Best Crossovers?

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#1
So I am thinking about implementing a crossover (actually, a HPF) for my main amps. My system uses two subs set up in parallel with the main L/R speakers. The subs cross using their built-in LPF around 45 Hz or so and the mains run full-range. I was going to build a real simple first-order in-line HPF (a capacitor) but am thinking I need faster roll-off and simply haven't the time to piddle like I used to do. So, I am thinking of a simple HPF or maybe full-blown crossover (so I could bi-amp again) that is not a small fortune. And wishing I had not sold my old ARC crossover many years ago... I may still just roll my own, probably an LC second-order design, but active would eliminate the source/load impedance issues and if I got a pair or a stereo crossover I could entertain the notion of scratching the bi-amping itch.

What are good ones to look at, say in the $100 - $300 range, under $500, and perhaps $1k and up range? Passive is OK but I would like second-order so it needs good coils (assuming it is an LC design); active must have good sound (low noise and distortion) because it will be in the primary signal path.

Curious, thanks - Don
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#3
Hi Lee,

Yes, at least the on-line pictures, but they are out of my price range and somewhat overkill for my needs. Interesting units, though; maybe when I get the kids through college! Have you used them?

I have a bad feeling anything I like is going to be more than I was hoping to pay, but that may just force me to build a couple of passive filters and call it a day.

Thanks,
Don
 

RogerD

Active Member
May 23, 2010
3,155
0
36
BiggestLittleCity
#4

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#8
Thanks guys, lots to follow up...

I have used Behringer in the past and found them noisy; I have not tried the new DSP units. I am leaning heavily toward analog because I do not need high-order filters, the AVR does room compensation, and I do not really want to stick another A/D-D/A conversion in the signal path. The Rane I have also used and it was a little noisy but may be a reasonable solution. I need to look at some of the other options.

Kal suggested a miniDSP a while back; it would be a kind of a neat thing to play with, anybody using one, or at least heard in operation? Keep in mind this is for a HPF and will be in line with the primary speakers, not the subwoofer (the opposite of what a lot of folk likely do).

Several folk suggested the Bryston unit, knowing my tastes and admiration of Bryston, but this is definitely a time when finances must rule.

I will have to check out the NHT and Marchand units. My main concern is not mucking up the sound of my Maggies, so I am very interested in hearing how the units sound in the midrange and main audio bands, not just down in the nether (suub) region.

Again, thanks all, and I have some reading to do this weekend! - Don
 
May 30, 2010
14,101
72
48
Portugal
#9
The Marchand crossovers are very highly considered in DIY forums and you can always tweak them to the level you want. As far as I know , they are used in very modified versions by the owners of the Tape Project in their multi amplifier tubed systems using Bottlehead modified electronics.

BTW, the ARC crossover you are referring is the ARC EC22? I also owned one with a card customized for Maggies at 3500 Hz.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,416
1
36
Seattle, WA
www.genesisloudspeakers.com
#10
IMO anytime you add circuitry ahead of the power amp, you are going to affect the sound - input impedance and output impedance of the active crossover unit, plus another two pairs of connectors, and interconnects. Since you already run your subwoofers in parallel off your Maggies, why don't you consider just adding a HP to the Maggies, and keep your subwoofers in parallel off the output of your power amp.

Then, you are just adding a capacitor in series and an inductor in parallel (for 2nd order) to your Maggies. It might be the cheapest and best sounding solution especially if you are looking at a frequency of 45Hz.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#11
microstrip; Yup, also for my Maggies, imagine that... :)

Gary: That was my original plan, but I am waffling mainly due to a lack of time (yes, despite all the posting here!) I was also thinking active to provide ease of expansion if I want to bi-amp my Maggies again in the future. However, I must note that an active circuit buffers the input and output so is usually much less impedance-sensitive than a passive design, though it does add noise and (usually) more distortion. We think alike, however; when I last bi-amped my Maggies I used an active sub crossover (LPF that I built, along with a servo circuit, ages before they became commonplace in subs), and built a passive filter network into my old SP3a1 preamp for the Maggies' bass and mid/tweeter sections. I did have to tweak in the passive networks to work with the amps' input impedances and preamp's (fairly high) output impedance. The only extra active electronics were in the sub path, and though they were very low noise and distortion, they were also only in the sub's signal path. I am more nervous about putting something else in the main path, though arguably with an AVR and amp it doesn't really matter...

