious pleasure of hearing one of the most absolutely horrible loudspeakers I have ever heard in my life - a Genesis 1 because the owner tri-amped it with external crossovers, added supertweeters, and removing the servo-feedback on the woofers. Just because the crossover is "perfect" and the amplifiers is "perfect" (over $1,000,000 in electronics) does not mean that the resulting sonic output would be perfect. A loudspeaker is a transducer after all.

This is a nice echo of the discussion in "How much is too much?": overly complicated and expensive system failing miserably by all accounts. But why did it sound so bad? Was it audibly distorting, or was the speaker designer in you just too aware that the drivers weren't presenting a coherent, integrated picture?

Frank

 

[DonH50]

Don,

I guess that my question to anyone who wants to use an external crossover for a loudspeaker that is not designed for an external crossover is "Why???" As you've already noted, the crossover points on the bass and midrange is more than an octave apart. Actually this is incorrect as the "crossover" point is where the bass crossovers to the midrange. I am guessing that what you are saying is that the -3dB point on the filters is an octave apart.

Unless the acoustic output of the drivers that you are integrating are absolutely flat across the passband, an active crossover that is absolutely flat plus a real-world driver would not result in a flat output frequency response. Then you have edge effects, phase effects because the drivers are not coincident, etc. etc.

Just a week ago, I had the dubious pleasure of hearing one of the most absolutely horrible loudspeakers I have ever heard in my life - a Genesis 1 because the owner tri-amped it with external crossovers, added supertweeters, and removing the servo-feedback on the woofers. Just because the crossover is "perfect" and the amplifiers is "perfect" (over $1,000,000 in electronics) does not mean that the resulting sonic output would be perfect. A loudspeaker is a transducer after all.

Your Maggies would be the same. Even if the electronic crossover summed up flat, or you have gain controls on each passband, it's a LOT of work to get it sounding right.

Hmmm... Gary, you might not have followed my ruminations about my sordid past. My Magnepans are designed to be bi-amped; they have an external crossover to make it easy to replace with active, and in fact I have seen them bi-amped by Magnepan and ARC (who used to have a close relationship with Magnepan, to the point of reselling them at one time) at various venues over the years. They did it, why can't I?

In the past, I bi-amped mine with very good results. Note the drivers are coincident as they are flat panels (planar speakers, dynamic instead of electrostatic, but the idea is the same). The panels are fairly forgiving (because they are both fairly wideband and have a nearly constant, resistive impedance) so it was not too hard to dial it in (bearing in mind I do have measurement capability; it was not by ear). Mine is a special case, yes, and normally I agree with you. I have heard endless systems bi-amped (actively, don't get me started on "passive" bi-amping) that have sounded great, horrible, and everything in-between. A lot of professional sound reinforcement systems are bi-amped, very few consumer systems. I have done both in the past. Yes, care (and equipment) is needed to do it right, but in my case I am confident I can do so. And perhaps deluded...

I think I got the crossover (ok, -3 dB points of the filters, you are right) frequencies wrong after doing a little more research (had the wrong model, or perhaps a mish-mash of models, in mind) -- they are not quite that far apart. However, Maggies in general do have asymmetric transfer functions, with a rapid roll-off on the high end of the bass panel, and slower on the lower end of the midrange. This is to keep the upper bass and lower midrange "fast" (per Magnepan, when I asked ages ago) as the bass panel's heavier wires and thicker panel did not respond as quickly.

To answer your last question first, the "why" back then was because many folk, myself included, used a fairly low-wattage tube amp that would not play terribly loud (though loud enough) and more importantly provided relatively poor control of the bass panel due to low damping factor (high output impedance). I (among others) choose to keep my beloved tubes (ARC D79) and searched long to find a compatible but bass-friendly bottom amp (Counterpoint SA-220). The combination, with an active crossover dialed in, was fantastic (in my somewhat biased opinion). The impulse response, distortion, and frequency response were all significantly better by my measurements, and the sound improvement was also quite noticeable (blind or not, and yes I did both back then, ca. late 80's). The top end was little changed sonically; measurements showed less distortion with the D79 relieved of bass but in the real world I am not sure I actually heard a difference. However, the bass was much better, audibly tighter and deeper with that big Counterpoint driving the bass panels.

