• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Do different vendor crossover caps really sound all that different?

Which speaker Crossover Capacitors do you prefer


  • Total voters
    78
You don't even need a battery, and many of the JBLs go without. They use four diodes to create a full wave rectifier to produce the dc current from the audio signal, then store this voltage in a capacitor which substitutes for the battery.

This is genius... Although the DC will be more or less modulated by the signal.

Do you have a schematic of this solution actually implemented by JBL? Or maybe some JBL models where this was implemented?

Thanks for the info.
 
The JBL 4367 has a charge-coupled crossover and uses this method. Note the diodes:

index.php
 
This also plays to what I was saying about the Bean Counters. Note the caps are not marked as per manufacturer. Also note that the caps are rated at 100 volts. And while charge coupling would double this voltage rendering a maximum input of about 600 watts (more than enough), Audyn caps are rated at 400 volts and Clarity caps at over 600 volts.

Don't you think that the JBL engineers would rather use some nice "boutique" caps" but the Bean Counters say this is "good enough?"
 
In saying that designers don't listen to components, I think you are thinking, perhaps, about more modern designers, who have a wealth of choices available to them off the rack, or to spec, and have crossover design programs on their computers.
But older speakers (which often are the most likely candidates for recapping), such as ARs, KLHs, Bozaks, and so on, were designed pretty much from the bottom up - I talked with (and bought a Mac MR 67 from) a retired MIT professor who had helped Ed Vilchur optimize the magnetizing of the drivers for the early models, and I remember reading about Henry Kloss with his room full of Waring blenders each with a different formulation of pulp for cones, and his little musical carousel whose tinging he used as a reference for the high end of his speakers. Certainly they listened carefully (and most people in hifi then actually knew what acoustic instruments sounded like, so their listening was informed listening). And they would have been hearing the ESR of caps when tweaking. So we are more likely to hear what they intended if we are aware of its importance.
We can blame them for not anticipating that electrolytic caps would degrade over multiple decades, but no one expected that speakers would be used more than a half a century after they were built.
 
I don't really believe that MOST speaker manufacturers even consider such things as ESR or the resistance of the inductor. They work with given vendors and use what's available to them to meet a price point set by the Bean Counters even if it means using a higher quality by-pass cap to "crutch" a cheaper cap, or using thinner wire for the chokes

Sure they "voice" the speakers, but they do it using the limited resources they are allotted.
Lets say this is true. Manufacturers don't consider ESR and DCR of the inductors. We don't know this per se, but lets assume it is true.

It is still so that if you change these parameters, you change the output of certain individual drivers thus giving the speaker system a different "voice"....right?
 
No doubt. But consider that the difference in impedience between two woofers pulled off the shelf is probably far greater than the difference in upgrading say 18 gauge to 15 gauge wire in the choke.

Also, cheaper speakers will use electrolytic caps and the variation on those could be far greater than other parameters, so would any two speakers really be voiced exactly the same to begin with?
 
I have not compared different brands on the same pair of speakers but If I did, I would use EPI 100s with a single 10uf cap for comparison. Would play some demanding material with pre-amp switched to mono and turn balance knob full left and full right to compare, if that makes sense. Differences would only be from about 1900hz and above though. I do have a pair of EPI 100 with npe caps and another with Dayton film caps, all selected to less than 1% tolerance and cannot tell the difference. ESR was about the same on both types, less than 1 ohm.
 
No doubt. But consider that the difference in impedience between two woofers pulled off the shelf is probably far greater than the difference in upgrading say 18 gauge to 15 gauge wire in the choke.

Also, cheaper speakers will use electrolytic caps and the variation on those could be far greater than other parameters, so would any two speakers really be voiced exactly the same to begin with?
Wether or not two off-the-shelf drivers of the same part number have differing impedance graphs and wether or not the tolerance of the ecaps are sloppy, the speaker still gets re-voiced when altering the parameters of the caps and inductors. We are talking about one thing...how speakers get re-voiced when changing components in the crossover....and you keep bringing up other irrelevant things. Interesting things nonetheless, but irrelevant.

To your question....

would any two speakers really be voiced exactly the same to begin with?

With tight enough tolerances in manufacturing of the drivers, crossover components etc....yes. With sloppy tolerances, the voicing of one speaker to the next of the same part number may vary.
 
Caps are not semiconductors. Transistors and diodes are.

Thanks..
I know the distinctions .. and the definition's enable the discussion .. "pass the signal unaltered"
Probably not a good thing for me to be tossing indiscriminately in to the middle of a good discussion ..

But given sufficient voltage everything's a semiconductor .. as in a "partial" conductor .. that's what the cap is doing ..
The "insulator" in the cap is conducting, but by design it's altering the signal.
 
Last edited:
Don't you think that the JBL engineers would rather use some nice "boutique" caps" but the Bean Counters say this is "good enough?"

I think that perhaps the engineers have determined that dielectric absorption is the main problem, and this is to be overcome with the charge-coupled solution, not via buying expensive capacitors.
 
I think that perhaps the engineers have determined that dielectric absorption is the main problem, and this is to be overcome with the charge-coupled solution, not via buying expensive capacitors.
...And don't you think that if cost were no object, they wouldn't charge-couple with expensive, boutique caps like some nice TOTL Mundorfs? Recognize that Stereophile Magazine only gave the JBL 4367 a "B" rating.
 
I have not compared different brands on the same pair of speakers but If I did, I would use EPI 100s with a single 10uf cap for comparison. Would play some demanding material with pre-amp switched to mono and turn balance knob full left and full right to compare, if that makes sense. Differences would only be from about 1900hz and above though. I do have a pair of EPI 100 with npe caps and another with Dayton film caps, all selected to less than 1% tolerance and cannot tell the difference. ESR was about the same on both types, less than 1 ohm.

I really like some of this idea. I don't like the "mono" part as I think you can't make any determinations on the imaging.

But talk about cutting to the chase!
 
I have not compared different brands on the same pair of speakers but If I did, I would use EPI 100s with a single 10uf cap for comparison. Would play some demanding material with pre-amp switched to mono and turn balance knob full left and full right to compare, if that makes sense. Differences would only be from about 1900hz and above though. I do have a pair of EPI 100 with npe caps and another with Dayton film caps, all selected to less than 1% tolerance and cannot tell the difference. ESR was about the same on both types, less than 1 ohm.
Primary difference would be that in accordance with the laws of physics, two speakers cannot occupy the same space at the same time so placement always comes into play in a typical room.
 
Primary difference would be that in accordance with the laws of physics, two speakers cannot occupy the same space at the same time so placement always comes into play in a typical room.
Perhaps I was not clear in my response. Each speaker would have a different brand or type of capacitor and I would only be listening for any differences in sound quality from one speaker at a time both having the same mono source. Suppose measurements could also be made, but I all I have for testing is a Stereophile test CD with warble tones and a SPL meter. Imaging, soundstage etc. would not be part of the test.
 
Perhaps I was not clear in my response. Each speaker would have a different brand or type of capacitor and I would only be listening for any differences in sound quality from one speaker at a time both having the same mono source. Suppose measurements could also be made, but I all I have for testing is a Stereophile test CD with warble tones and a SPL meter. Imaging, soundstage etc. would not be part of the test.

I've done similar. Change caps, with different brands in each speaker. I've narrowed down different brands, testing this way. Now, I typically only use Mundorf, after comparing at least 6 brands of caps in A/B tests.
 
Perhaps I was not clear in my response. Each speaker would have a different brand or type of capacitor and I would only be listening for any differences in sound quality from one speaker at a time both having the same mono source. Suppose measurements could also be made, but I all I have for testing is a Stereophile test CD with warble tones and a SPL meter. Imaging, soundstage etc. would not be part of the test.
I understood, maybe you didn't.

In order to instantaneously A/B the speakers they need to occupy the same place in space or they will not sound exactly the same and this difference may be enough to overshadow actual changes associated with the manufacturer of the cap.

If they are not in exactly the same place, room acoustics, or even just an inch difference in height relative to your ear can make a difference in their sound. And if you do try to put them in the same place, you can't A/B them without some time gap/audio memory loss that could easily obscure any real differences.

Then there is the very real probability that both channels of the amp are not identical. Read equipment reviews in Stereophile to see that both channels are rarely identical.

The way I do it is to instantaneously A/B the caps in the crossover. Use two sets of three long wires, each set with one common connected to the other two other wires (A and B) through a DPDT switch with alligator clips on the open ends. Solder one leg of each cap both in the same place in the circuit. Connect one of the the A and B wires to the open leg of each cap and connect the common wire to where the caps are to go. Then I can sit in the sweet spot and flip a switch back and forth and instantaneously hear if there is any difference.

Nothing else changes and there is no time gap.
 
Last edited:
Well if we are going to delve into minutia..the continuous changes the air between the speakers and listener make any moment-to-moment comparison impossible.
 
To say nothing of psychology. We are intrigued by difference, and not alert to similarity.
And then there is the difference between every day listening and the hyperfocused listening we do when trying to decide. Which is more accurate, long term?
 
Back
Top Bottom