Before anyone asks, my old set-up was:
TT (forget the model but still have it; a very good Linn copy with heavier bearings)
Tuner (Yamaha T-1 or T-2, been a while)
Cassete deck (cheap'ish)
Several R2R units (Pioneer RT-707, Studer B77, ML-modified Studer 1/2" two-track, etc.)
ARC SP3a1 preamp (modified)
crossover -- was the ARC unit, later passive in the SP3a1 and I sold the ARC
ARC D79 (upper amp)
Counterpoint SA220 (bass amp)
Custom subwoofer using a 12" Infinity QLS driver in a ported enclosure, custom electronics (crossover and servo circuit), and bridged Hafler DH-220 (not a great amp, but with the servo did fine)
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#12
Well, I am leaning toward a Marchand passive unit (XM46) for now, configured as a dual HPF to roll off the bottom end going into the Maggies. The flexibility and buffering of an active unit is still very appealing, but it looks like I would need to spend ~$1k USD or so to get a unit sufficient for present and future needs, and the less expensive units do not have the features I want. The Rane units look interesting, but the HPF only goes down to 70 Hz and I'd like a little lower. The NHT unit is interesting but the NHT site claims it is no longer made. The DEQX is overkill, in performance and price, and I worry about the SQ of the miniDSP for the mains though am considering a stereo unit to compensate my subs (waiting to see how I do with what I have, especially since the room is so heavily damped at this point). The price is low enough I may get one to play with and see how it works out. The Behringer unit (or DSP-based Rane units) would again add another conversion in the signal path I was hoping to avoid, at least for now.

The Marchand passive is 4th-order Linkwitz-Riley which means I could leave the crossover down around 50 Hz and still get some rejection where the Maggies roll off (-3 dB around 35 Hz). I'd like to keep the crossover as high as possible, within reason, and around 50 Hz is where I have had it before and where the subs cut in now. I worry a second-order (passive or active) is marginal; my old active design was third-order. I could also cascade a couple of Butterworths on my own, but it would be nice to have it made into a nice little box. My DIY days are largely behind (and ahead!) of me for now due to a demanding work schedule and lots of other things going on with the kids, parents, etc.

For about the same price, though, I could get a 2x8 minDSP unit... Has anybody here tried one, what did you think? Kal suggested one to me a while back, but I am not sure he has heard one either.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#13
The Hafler amps, with their MOSFET outputs, tended to be bass-shy audiby and measurably, and did not have real good (high) damping factors. That made them a good match with the tube gear I had at the time, but less than ideal for driving dynamic loads with lots of charge kickback. The servo circuit took care of most of that in my design (a simple LDI feedback using the second voice coil, nothing fancy -- this was during my college days). There were some other minor issues, including a little hysteresis in the electrolytic input coupling cap (solved by using a better cap) and minor wiring issues (tweaked the gauge and dressage).

An interesting aside is that a friend had the same amp in his system (not bridged) and we heavily modified it with new caps, bus bars for power supply wiring, etc. All our worked helped a little bit... The thing that made BY FAR the largest difference (audible and measured) was to build a bank of power supply caps in an external box wired to the amp through very heavy cables. It turned the relatively anemic bass into a powerhouse. Bear in mind at the time the other amps we were comparing including a Phase 700 (mine, a love/hate affair) and various ML, Krell, and Threshold SS amps. We did not achieve the same level of performance, but awfully durn close, and for a fraction the price.

Back on-topic: The miniDSP has mixed reviews on AVS but a couple of other sites (which I did not bookmark; need to not stay up so late on Saturday nights staring at this little bugger!) were quite favorable as far as SQ goes (bit of a surprise, but I'll take it!) One of the drawbacks seems to be the low'ish output level, 0.9 Vrms for the basic version. The miniDSP forum has comments from the designer(s) saying it is due to the USB 5 V supply limit; reasonable, but I do not see why they don't provide a higher-output 9 - 12 V version -- it can't be that hard! I need about 1.2 V to drive my power amps to FS so would get the balanced version, with its strange connectors, or step up to a bigger unit (more costly, and not available in a box so more work). Annoying... I am contemplating getting one and customizing the output opamps' gains to provide a bit more output, understanding I will use a 12 V wall-wart to power it. Being able to tweak and play is just sooo enticing... :D

The Marchand units seem to be almost universally praised, but $300 for a few coils in a box is irksome. Worth it, given the price of everything else, but makes me think I should just build my own. In the spare time I do not have... I would also prefer a standard Butterworth for this application since the AVR sill compensate the phase match and I would like faster roll-off around the crossover than a L-R design. Plus the sub has a phase control knob as well, so in this case I don;t really need L-R for the HPF to the Maggies. OTOH, L-R is perfect for bi-amping the Maggies later, assuming I decide to buy another Marchand (or whatever) for that.

Hmmm.. Wonder if my wife would allow me to pick up another XPA-2 during their winter sale so I could bi-amp my mains? I don't really need it, but it would sound cool, and there're all those clothes she buys and doesn't use. I could point that out, but hospital bills would really crimp my audio budget... :D
 

RUR

WBF Founding Member
Apr 21, 2010
647
0
0
SoCal
#14
....One of the drawbacks seems to be the low'ish output level, 0.9 Vrms for the basic version. The miniDSP forum has comments from the designer(s) saying it is due to the USB 5 V supply limit; reasonable, but I do not see why they don't provide a higher-output 9 - 12 V version -- it can't be that hard! I need about 1.2 V to drive my power amps to FS so would get the balanced version, with its strange connectors, or step up to a bigger unit (more costly, and not available in a box so more work). Annoying... I am contemplating getting one and customizing the output opamps' gains to provide a bit more output, understanding I will use a 12 V wall-wart to power it. Being able to tweak and play is just sooo enticing... :D...
Don, in addition to the "Rev A" 0.9 Vrms version you describe, they also sell a 2.0 Vrms "Rev B" version, which is the weapon of choice for typical preamp output. I installed one of these, along with the PEQ 21 plug-in, into a friend's system a few weeks back. Inserted between an ARC Ref 3 (2V @ 0 dBfs) and monoblocks/sub to tweak for best mains/sub crossover frequency/slope, as well as to EQ 3-4 FR bumps in the mains.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#15
Yes, I looked at that, but the RevB version allows 2 V input; its output is still limited to 0.9 V (I missed that the first time I looked at it). You have to go to the 2x8 or 8x8 to get higher output. The usual solution seems to be to use the balanced version, which gets to 1.8 V input and output levels when run single-ended. You also have to rig connectors since the balanced unit uses pin connectors instead of standard RCAs or XLRs. It is frustratingly close to be plug-and-play on the HW side... Still very appealing for the price, and would be a cool toy to play with!

You notice no sugnal degradation in your mains? That is/was my main concern with such an inexpensive unit, but a lot of folk swear by it so I am thinking of giving it a try.

Thanks! - Don
 

RUR

WBF Founding Member
Apr 21, 2010
647
0
0
SoCal
#16
Well, I'll be jiggered! Full output for his Wolcott's is 1.5V rms, but I just called him and he's not missing any volume at his typical listening levels. If he decides it's problematic, we can pick up a Cleanbox Pro or move up to a balanced version mini DSP. Thankfully, they're cheap!

As for signal degradation, although I've heard his system many times, I'm not familiar enough to say anything definitive without a fast A/B, and that would require a switching device I don't have. Certainly, nothing adverse was noticed by three listeners when it was installed, and I suspect that degradation, if any, is dwarfed by the improvement in sub integration and flat FR - no emphasis, no masking.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#17
Does he have the A or B version? Though you give up signal and noise by attenuating the front end with the B version, I am thinking that is far better than clipping the ADCs' inputs. As for output, I suspect the only thing missing might be the very loudest peaks, and chances are nobody is missing those even if they aren't there (clipped). After all, from 1.5 down to 0.9 is only about 4.4 dB change in headroom. My Emos only need about 1.2 V so it's even a smaller change. It is debatable whether he (or I) would notice if the thing actually put out 2 Vrms... I only need the basic 2x4 unit for what I have in mind, now or in the future, so stepping up does not really seem worth the price or hassle but I am still cogitating. I am also wondering if anybody has tweaked the resistors to get more gain, but need to know if the supply is regulated on the board (otherwise applying 9 - 12 V would not help).

