|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
76.112.135.123
Anybody heard or experimented with this?? My friend asked me about it and seems like JBL is using this crossover topology in their top line of speakers.
Edits: 03/04/09Follow Ups:
Old thread.
This took me 2 minutes to find. JBL idea?
Edit - add:
I read that using the charged caps helps the low level details. I also read that using PLLXO also increases low lever details.
For those who have experience with both, can you offer further comments on how things sound between these two approaches?
I have been fortunate to meet Greg Timbers of JBL on several occasions. He has mentioned the battery biasing of capacitors more than once, but just as with RCAfan mentioning conical horns it took a while for the importance of the information to make it through my thick skull. As with the conicals, when I finally experimented I was delighted with the results of this "charge coupling" technique.
Greg describes the technique creating a class A situation. I think this is correct. I have long thought that the remarkable retrieval of fine detail possible with single ended tube circuits is due to the avoidance of a zero crossing point with zero signal, i.e. the transformer(s) are biased up to some point between cutoff and saturation by plate current. Instead of tiny microvolt signals (reverberant tails, etc.) being distorted by nonlinearities in the crossover region where magnetic fields are collapsing and reforming in opposite polarity with every audio half cycle, these tiny signals sail on through the most linear operating region of a biased transformer as just a tiny modulation of the quiescent bias point.
So now it seems we have the same situation with capacitors. In the usual passive crossover, to pass the audio signal the electrostatic fields have to collapse and reform with every half cycle, and the smallest signals live in the most "dirty" nonlinear region of the device. Biasing the capacitors involves extra cost and expense, as one needs to two capacitors of double the value of the unbiased circuit (four times total capacitance) to achieve the same filtering. Rich and I have found that with the high pass crossover to our midrange / high frequency driver the improvement in sound is not subtle, more of a "Wow" experience.
Greg Timbers discusses the technique in the third post of the Lansing Heritage thread linked to below.
It would seem at first thought, that what you describe might be far more critical with high efficiency drivers, as the signal levels we are seeing are quite small, relative to a "traditional" speaker.
With 106db horn speakers, we're not talking watts, those "fine details" you speak of may represent only a few milliwatts of power, very small signals, which could easily get distorted in a complex crossover.
Well worth a try!
OK Rabbits, happy now?
Hello Ivan (and rabbits),
I think you make a good point about this technique perhaps being particularly useful with high efficiency speakers. We live in millivolts and below nearly all the time. With the efficiency of most compression drivers, hall ambience, audience coughs etc. are down in the microvolts region. Combining this with the superior IMO resolving power of these devices it is not surprising that crossover improvements loom large.
Wonder what all of this means when it comes to selecting capacitor brands and types for use in X-over designs?
Did the capacitors we thought were the "best sounding" sound good because the resisted the very problems with biasing now eliminates?
Do we get to start a whole new era of "cap rolling" in out x-overs?
Inquiring minds what to know!
OK Rabbits, happy now?
I've wondered the same thing, but haven't yet experimented. What if some charge coupled doggy-doo electrolytics would outperform non-biased living legend pixie dust caps?
JBL seems to like a combo of Solens and electrolytics in speakers using this technique.
Not exactly Duelunds!
OK Rabbits, happy now?
Here is the link to a profile of Greg Timbers. BTW, he is about the farthest thing possible from the wild eyed audiophile who obsesses over every questionable nuance, so I should have picked up on his enthusiasm for charge coupling sooner.
.
what kind of caps does this circuit like best? - electrolytic?
JBL uses Solens in this system for the Mids and UHF. Use Electrolytics for the LF section. They tend to split it up that way. In the Everest 2 they use a combination of Electrolytics and Solens on the woofers and all Solens on the Horn and UHF.
Edits: 03/08/09
thanks!
Here is a link to check that shows a schematic if you have not already...
more...
In this arrangement can you replace non-polarized electrolytics with standard polarized electrolytics? looks like so from the JBL cross over pic.
I am not grasping what charge coupled means
S
http://audioheritage.org/vbulletin/showthread.php?t=43&highlight=battery+bias
http://www.lansingheritage.org/images/jbl/specs/home-speakers/1993-k2-s5500/page11.jpg
Very interesting!
I got a charge out of it........
I've been using it with our Cogent horn system for a couple of months now, it works just like Robert says. Well worth the effort.
Hey Rich!
Very cool, If applying a charge to cables to improve (stabilize?) dielectric properties has a following it only seems logical that parts that require a dielectric to operate would also benefit. For years people have been saying capacitors sound better with DC on them and it great to see people experimenting with this idea in the speaker world. I wonder where on earth you are going to find 12V or so to bias your caps tho' :-)
dave
Hi Dave, we got yer 12 volts right here! We're using 9 volt batteries for the crossovers though. As with the battery biasing of tubes, there is no load on them and the batteries last their shelf life. Best, Steve
Rather than use two identical caps and the large value resistor, it would be interesting to try the proper valued cap and a series battery (which could then be bypassed with a small cap. This would actually be really easy to try in any existing crossover design.
dave
Dave I like your out of the box thinking, though I don't think this would work due to the equation for total capacitance of two capacitors in series, C1 X C2 divided by C1 + C2. Whaddya think?
the battery has a very low AC impedance so in theory it behaves like an infinite cap. The .1u cap is there as a local bypass
dave
Hello Dave:
I'm willing to try your implementation so can you please let me how I go about implementing this in the capacitors of the attached x-over schematic?
---thrax
lemmme see, where would he have a source of 12V in that design?...yer obviously kidding I think.
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.
.
They were a carnival of American decay on parade, and they had no idea of the atrocity they had inflicted upon themselves–Henry Chinaski
I have built several networks using it and it works. Better lower level detail. Can get expensize with twice the capacitor coumnt at twice the value.
Could it be that you could might be able to get away with using cheaper caps? Could this be the reason JBL/Harmon is doing it?
IIRC, JBL uses electrolytics in their crossovers.
OK Rabbits, happy now?
I have always used Poly caps, so I have not tried it with the electrolytics. Should work there as well. Many of the JBL CC networks are Electroltics or combinations of them with Polypropylene and Mylar caps mixed in.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: