Charge coupled crossover

[Phillip Jeffries]

This weekend I tried a charge coupled capacitor arrangement for the big capacitor in the Altec series style crossover.

I’m pretty impressed. If I were to try to describe the sound, it’s like doubling your amplifier wattage. Just clean, articulate.

Maybe because I wasn’t using a great capacitor there in the first place? I don’t know. Haven’t swapped the others because I’ve got nice KBG PIO’s there.

Apparently, JBL uses this trick in their high end speakers….it’s putting two caps in series with a dc bias on the center between them.

Wondering if anyone else has tried it.

 

[MrEd]

This weekend I tried a charge coupled capacitor arrangement for the big capacitor in the Altec series style crossover.

I’m pretty impressed. If I were to try to describe the sound, it’s like doubling your amplifier wattage. Just clean, articulate.

Maybe because I wasn’t using a great capacitor there in the first place? I don’t know. Haven’t swapped the others because I’ve got nice KBG PIO’s there.

Apparently, JBL uses this trick in their high end speakers….it’s putting two caps in series with a dc bias on the center between them.

Wondering if anyone else has tried it.

I'm interested but don't quite grasp the concept.
Can you share more details?

 

[Phillip Jeffries]

Yes, here are some links…

Charge Coupled Crossovers Article By Jeff Poth

Charge Coupled Crossovers Article By Jeff Poth, DIY AudioProjects And Articles On Enjoy the Music.com

www.enjoythemusic.com

This one has links to the JBL documentation…

Charge Coupled Crossovers

For some years Greg Timbers the designer of JBL's statement speakers (Everest, K2 etc.) has advocated the use of Charge coupled crossovers in both these new JBL's and as an upgrade to older models. The benefits are supposedly to achieve as near a Class A biased crossover as possible...

theartofsound.net

 

[240sx4u]

I was going to do this with my current speakers but ordered the wrong caps and then forgot about it. Thanks for the reminder!

 

[Michael V]

This weekend I tried a charge coupled capacitor arrangement for the big capacitor in the Altec series style crossover.

I’m pretty impressed. If I were to try to describe the sound, it’s like doubling your amplifier wattage. Just clean, articulate.

Maybe because I wasn’t using a great capacitor there in the first place? I don’t know. Haven’t swapped the others because I’ve got nice KBG PIO’s there.

The last bit is exactly what I concluded when I tried this (using 9v batteries). The charge coupled capacitor trick made lesser capacitors sound tolerable. For example, I used some big oil caps that sounded slow and overly dampened by themselves, but with charge coupling the energy and dynamics were much improved.

But I preferred the sound with really good caps and no charge coupling (in my system, etc etc). This was in the high pass of my crossover with TAD 2001s and various horns.

 

[Phillip Jeffries]

The last bit is exactly what I concluded when I tried this (using 9v batteries). The charge coupled capacitor trick made lesser capacitors sound tolerable. For example, I used some big oil caps that sounded slow and overly dampened by themselves, but with charge coupling the energy and dynamics were much improved.

But I preferred the sound with really good caps and no charge coupling (in my system, etc etc). This was in the high pass of my crossover with TAD 2001s and various horns.

I’d like to hear some TAD drivers some day. What caps did you end up with?

 

[Michael V]

That's a good way to derail this thread entirely, but seeing as though you started the thread...

I am using vintage caps made by Western Electric mid-century. I get away with really small values, so they're not expensive and sound better than anything I've tried. @hifitown had a nice thread about vintage caps, and deserves full blame credit for my starting on that path.

 

[EarlK]

Due to being an almost charter member of LHF since its inception over 20 years ago, I've used DC-Biased networks ( speaker crossovers plus amplifier coupling capacitors ) for a good 20 years ( though not for the last 8 months ).

I have quite a lot to say about the topology.

( When time & interest permits ) I'll post a lot more later about the Pros & Cons as I see them coming from my own experiences.

 

[Wntrmute2]

Got any pics/schematic of how you implemented this?

 

[MrEd]

Reading most of the comments it seems that the effect is dependent on the type and quality of the caps used.
My preference for oilers and Teflon not specifically mentioned, but I believe I have enough on hand to give this a try.
Any one have thoughts on the resistor wattage rating needed ?

 

[Wntrmute2]

Since the 9V batteries are supposed to last practically forever, I'd guess they could be pretty small. I'd probably go with some 3W jobs and monitor their temp for a few hours.

 

[EarlK]

Reading most of the comments it seems that the effect is dependent on the type and quality of the caps used.
My preference for oilers and Teflon not specifically mentioned, but I believe I have enough on hand to give this a try.
Any one have thoughts on the resistor wattage rating needed ?

2 to 3 meg Ohms is what I believe JBL first used some 25 years ago.

More recently ( last decade ), JBL got rid of the battery and used ( parasitic ) voltage derived from the woofer circuit ( partially rectified and bled into the HF network by use of a zener ?? diode and small resistor ).

Search out the JBL 4367 network and you'll find some of the necessary details of the newer topology.

Something to consider ( a "CON" if you will ) that applies to running AC signal through a cap that's holding a DC charge, is you should find ( based on your own aural acuity ) that the "Personality" of that capacitor is greatly diminished ( or completely stripped away ).
- IOW, if you paid big bucks for a capacitor that has a distinct voicing, then it's going to change to just what that raw dielectric should sound like .

Only you can decide if that's a good or bad thing ( at the very least, it's quite educational in comparing different dielectrics ) .

IME, once a cap has been exposed to ( holding ) DC ( & upon removal of that DC charge ), it never fully returns to it's as new condition ( essentially, it's been "broken-in" >> inevitably losing some sparkle, especially with cheaper MPP type capacitors ).

 

[Phillip Jeffries]

I’m using (2) 1/8 watt, I think, 1 meg at each end of the battery. I kind of want them to be like fuses in case of some kind of shorting accident.

When I was comparing my 13 uf capacitor, which was actually 3 capacitors in parallel, to the charge coupled 13 uf, which was actually 4 capacitors…20 uf + 6 uf, in series with another 20 uf + 6 uf, in mono, what seemed to go away, was this upper midrange hash. That hash, actually kind of made horns, the instruments, sound a bit more brassy and realistic, though. Maybe even a bit more dynamic, like the charge coupled was somewhat muted in comparison.

But, having listened for a good number of hours, not just acoustic jazz, but dense mixed rock, etc I’m convinced it’s a good step up. Really pleased with it, although it seems wrong since it’s not shortest signal path. Kind of thinking about doing one of the other two capacitors, actually.

What I found in my experiments with Altecs, is that the recordings that just never sounded good were the really dynamically compressed ones…modern pop style music especially…to the point of being unbearable, based on sound, not content.

This mod, really fixes it. Listening to the exact same recordings, that hashy quality is removed, and more details of the recording are revealed.

But, yes, when I upgraded from polypropylene or Mylar caps I had on hand to Kbg PIO’s it also reduced those hashy qualities. So, maybe the mod is only effective on certain types of caps? Aren’t PIO’s actually higher dielectric absorption than plastic film types? Still makes me curious to try it on PIO’s, even though I’m really pleased with where I am already sound wise. Maybe it can get better.

 

[targeteye]

I had not come across this concept until now. My initial impression is that if the goal is to maintain the Dielectric in a 'pre-stressed' state to prevent complete relaxation at the zero-crossing point, 9 Volts may not be sufficient. The later design that utilizes the peak voltage delivered to the woofer seems more likely to achieve the intended objective.

Its very interesting to contemplate but to be a "believer" I'd need to see some analysis of Dielectric absorption vs voltage. Do these graphs exist?

 

[EarlK]

Put a volt-meter on the output of ones amplifier and most people find they listen ( at normal listening levels ) to less than a volt ( 2 max ).

9 Volts is way more than enough >> though JBL's initial implementation was actually 18 volts DC.
- Frankly, I don't like listening to this effect when there's any more than 1 volt DC within the capacitor-pair ( but then it seems all of my Solen capacitors display a tiny bit of leakage after a couple of weeks ).

Even a couple of DC millivolts ( measured across my Altec 288's ) drive my ears to distraction as to "what's wrong here??"

There's not much to believe here >> if one can't hear the subtle differences between dielectrics of different materials >> then this circuitry is an absolute waste of that persons time and attention ( let alone expense ).

It's effect just won't be audible.

 

[EarlK]

Got any pics/schematic of how you implemented this?

Here's one of JBL's earliest implementations ( for DC-Biasing caps in crossovers );

 

[targeteye]

Put a volt-meter on the output of ones amplifier and most people find they listen ( at normal listening levels ) to less than a volt ( 2 max ).

No offense but putting a voltmeter on the output of you amplifier will not tell you much about the complex waveform being fed to to it.

I'm gonna bow out of this. If it sounds better to you then that's all that really matters.

 

[BMWCCA]

🙀 This reminds me it's probably time to change the batteries in my custom charge-coupled crossovers in my JBL 4345s from an updated schematic offered by the speaker's designer Greg Timbers.

 

[Audionut]

🙀 This reminds me it's probably time to change the batteries in my custom charge-coupled crossovers in my JBL 4345s from an updated schematic offered by the speaker's designer Greg Timbers.

Care to post pics of said 4345’s, asking for a friend.

 

[BMWCCA]

How about a glamor shot from the guy who built them?
They are not original JBLs but clones build by the Rick in Kenrick who is their cabinet maker who actually lives here in the USA. Mine are one of two pairs made with black ash baffles, which I really like since no one will take them for original JBL studio monitors of which all 200 were blue. They were populated for the first owner with new or JBL serviced drivers by Dave Katz who built the crossovers from the Timbers circuit design. The diffraction lens came from Japan. I bought these in Connecticut during a snow storm on my birthday 15-years ago and drove them back to Virginia.