As to your second paragraph, I agree that as long as no objectionable noise/hum is added the improvement will far outweigh any downsides. Since my Pio does not EQ the sub, I am considering the bigger 2x8 unit so I can piddle with it. The downside is that in my room, and with its own phase adjust and single-band EQ, it's already fairly flat so there's not a lot of justification for the extra effort.

Pondering... Older boy goes back to college this week and I just got the bill, so I may be pondering a while! - Don
 

tony ky ma

Industry Expert
Aug 22, 2010
626
0
0
Whitby Ontario Canada
#18
LC passive xover in diy

Don
Bi amp all ways the most effective up grade, I did use tube and SS active xover before, but found the best sounding is LC passive type. now in my system's xover is 4 ways 600 ohm LC filter drive by a 300B in a 2.5 K to 600 ohm out-put transformer, another step up in-put transformer to cover the losing gain, maybe this way will out in your budget, or you can use a current op-amp to drive the filters for lower the budget. I know you can work out the value of the LC in a 600 ohm filter for a frequency that you looking for, the only difficult part is to find the right choke, custom order is the only way, I got some adjustable choke from junk shop, that is good for HPF, choke in sub side has to be order make in iron core for the high H value
cheers
tony ma
 

Attachments

DonH50

Member Sponsor & WBF Technical Expert
Jun 23, 2010
3,537
4
38
Monument, CO
#19
Oooh, I am jealous of those chokes, Tony. Trust you to have the good stuff at hand! Yeah, that is really what I should do, I am just lazy (I say busy, but I find time enough to type...) Of course, I would have to buy another amp... One of those things I do not really need but it would be nice to have. It would also get the amps out of the console and on platforms behind the speakers, using very short speaker cables.

Are you using second-order (12 dB/oct) LC or higher order? Ideally I would use 3rd or 4th order Butterworth for the low-end HPF, then probably 4th-order L-R for the bass/high panel crossovers for phase matching. I would leave the factory crossover from HF panel to ribbon tweeter in place (don't want a third amp channel). I can probably live with the gain loss from a simple 2nd-order, not sure about higher. I would buffer with op-amps, don't have all the stuff nor really want to deal with tubes anymore (can I mention again I wish I'd kept that old ARC crossover?) Maggies cross over low and slow so using a pair of XPA-2s would be a reasonable choice; the LF/HF amps must match decently to preserve the speaker's sound.
 

tony ky ma

Industry Expert
Aug 22, 2010
626
0
0
Whitby Ontario Canada
#20
Oooh, I am jealous of those chokes, Tony. Trust you to have the good stuff at hand! Yeah, that is really what I should do, I am just lazy (I say busy, but I find time enough to type...) Of course, I would have to buy another amp... One of those things I do not really need but it would be nice to have. It would also get the amps out of the console and on platforms behind the speakers, using very short speaker cables.

Are you using second-order (12 dB/oct) LC or higher order? Ideally I would use 3rd or 4th order Butterworth for the low-end HPF, then probably 4th-order L-R for the bass/high panel crossovers for phase matching. I would leave the factory crossover from HF panel to ribbon tweeter in place (don't want a third amp channel). I can probably live with the gain loss from a simple 2nd-order, not sure about higher. I would buffer with op-amps, don't have all the stuff nor really want to deal with tubes anymore (can I mention again I wish I'd kept that old ARC crossover?) Maggies cross over low and slow so using a pair of XPA-2s would be a reasonable choice; the LF/HF amps must match decently to preserve the speaker's sound.
Yes I am using second -order 12db/oct, I did try 18db/oct and 24db/oct with those tube and SS active xover, to me 12db/oct is the best for my system, I can't say same to other system different room and different system will effect different. the crossover point for the crossing channels should not be the same point, it should depend how sharp drop of the curve and effect of the room and speaker too, I found out the best xover points for each channel is by listening with a selectable xover first, and then make a fixed one. too many filters will hurt the sound quality too, so I only use add on style, one filter for a channel, roll off the other side by speaker itself that is why all crossover points had to be decided by listening. in phase match, I moved speakers front and back for matching also adjusted by listening ! this is the way what I did and hope it can give you some ideas
tony ma
 

About us

  • Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing