Active crossovers

There's an abundance of info and opinions on just about every topic under the sun in audio, down to the smallest minutiae.

I have problem finding opinions on active crossovers and which sound good, or not. Or which offer good bang for the bucks. Or whether tubed crossovers are better than SS.

So what's the skinny on this?

Oh, and I'd like to ad a Stupid Crossover Question of my own:
The XO will match the levels of the low/high amps, but each amp will have a different sensitivity. Would you have to match the amp levels each time you change the volume?

There are serveral good active cross overs. Bryston 10B and the Marchand are the two that both reasonably priced and sound very good. They are both solid state. I bought the Bryston--at the time Marchand did not make a balanced cross-over. I'm pretty sure they do now. As to tube cross-overs, I can't really comment. I know of a few, but have not heard them. As to your "Stupid Question", it's really not a stupid question. If your speakers are designed to be actively crossed over then their response and impedance curves should be reasonable enough that it will not cause a significant change at various levels. However, if the two amps are not reasonably well matched in terms of sensitivity and possibly more importantly their current delivery (depending on the speaker), then you could wind up with a problem. I have not experienced this problem, but my differential between high and low end is only 1 dB. The Bryston only allows a +/- 5 dB, and this may be the reason why--if it were more you could run into some non-linear problems as you suggested.
To add to the post above, most amps are pretty linear in what is called their "gain curve". This means that what you put in is amplified at the same level regardless of how much drive is applied. To put things into generic numbers to make things easy to follow, putting .1 volts into the amp would give you 10 volts out. Putting 1 volt into the amp would give you 10 volts out. Putting 10 volts in would give you 100 volts out. In other words, it SHOULD remain a consistent LINEAR ratio.

If that were not the case, momentary peaks would be REAL loud and quiet passages would be REALLY quiet if you adjusted your listening level by the average output. Obviously, linear amplification can take place so long as the amplifier is not driven beyond its' capacity and there is enough signal to begin with to drive the amp. I have seen situations where there was enough signal to drive the amp, but not enough to drive it to full output. This tends to sound "flat" due to the reduction in dynamic range and "squashed peaks" that occur in such a situation.

With that in mind, setting the drive levels for each amplifier at the crossover SHOULD produce relatively even output through-out the various frequency ranges. This should take place regardless of volume adjustments at the preamp. Obviously, something to check before buying a crossover would be the amount of adjustability in output from band to band that the crossover would allow. It is possible to use amps with varying levels of gain so long as one could gain match them to begin with. The fact that the individual amps may be being driven harder or easier than one another to achieve initial equal output should not come into play so long as they are TRULY "linear" in amplification and have enough headroom to linearly amplify the signal presented to them.

If the amps were NOT linear in amplification, another factor that would need to be factored into the equation is that the impedance that each amp sees at the output would also vary. As such, power potential for each individual amp would change too. It is quite possible that one could be using two identical amps to actively bi-amp with and still have two different wattage potentials. This is due to the differences in impedances that the amp sees in each frequency range.

For instance, a speaker using a single tweeter and dual mid-woofers ( like a D'Appolito or MTM array ) might present an 8 ohm load for the tweeter and a 4 ohm load for the mid-woofers. While some of this would have been addressed in the passive crossover components, one could not make adjustments for this at the amp or active crossover other than to initially make sure that the output levels were balanced. After that, it would simply be a matter of each amp having enough power potential to work within the volume demands that the end user dialed up.

Keep in mind that when multi-amping, it is possible to "clip", "overdrive" or "saturate" one amp and not the other being used in a different frequency range. This explains why one can TYPICALLY get away with using a smaller amp on tweeters than one would use with woofers. Different frequency ranges put different levels of power / current demand with low frequencies requiring the most "juice" to operate properly. While some e-stat's present a VERY TOUGH load at high freq's, there usually isn't enough musical content up that high to make it overtly demanding. That is, unless you expect high volume levels out of such a design. Sean
thank you for the excellent responses.

The speakers in question are von Schweikert VR4's; I've learned they have an impedance low of 3.8 ohms at 25 Hz and a high 0f 19 ohms at kHz. Crossover between mids/highs and woofers is at 150Hz. High/mids and woofers are physically separate (stacking design).

I would be using Rogue Magnum's for the high/mids, bass amp(s) not yet determined. The Rogue amps have an input resistance of 200k ohms.

KP, i would contact Albert and see what he thinks about active crossovers and see what he recommends in terms of necessary mods to the speakers. Most of the benefit of active multi-amping comes from going "direct drive" from amps to speaker drivers and bypassing the internal passive crossovers. Going active while still retaining the internal crossovers should offer some benefits but not all of going "direct". Sean

PS.. Even if going to "direct drive" and bypassing the crossover or "dividing" network, it is still quite feasible and recommended to leave impedance compensation ) Zobel ) networks or notch filters in place. Doing so would require removal of the actual crossover network while still retaining some of the parts in the circuitry. That is why i suggest contacting the manufacturer before attempting to do anything. Putting their advive and knowledge to work for you could save you a LOT of time, money and headaches.
Hello, Sean, I'm not sure I understand (would not be a first!).

I thought, in a case such as the VR's, when the lows/mid are physically seperate (the speakers must be bi-wired or jumped to work) there was no crossover (between mids/woofers) to worry about.

I know there's other crossovers inside the speakers, but I don't know how practical it'd be to by-pass those (between the mid/tweeter). Or is there another croosover or filter involved right after the binding posts (ha, I hadn't thought of that!, maybe I'm begining to undestand?)

Search your topic "active crossovers" on and you'll get gobs and gobs of info. "biamping" is another search that'll pull up alot of related info too.
KP: There is a filter prior to the driver. It is effectively the cross-over. Essentially a high pass for the highs and a low pass for the lows. As Sean points out, you should bi-pass these. However, there is a word of warning here. While the cross-over is reducing the benefit of active bi-amping, these filter networks can be part of the voicing of the speaker. Eliminating them may change the characteristic of the speaker--sometimes for the better, but depending on it's design it can be for the worse. This is an area where I would contact the speaker manufacturer and ask them how (and even if) they recommend active bi-amping. For some speakers active bi-amping just won't work very well, but I'm not familiar enough with yours to know whether or not that is the case.
Have you thought about converting them to the vr5's?

Might be a more rewarding project. Again, I'd talk to Albert or VSA about it.
Hello, Rives, I think I understand now. I have sent Albert an email asking about how to best bi-amp. From a previous email, he strongly suggested biamping but did not get into technical details.

Drrasta, yes, I have though of the Mod 5 conversion. I still would like to bi-amp them, in either form. I've checked with Albert and the crossover point between the bass and mids would not change, so the biamping would work fine before and after conversion.
Have you heard them before/after the conversion?

KP, it looks as if you were thinking about passive bi-amping. While there can be gains made by doing so, they do not in any way, shape or form compare with the benefits of active bi-amping. That is, if everything is set up right and the speakers do not have a ton of "passive parts" in them to begin with. Sean
I though having an active crossover would handle all the problem, since it would sned the "highs" and "lows" to the resprective amps, and them have one amp on the high/mid module and another on the bass.

What I'm reading is, then, that even after splitting highs and lows into the correct speaker input post, I still have to worry about the inside of the speaker?

Wish I had a dry erase board we could share, it'd be so much simpler! I'll try anyway, this is what I thought!

pre---xo---hi amp---self-contained speaker box (high/mid)
|---lo amp--- " " " (low)

So in this case the mids/lows are phisically seperate at the speaker level. This won't work?

Since the speakers already have their own integral crossover, you would be crossing over electrically BEFORE the amp and then again after the amp inside each speaker section. In effect, the passive crossover parts that divide the lows from the mids would just be "exra junk" in the way of the signal and redundant unless you took them out of the circuit path.

I and others are talking "theoretical best performance" here, so keep that in mind. I guess that we figured if you're going to go to that much trouble to buy the amps and crossover, you might as well get the best results possible. You could actively cross and leave the speakers alone internally. It would be a step up in amplifier efficiency and power handling for the speakers, but i don't think it would compare to the total benefits of "direct drive".

Keep in mind that even if you did actively bi-amp, you would still be relying on the passive crossover components between the mid and tweeter if you simply divided the speaker between highs and lows. As such, you REALLY do need to discuss this with Albert and see what he thinks is best. After all, he should know his product better than anyone else. At least one would hope so.. : ) Sean
Sean, I see what you are saying.

I've seen your other post. You, my friend, aree the Poster Chilp for multi-amping. I was hoping you'd jump on this thread!
So in my case, I'd need to at least tri-amp for max results (that beign the total removal of all passive crossovers). question mark.

I'm waiting for Albert's reply.

If at all possible, you would want to actively tri-amp. Take into account that this is a LOT of set-up to say the least with a pretty massive amount of cabling, amplifiers and rack space taken up. It can and will change the sound of your system, possibly to a very large extent. My experience is that doing something like this produces FAR more liquid and transparent sound IF you can get everything dialed in. This can be a VERY time consuming "if" and one must have both patience and some help to do this. If it sounds like i'm trying to talk you out of it, i almost am. You REALLY need to think about what is TOTALLY involved in doing something like this.

My one concern about your speakers is that they have an "extra" rear mounted tweater also. Do you know if it crosses over at the same frequency / slope as the front firing tweeter ? Sean

PS... I am the poster child for "something", but i don't quite know what it is : )
Ah! good point, I'd have to ask Albert.
Now, let me ask:

Bass is actually 2 woofers, crossover differently.

Alber says to get the most from his speakers you should bi-amp, tubes on top SS on the bottom.

So, if I were to say, biamp now, could I get real benfits and then if/when I tri-quad amp really go into the stratosphere with dynamic power? Hum, why doen't this sound farfetced to me anymore?? SET for the highs, ogues for the mids, 2 SS amps for the bass... ahhhh.... I hear the siren call... maybe this is "the direction" question mark.

And, yes, "Poster Child for Multi Amping", I'll stick to that! Oh, and certainly you'd be in the group photo for folk who make this hobby fun. But heck, that's just my opinion and I'm sure it's full of holes (also part of the fun, huh?)!

Taking a break from the music to keep up; tubes are nice and hot; breeze and cool outside!


Think first about what active crossovers are all about. There are a couple of benifits; 1) crossing over before amplification means the amps (and speaker cables) don't have to see the whole audio spectrum and can put their power into a more limited frequency range hopefully with lower distortion, 2) active crossovers can be provide alternate filtering as compared to the built-in crossovers, but you need to be careful here the speaker designer most likly spent a great deal of time blending the sound from the different drivers to give a smooth transition between frequencies and 3) active crossovers can be used in conjunction with equalizers or as just a two band equalizer its self to provide room correction.

I went to the von S web site and read some about your speakers. It looks like what you suggested about an active crossover to split the signal into a piece for the base module and one for the mid/treble module will work. It won't give you all of the benifits of active crossovers that can drive each set of drivers for each frequency range but you will still get the advantage of making your amps and cables work over a more limited frequency range. The signals will still pass through the internal crossovers but I think that is also good in your case as von Schweikert went to a lot of trouble in the design of their crossover and I would think that your speakers would not sound as good without their internal crossovers.

So with the goal of offloading the amps and speaker cables I think that your idea would work and I think that this goal would also put a limit on the quality of the active crossover that you would require.

For what its worth this is how I see it.

Good Luck,

If you decide to use and active crossover (as I did) and can afford it, go for the Krell KBX.
Krell makes the active crossover for such speakers as the $40,000 four way B&W Nautilus, and has made crossovers for other companies as Martin Logan.

Have fun!
Guys, you are great!

Couldn't find info on the Krell. I need SE, not balanced. What does the Krell provide?
Any other especific xo's to look for?
As per Rives, looked up Marchand, saw the XM9L for $400. How's this model?

The Bryston 10 seems to go for 1200 or so, the Krell for 1500 (used on these).

The Marchand is quite good. Some have indicated it's better sounding than the Bryston, but I have to admit I have never done a direct comparison. I think it's fair to say that $ for $ it's probably the best out there. If your budget allows for more, then it's difficult to tell which unit is best. As to the Krell, they did have some 60 Hz noise problems at one time. Perhaps that has been fixed, but if I'm not mistaken they quit making the KBX some time ago and getting the boards for the various cross-over points might be difficult. You should look into this carefully before going that route for sure.
Pardon my ignorance. I'm not even sure if this is the proper thread for my question, as it's not related to specific active cross-overs. Some speaker maufacurers compensate for time (distance), phase and even impedance via thier cross-over designs. I would guess that using mulitiple amps would negate the impedance issues but how about time and phase?
here's a curve ball:

Any thoughs on using the Innersound bass amp from the new Eros?
Roger told me he can change th xo point to anything I want.

It also includes a remote volume control; 600 wpc.

Anybody know how crossover-less Gallo Ref's would factopr into eqauation.They have no X-over persay on a supplied Cap that can be inserted between Bi-wire posts.
If you can buy the Krell KBX for $1500 ... BUY IT!
The Krell lists for about $4K. It works either balanced or SE. It costs only $200 for Krell to make you a custom board for your speakers. If you change speakers Krell can make you a new board, making it immune from being obsolete.
E-mail Krell and ask for specs. (
Tell Steve Leckrone that I suggested you contact him. He has been a great help for me.
Actively biamping my system is the single biggest improvement that I ever made.
Also, I was just at the Krell factory and they do still make the KBX, but since it is a low volume unit, they are not advertising any more. Getting boards should never be a problem.
Richard Diamond
I'm using a Marchand XM9-3 Deluxe to tri-amp one system. I'm pretty pleased with it over all. The biggest drawback to this design is the lack of variable crossover points. You basically have to install very inexpensive "resistor packs" to adjust frequency. While it is not really a big deal, it's nowhere near as convenient as having an adjustable pot on the faceplate. It is probably a LOT more precise though. Other than that, i've only used professional grade crossovers and they did not sound as clean or focused. Nor did they offer the added versatility of the "blend" or "Q" adjustment that the Marchand has.

Mike Bates ( aka "Magnetar" on AA ) has used a Marchand that he built as a kit along with a Behringer model. He said that the Behringer has some nice features and works better on the low frequency range whereas the Marchand is cleaner up top. He also said that the Behringer has some features that he's found to be quite useful that are not available on the Marchand. He's a very experienced "tinkerer" who's judgment i trust and respect.

There was a Pioneer crossover that was made that seems to be relatively sought after. I noticed that while i was looking for one, there were other "wanted" ads posted for it on both A-gon and Audioweb. Can't remember the model right now, but it offered you the choice of four different slopes ( 6, 12, 18, 24 db's per octave ) along with some other nice features.

My brother is currently using one of my older professional grade crossovers that i used to use when doing pro sound reinforcement at concerts. It offers variable slopes ( adjustable from 1.5 to 12 db's per octave ) and variable crossover points for each band. This one comes in very handy as it makes it easy to find optimum crossover points when building speakers from scratch. Once he's got the cash, he's going to go into a Marchand. I hope it's soon, as i've got a large line array ( 54 drivers per cabinet ) that i need to start "fine tuning" in terms of crossover points. Sean
BTW just saw on ebay somebody paid $3500 for a set of vintage Marantz x-overs!!Seems a bit steep but....
Active crossovers are a tricky topic, imho.

I'd recommend using an active xover for subwoofers and a passive xover for the rest in general.The advantage here is that you can use a clean, high power amp for the sub bass and select an amp that is optimal for the main speakers (as compared to driving everything from one amp)

It is very easy to muck up some very high quality sound with a not so wonderful HP filter made from not so wonderful opamps and caps (not to mention the power supply).

On the other hand it is possible to implement surprisingly clean and good xovers with both tube and solid state parts. Regardless, it is a good practice to compare the speaker in question with and without an active xover in line.

Much depends upon you, your ears, and your system as to wheather or not an electronic xover will be an improvement or a deficit. Which means there is NO single answer.

In general, the commercial "pro sound" products do not have truly audiophile "sonic signatures" so I would be cautious about them in general.

The advantage of an electronic xover for a DIYer is certainly one of flexibility and ease of adjustment, so it has merit on that point. On the other hand you do need an extra amp or so to use it, so perhaps the money is better spent in other directions?

Ymmv with electronic crossovers in general.

As has been touched on above, the "purist" solution for the VR4 is to passively biamp. Putting an active xover in the signal path is going to introduce unnecessary processing. The top and bottom modules of that speaker already have high pass and low pass filters (respectively) that allow a full-range amp to drive each. The only "compromise" that will be needed is some sort of signal splitter at the output of your preamp, unless you are luck enough to have 2 outputs on the preamp. I would strongly advise against trying to bypass or modify the internal filters of the VR4, which have already been optimized for module integration.
Interesting, Ral, Albert sent me an email staing as much.
The trick then, is to find amps with similar sensitivity.
So now, to look for an SS amp to match 99 magnums?

There is an active crossover, that can be ordered in a 2-way or a three-way configuration,from the Australian company
Symfonia.It´s called the Symfonia AC 1, and the company
has a testimony from Duntech on their website,that Duntech
finds the AC 1 superior to the Krell KBX.
I find this crossover interesting, but actually, I have
sent a mail to Symfonia three times,without recieving an answer!
Regarding the general question on active crossovers,I think
that it depends on several factors,e.g. the need for high
sound pressure level requirements, synergy with other components,knowledge by the user etc, if the end result is benificial compared to a passive solution.
But I agree to the statement, that it´s generally a good thing, to have a active crossover for the lowest part.
I can honestly say that by using this approach in my current loudspeaker-project, I have,
with some tweeking, succeded to get possibly the best
bass reproduction that I have heard..
KP: Yes, matching the 2 amps will be challenging. If you can find out the overall gain of the Rogue that would be a good start. Most solid state amps will have gain in the 28-30 dB range. Might be interesting to see what Albert uses for his SS/tube combo. Even with gain and sensitivity matching, a tube amp can still sound "louder" than the SS. However, given the low xover point on the VR4 (150Hz) it shouldn't be terribly noticeable. By the way, I have been thinking about the same approach to driving the VR4, if I decide to buy them (I'm still auditioning some locally). Good luck and please let me know what you find.
Ral, I will.
I'm looking for ss amps with adjustable gain; should make life easier.

Matching two amps is trivial.

There is no matching to be done, other than level setting. Just about every electronic crossover has level adjustments for every band.

Regardless of the "gain" of an amp, when you increase the input by 3dB, the output goes up 3dB. So, all you have to do is to put a level into amp 2 that when compared at the speaker output using pink noise gives the *same* level on your SPL meter as does amp 1. Of course, your ear will be the final arbiter of the precisely perfect relative levels.

If your electronic xover does not have level controls, and your amp does not either, you can either build a fixed pad into a small box - after measuring the dB difference with no pad. Or you can buy an after market in-line adjustable attenuator box to put in line with the amp that is "too loud" to reduce its level. Alternately, you can drill some holes into your amp and put a small rotary switch with MF resistors and make your own attenuator built in,or just put a decent pot there... I do that all the time on Phase Linear 400s that are used as subwoofer amps from time to time (on the front panel of the PL400, to match the PL700).

All done.

Yes, having an amp with adjustable volume control gains, like the Parasounds (God bless em for doing so!!!), makes live way easier when you biamp!
Hey, also let me know which active crossover you are ending up with ultimately!!! I would like to know for my own references/uses! The only one I can recall having heard implemented is the Krell, and am currious to hear how the Bryston, or others, sounds!
If you use the Krell KBX, it has gain adjustment for each Highs and Lows, making the need for adjustable volume on the amp totally unnecessary. In fact, the less in the signal path the better, so unless I could bypass the amps volume control, I would prefer one without it.

OK, I posted this in the wrong thread, hehe! Hope I didn't cause too many blank stares!

I asked Albert the best way to bianp the VR's, here's his info:

Hello, again,
for facilitating active biamping of the speakers, does anything need to be
done to the speaker itself?

No. The internal circuitry divides the signal between the two amplifiers, so you don't need to change the speakers or buy an electronic crossover. Just connect the preamp or CD player to both amps with a "Y" connector and interconnects. Just make sure that both amps have a similar input sensitivity, so that they play at the same volume level. The output power has nothing to do with this, we use an 8 watt SET amp on the mid/tweeter unit and a 200 watt solid state amp on the woofers with complete integration! Albert
Let's be clear, these are LEVEL controls, NOT "gain" controls!!

We call them "gain" but 99% of the time they set the output level of a fixed gain circuit. To change the gain of a typical solid state circuit you need to change the feedback resistor. Few circuits (for good reasons) actually permit you to adjust the gain of a stage directly.

It is standard practice to build in "excess gain" so that you can set a level control midway - in fact the volume control only LOWERs the level down from maximum.

On power amps 99.9% of all level controls are right at the input of the amp, before any active circuitry. Again they only serve to lower the input level to the front end.

There are some exceptions, like the Crown Macro Techs which if I recall actually has a neat circuit that adjusts gain, not level, in order to *set the level* then sent on to the fixed gain remainder of the amp...

Again "gain" is not an accurate description of this control, although we all tend to use the term!