Active Crossover/Biamp for 604 duplex?

[lhgin]

Hi guys,

Has anyone tried biamping the 604s? Or tried active crossover for the 604s?

I've been thinking of building the MasteringLabs' crossover for my 604D and thinking of biamping the 604s with 2 SETs of 1W each.

Rgds,
lhgin

[David Deratany]

Been there, done that (biamp) with my 604B's. Will never go back. I think that ideally you would want to use an active crossover when biamping. There are units out there you can build that will minimize the cost, if that is a consideration.

[marcuswilson]

Biamping can be excellent, but can take a little time to get right. Nothing can beat a high quality electronic crossover. Many crossovers that are fine for PA do not do justice to the speakers in a domestic setting. If you are well off get a Bryston. I have seen some kits a friend of mine got through a website, they are very good (not cheap for a kit but cheaper than average quality new ones and much better). Because they are so simple, it's hard to beat an old JBL 5200 series if you upgrade the components, including fitting TL075 ICs. I have found making good passive crossovers superior to using standard 'commercial' grade active crossovers and bi-amping. I make 24 dB / octave passive crossovers with giant air cored inductors and oil *****d film capacitors and they transform speakers like 604s and Tannoys. Many old crossovers can be improved by the addition of an impedance compensation network across the woofer. This is a resistor, the resistance of which is equal to the DC resistance of the woofer voice coil, in series with a capacitor of 25 to 100 uF. An old 604 would need about 22 - 50 uF for 8 ohm driver, half that for 16 ohm and 1/3 of that for 24 ohm. Later speakers may need higher value capacitors. Do this only if it is not already fitted to the crossover.

[lhgin]

Thank you for your reply guys!

Hi Dderat,

What crossover did you use for biamping? The Mastering Lab's crossover? With over 100dB sensitivity, will 1W + 1W enough for biamping?

Hi Marcus,

I've been thinking of this crossover:
http://www.soundpractices.com/images/604xover.pdf

It has already included the compensation network, similar of what you've said. What do you think of it?

Rgds,
lhgin

[David Deratany]

I am using Altec 1631A single channel active crossover on one channel and I have an EVI XEQ-2 on the other. They are identical, as Altec made them for EVI. I think it does the job well.

I used at first a DBX 223, which I liked and which has great flexibility: separate level controls for input, low and high, also variable crossover frequency, phase reversal switches, and switchable 40 hz low pass filter. Not a bad unit. 24 db/octave slope. Not expensive at all, less than $300. I switched only because I got the others for real short money.

1w & 1w ? Many are happy with SET's of low power. Your results may vary. I am blessed to live in an area where I can and do play them at substantial levels; I could never make do with 1W & 1w, even if the nicest amplifiers ever made.

[joyspring]

Hi,

The Mastering Lab crossover is a passive network and would not offer the benefits of biamping.

An excellent active crossover is the Rane AC22; this crossover includes a delay for the LF section allowing time offset correction between the 604's LF and HF drivers.

In any event, choose an active crossover that includes an adjustable delay for the LF section. Once this axial offset is corrected, the on-axis frequency response, power response and overall sound quality will be markedly improved over most passive networks (particularly the stock Altec units).

The approximate offset of the 604 is ~7.2cm between LF and HF voice coils (be warned that this isn't necessarily a completely accurate measure of acoustic centers between drivers) - 0.2mS delay for the LF should be in the ballpark.

Regarding recommended amplifier power... this depends on the type of source material, room size/acoustic treatment, distance from the loudspeakers, desired peak-to-average ratio, etc.

Ignoring room loading and assuming a 16dB peak to average ratio (easily attainable with digital sources) and assuming an average listening level of 85dB at 2 meters with the 100dB 1W/1M efficiency of each 604, this would imply an average power requirement of ~0.1W/channel but a peak of 4W/channel.

Particularly for the LF section, I feel that a 1W amplifier would be far too anemic.

BobR

[marcuswilson]

lhgin,

Had a look at the Mastering Lab crossover. It is made with a completely different philosophy than mine. It has been tuned by the Sax brothers for the 604s. My crossover uses high slopes and very low loss inductors, to give the highest efficiency and the highest power handling (not a factor for you). I'm sure the Mastering Lab crossover is excellent.

