• 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

Crossover Capacitor Quality and Kirchhoff's Laws

Agree or disagree

  • Agree

    Votes: 3 75.0%

  • Disagree

    Votes: 1 25.0%
  • Total voters
    4
  • Poll closed .
H

hh66

New Member
In almost every thread regarding recapping vintage speakers, people refer to capacitors which carry the signal "through" vs. those that "bypass" a driver and therefore don't benefit from a higher quality capacitor. It's my opinion, grounded (no pun intended) in Kirchhoff's laws, that this is complete nonsense. Any circuit element is going to have an effect at each circuit junction in which it is connected.

Film capacitors are generally regarded as being superior to electrolytic due to the dielectric having less store and release activity which can colour the sound by leaving time delayed artifacts in the form of signals that are additive or subractive to the original applied signal depending on instantaneous phase relationships. This results in smearing and a resultant loss of fidelity, clarity, definition, or imaging. Per Kirchoff's laws, this effect will be of equal significance regardless of whether the capacitor in question is in a "through signal" or "bypass" position.

I'd like to hear any argument of how this could be an incorrect conclusion.
 
Register to hide this ad
It a bit of a flame bait topic, hh66.

Nonetheless:
1. Capacitors which have constant voltage across them cannot produce distortion, so their quality matters less.
2. Capacitors from the signal path to ground generally do not meet that condition and may contribute to signal degradation.
 
Sorry, I had no intention of flame baiting. I'm trying to see the other side of this argument but I can't so I'd like to know what the reasoning is because it seems most people have convinced themselves that bypass matter less, so something must have led them there. I believe power supply capacitors wouldn't benefit from lack of dielectric deficiencies though because of this reason. I don't think a crossover cap would ever have constant voltage across it of significance unless the amp has a DC offset problem, so not sure how that enters into it.
 
Last edited:
Logically the effects of any component will vary depending on where they are in the circuit. A coupling capacitor is much more likely to have an effect on the signal that an RF blocking cap on the power rail, for instance. De-coupling caps shouldn't have much effect on signal either because they are a drain -- signal goes in and nothing comes out. The will have some effect, since they are electrically connected, but without knowing the math, I'd be willing to bet any effect is very small, much less than capacitor defects directly in the signal path.

One thing I do know to be a fact is that capacitor networks are NOT simply additive, and the deviations from value and impedance are strongly dependent upon the values of the capacitors in parallel.
 
I don't think RF blocking caps are a fair comparison because their direct effects are outside the audible range. In a crossover, all the caps are functioning in the audible range. The qualities and deviations from perfection of a cap across a midrange driver to block high frequencies has as much effect as a cap in series that blocks low frequencies.

You could argue that a cap across a woofer might not be as big a concern beause the ear is less sensitive at those frequencies, but on the other hand, if the dielectric is emitting out of phase high frequency signals, what's stopping them from contaminating the signal arriving at the midrange or tweeter?

ESR is of course very important and film caps perform much better in that respect, so if a film across a woofer were replaced with an electrolytic, it would be similar to adding an up to several ohm resistor in series which has to be significant. I could see a crossover which when new performs well with electrolytics on an analyzer would show significantly different results when the damping effect of the ESR is removed by a film capacitor.

This stuff is always more complicated than it seems at first.
 
Thats why JBL with their DD series Synthesis speakers use batteries to make sure the capacitors are always in a state of readiness. I prefer to tri or bi-amp myself with the passive crossovers being removed with some speakers. I mainly do it to get better coupling between the amps and the woofers.
 
In my experience it is absolutely possible to hear the difference changing from electrolytic or oil-filled capacitors to PE, PS, or Mylar types because of their lower DF/ESR. The result will be an increased sound stage and “crisper” response in the mid to upper frequencies. But will there be an improvement from changing only the bypass capacitors, particularly in the woofer circuit? Although I haven’t tried it in only the woofer circuit, I must assume there will be a difference because a lower DF also means lower resistance (ESR), inductance and phase shift, along with higher Q, because ALL of this is what you get with lower DF stuff. (for interesting reading, click the following link to see an article on capacitors at:

https://www.electronics-notes.com//...nce/esr-dissipation-factor-loss-tangent-q.php

As has been suggested in earlier threads, such “changes” may not be so perceivable in the lower frequencies, but audible or not, the changes are there. But how far to go? Even changing to simple PE capacitors in the mid and twtr circuits will certainly bring a smile to your ears! The need for further changes will be dictated by your other equipment and purse.
 
Thanks for all the replies. I'd like to reiterate a subtlety that I think may have been overlooked: although the woofer bypass cap shunts high and mid frequencies, hopefully leaving only the correct low frequencies, it's my contention that the imperfection of the shunting will affect the high frequencies going to the mid and tweeter as much as the "in circuit" caps going to those drivers due to Kirchhoff's laws. There will be some high frequency ringing from the dielectric releasing these signals out of phase with respect to the original signal. That's why I think it's wishful thinking that corners can be cut on the woofer bypass cap. (No I don't sell large value film caps, LOL.)
 
hh66 said:
Thanks for all the replies. I'd like to reiterate a subtlety that I think may have been overlooked: although the woofer bypass cap shunts high and mid frequencies, hopefully leaving only the correct low frequencies, it's my contention that the imperfection of the shunting will affect the high frequencies going to the mid and tweeter as much as the "in circuit" caps going to those drivers due to Kirchhoff's laws. There will be some high frequency ringing from the dielectric releasing these signals out of phase with respect to the original signal. That's why I think it's wishful thinking that corners can be cut on the woofer bypass cap. (No I don't sell large value film caps, LOL.)
Click to expand...

I agree. But, large value bass shunt capacitors, tend to be expensive if a poly cap is used, and physically large too. So, I either use a combination of NPE and poly, or an NPE cap, with a bypass cap in parallel. Somewhat similar to one another, though pricing is lower with the NPE/bypass combination.
 
I still think that the presence of the NPE, bypassed or not, will cause phase smearing that the film would be much better at. BTW, partsconnexxion has 60uF film caps at 91% off right now through December. I just bought some for my Boston A150s and might use the other in one of my Acoustat pairs.
 
Is it the case that the current is different at each node due to the difference in impedance of the bypass capacitor and the driver voice coil? So you would have less current through the bypass than the driver.
 
The higher ESR of the NPE is going to make it less efficient at filtering which the film bypass will help, but I think a single film (or several to obtain the correct value) will perform better. The lower capacitance film bypass can only function at the higher frequencies, so it's likely the higher its value, the better.
 
Of the speakers I actively use, my AR90's have a 350uF shunt, my AR92's an 80uF, my AR58S's a 100uF, and my Martin Logan's, a 180uF shunt. I guess I could bundle caps together, starting with those 60UF caps? Off to Parts Connexion to find those caps on special...! :)
 
hh66 said:
the woofer bypass cap shunts high and mid frequencies [...] it's my contention that the imperfection of the shunting will affect the high frequencies going to the mid and tweeter [...] due to Kirchhoff's laws.
Click to expand...
You've got your "wires crossed" somewhere...
 
hh66 said:
...people refer to capacitors which carry the signal "through" vs. those that "bypass" a driver and therefore don't benefit from a higher quality capacitor. It's my opinion, grounded (no pun intended) in Kirchhoff's laws, that this is complete nonsense.
Click to expand...
You are correct. I've said the same thing myself several times on this forum but often get told I'm an idiot.
 
You must log in or register to reply here.
Back
Top Bottom