Today, with my D79 in new hands, the SA-220 in its box and unlikely to come out, and a new SS amp in the system, I actually have little need to bi-amp. It is more out of curiosity, and perhaps a childlike longing to set up my system of yore. I do not expect anything significant this time around; I am curious how much improvement in bass can be had by eliminating the passive crossover, but it is just that, curiosity -- I do not expect audible improvements. I would be happy to be proven wrong, of course. However, bi-amping is not my first goal...

One point not clear in all this is that my quest began, and may well end, with the desire to roll off the bottom end of my Maggies. Like any speaker, they distort more when driven beneath their low-end cutoff, and my present system uses them full-range with a pair of subwoofers in parallel. (To forestall the next objection, I have explained the technical "why" elsewhere, and I like it that way.) My current goal is to provide a HPF around 50 Hz or so to the amp driving my Maggies. A passive would be the easiest way, but is less flexible and would take some time to tweak in since it depends upon the load. An active would be easier to fit in, offers the flexibility to play, and the promise of future bi-amping if I decide to try that again.

Does that help? Sorry for all the ink, but I'm too lazy to edit it down! - Don

 

[garylkoh]

This is a nice echo of the discussion in "How much is too much?": overly complicated and expensive system failing miserably by all accounts. But why did it sound so bad? Was it audibly distorting, or was the speaker designer in you just too aware that the drivers weren't presenting a coherent, integrated picture?

Frank

It was much more than a lack of coherence or integration, it was an ugly, jarring sound.

 

[garylkoh]

Hi Don, Sorry - I was away from the forum for too long - first finishing off a series of new loudspeakers, then a 3-week vacation. Now I recall that in another thread you were looking for a HPF for the main amp to the Maggies. Yes, that makes sense if you are looking to relieve the amp (and more importantly the panel) of the low frequency load for better high-frequency clarity.

I've got a pair of 3.6's and they do sound better biamped using the dual inputs on the stock MG crossover even when I have a crossover. I haven't tried a HPF on the low-frequency input, but my latest experiments having a pair of 1440W prototype amps into each input certainly belies the "truth" that Maggies can't play softly with dynamics. If I do find the time to build one and tweak it, I can certainly send it to you after I'm done with it.

 

[DonH50]

No worries! And yes, I am wanting to relieve the panel, much more so than the amp. Once you start down the crossover path, it quickly turns into "anything worth doing is worth overdoing" and so I started thinking about full bi-amping again. And so it goes...

I have a theory that, despite their avowed "ease" of driving, in fact the panels need very good amp control (high broadband damping factor) to control secondary "ripples" in the panels during operation. That would explain why they sound better (cleaner, anyway) with higher-power amps even when the power is not needed. Of course, since they are like 82 dB/W/m, a little extra power doesn't hurt, either...

Hey, if you build something you like and can part with, I'd love to play with it! - Don

 

[mep]

Well, shoot, the price is right, ship it to me! Let me know the cost and I will drop you a check for the shipping. My plan is to use it as a HPF for the amps driving my Magnepans, getting the very LF stuff out of the panels (they are full-range now, I am using the subs' built-in LPFs but not touching the Maggies). They are a little happier that way, at least measurably. I am still not sure the difference I heard was real or wishful thinking, but I figure what the heck. I already have 1/4" TRS to RCA, just need to pick up XLR to RCA. I may even have some of those in my bag, though I got rid of most of my patch cables when I (pretty much) quit doing live sound.

As a two-way, that also prevents me from running out and buying another amp to bi-amp the Maggies immediately; a good thing the way college costs have shot up! This way I can try out the unit and see how it sounds to me before deciding the next step.

Thanks! - Don

Don-PM with your address and I will get it shipped out to you.

Mark

 

[DonH50]

So I did the math... The stock MG-IIIa external crossover is LPF = 300 Hz, 18 dB/oct (3rd-order) and HPF = 500 Hz, 12 dB/oct (2nd-order). The asymmetry imparts about a +1 dB hump around 220 Hz and about -1 dB dip around 450 Hz. This is for the electrical transfer function, of course, so there is no guarantee the acoustic output exactly matches, but we'll run with it for now.

The filters are 6 dB down around 380 Hz, so a 4th-order Linkwitz-Riley crossover centered at 400 Hz is pretty close to providing the same effective input power band and results in no ripple across the band. I might post pictures later but have not exported them out of my analysis program. So, if I want to bi-amp, I need a 400 Hz 4th-order L-R crossover to do the trick.