I agree with BobR about the power, 1 watt is inadequate for the bass. Remember that the sonic signature of the low frequency amplifier is no where near as prominent as that of the high frequency amplifier. Also each amplifier is handling only part of the frequency range, so there is much less intermodulation and similar distortions.

I disagree on the amount of signal delay required for correct phase alignment as the delays in the filters of the crossover used makes the mere physical measurement insufficient. You need to do acoustic tests to get the delay time correct.

[joyspring]

Hi again ;-)

I concur with Marcus that delays in an active filter as well as a small amount of driver delay make physical measurements somewhat imprecise although they will get you in the ballpark... Having continuously-variable control over the delay line is convenient in this instance.

However, an active crossover with fourth-order Linkwitz-Riley filters (as the Rane AC22 and most recent examples possess) exhibits the same group delay for both high- and low-pass sections, so physical measurements will get you very close.

Time-domain measurement (now relatively inexpensive with PC-based tools) or impulse/square wave tests, and careful, critical listening (particularly of transients) are necessary to fine-tune this delay.

Despite the benefits of active crossovers, I will admit that I still use the passive Time-Aligned(tm) filters for the 604-8Ks in my UREI 811As, though I use an active crossover for the subwoofer.

Old preferences die hard ;-)

BobR

[lhgin]

hi guys,

these are quite some interesting and informative posts out there! thank you for all the feedbacks!


after reading the, i think i'll go with 1W (6C45PI) for the HF and 15W (6C33CB) for the LF. i've already got the 6C33CB built but have not for the 1W 6C45PI.

since active crossover is far much better than passive crossover, could you guys please recommend me one that i could DIY? perhaps with using tubes?

marcus,

could you please share with me what you're using for your altec? i find that the tapped inductor for the mastering lab is pretty hard to get.

cheers,
lhgin

[billfort]

Hi Ihgin. I use 604Gs and am in the same investigation phase you are in with regards to crossovers. I'll share a little of the thought process I have gone through but it might not count for much as I haven't done any of this yet. I'm building new cabinets first and using the stock crossovers externally (wired strait to drivers - no binding posts) so I can play around with different configurations.

My first thought was to use the Mastering Labs design as it uses compensation on the horn, which makes sense to me, and attempts to correct for the above mentioned time alignment problems. My concerns are all that "stuff" in the signal path, the complexity of the design and use of custom unobtainium components. I might have a source for building one of these here in Toronto - someone who is comfortable with hand winding those mult-tapped indictors, and will check this out. You keep hearing this is THE way to go passively with these drivers so I haven't ruled this out yet but?I keep thinking bi-amping.

The problem with bi-amping for me is that I'm a tube/analogue guy. I don't want sand circuits anywhere in the signal path and definitely don't want digital junk playing with the signal to accomplish delay. For me, it has to be a tube crossover, maybe with horn compensation. I might start by trying passive line level as active crossovers that do the above are pretty hard to find. The tube Marchand can apparently be configured for 12db slopes with compensation but this is a custom version of something that is already hugely expensive - even in kit form. Wayne at Pi speakers is apparently working on a tube unit with compensation and it might be available DIY - I'm watching where this goes - maybe exactly what I'm looking for.

On amps for bi-amping, I've been advised to proceed very carefully. These drivers need a crossover somewhere between 1200 and 1500hz and some that have tried this tell me that with a cross this high, you need 2 amps with the same sonic characteristics or it'll sound disjointed. I hear don't mix sand and tubes, don't mix push-pull and SET, don't even mix tube type like 300B and 2A3. Some even go so far as to say you shouldn't mix circuit layout - stick with the exact same amps! I guess you won't know till you actually try this stuff yourself but I'm going to take some of this advice - I'll use 2 300B amps (with different circuits) and see if I like the results.

Don't have much experience with real low power amps of less than a few watts (at least not with the Altecs), but I think you need a little more power for that 15" woofer. Maybe I push my 604s a little hard, but I like the dynamics I get with a whopping 7w from a 300B . You might want to give a low power amp a try first before committing to 2 flea power amps.

Again, this is all just speculation and random thoughts. I love the 604Gs with the stock crossover and an all tube 300B SET system (vinyl/SACD front end). I just want a new crossover scheme that might refine things a little bit and maybe squeeze a little more magic out of my system without throwing the baby out with bathwater.

Billf