Plus the subwoofer filter, of course, which I was planning to be around 50 Hz. That would be -6 dB at 50 Hz with a 4th-order L-R, and about 10 dB down at 40 Hz, so I might move it down a little (the panels are rated -3 dB at 35 Hz but I would like them to not work so hard down there). Moving the sub crossover to 45 Hz puts the panels down 6 dB at 45 Hz and 10 dB around 38 Hz, probably close enough. The advantage of the higher-order filters are that I can get less overlap and move the crossover down a little to help avoid the usual integration issues between planar panels and the dynamic sub.

Onwards - Don

 

[DonH50]

Update: I installed the dbx crossover Mark graciously provided last weekend. Unfortunately, not a lot of listening, story of my life. The crossover is a stereo two-way unit, and I set it at 50 Hz to cross from my main MG-IIIa's (driven by an Emotiva XPA-2) to my subs (a stereo pair of Rythmik F12's co-located with the Maggies). The goal was to off-load the very lowest frequencies to the sub and away from the panels, reducing their excursion and thus improving their linearity and increasing their effective dynamic range.

After hooking everything up (a much bigger chore than I anticipated, for various reasons), I did not hear a lot of difference. This was not unexpected; I listened mainly to music which has less very deep bass than some movies. I did not hear any added noise or distortion, a very good thing. I can convince myself I heard improvment on some of the loud drum licks but without a DBT who knows. In the past I was able to measure lower distortion using a similar setup with these same Maggies, but even back then it was often difficult to hear unless I knew exactly what and when to listen. I need to get some good piano and organ music on CD… I also plan to try a movie or two this weekend; that is more likely to demonstrate audible improvements.

Other things in Life have put bi-amping them, the next step, on the back burner for now. Just found out the house needs a new roof, so the funds for a new amp are going towards a new roof instead… I may play around with the crossover, though in the past found 50 Hz was a good compromise for me so am in no hurry. I had to mount the crossover behind my center speaker, making it less accessible than I would have hoped. OTOH, it’s not something that gets tweaked after the initial setup.

Thanks again Mark! - Don

p.s. Less anyone misunderstand my excitement, with 24 dB/oct filters the input into the Maggies is down 6 dB at 50 Hz and ~24 dB down at 25 Hz. This is a major reduction in LF energy and should really improve the low-end performance of the system. A lack of listening coupled with our (humans') relatively low sensitivity to LF distortion makes it less audible than higher-frequency effects, but the benefits are real and I have, in the past, measured substantial reduction in distortion.

 

[mep]

You are welome Don. I'm glad it's working in your system and not sitting in my workshop room gathering dust.

Mark

 

[tdh888]

As some may have noticed I am looking at crossovers to roll off the low end to my Maggies where my subs take over. Now, I am thinking of bi-amping again, but my old crossover system is dead and inexpensive (and most expensive) crossovers these days all seem to be Linkwitz-Riley 4th-order and lack the ability to provide independent asymmetric upper and lower slopes as specified by Magnepan (for mine, 300 Hz, 3rd-order LPF and 800 Hz, 1st-order HPF).

Has anybody used a "standard" crossover (e.g. two-way at a single frequency and equal hi/low slopes) on their Maggies and if so how did it sound?

Curious, thanks -

Sorry to revive this old thread. I'm currently using my MG3A in biamp mode using an active x over and identical amps ( ARC EC21,a pair of ARC D79B"s for amplification) Cutoff of the EC21 was set in Magnepan"s recommendation 300/800hz 18/6dB bass slope. Capacitor kit for the EC21 is available through ARC.I'm very happy w/ the sound and I prefer it than using either my D250MK2 servo or my Classic 150 monoblocks driving the MG3a full range using the stock outboard x over provided by Magnepan.

Tdh888

 

[yawg]

I've been bi-amping my 3.6 Maggies with an excellent custom-built conventional passive XO since 5 years, very satisfied. See my sig for the amps I use.

I will add a photo.


The Musical Fidelity MA-50X monos you see on the photo were too weak for my max. volume and shredded one of the ribbons. Both the ribbon and the fuse blew. Now I use big 200W EAR tubed monos - those babies were made to drive the cutting lathes of recording studios.