have you ever tried to use bypass capacitors in parallel to other
capacitors of the crossover?
Yes.
Do you think this could have a positive impact?
I thought there would be some impact, but coudn't hear it - no matter how much I TRIED to hear the impact.
I have found significant & positve impact using bypass capacitors in power supplies, and coupling circuits. Unfortunately I found no impact using bypass capacitors behind my tweeter.
I believe the reason for this is the environment of application.
Given a power supply electrolytic the smaller metallized poly bypass will reduce ESR siginificantly. This helps drain the grunge present in the power lines. It also helps drain the ringing/ripple from the diodes. Better diodes will generaly have less ripple, but there is still some ripple. I was skeptical about the impact of bypassing in power supplies BEFORE trying this experiment. After the experiment, this changed. Using smaller "high quality" bypass capacitors around bigger electrolytics works very well.
Given a coupling application, the circuit impedance is very high. This high impedance exacerbates the importance of Time Constant. Time constant - the rate a capacitor fully "drains", is dependent upon the size of the capacitor, and the impedance of the circuit. Bigger capacitors drain slower. Higher impedance circuits drain slower. Hence the importance of using a smaller capacitor in high impedance circuits becomes very important when considering time constant.
I do realize there is a plethora of discussion about Time Constant online. I believe much of this discussion among laymen is based on hear-say and liberally applied generalities. I find this somewhat irritatating. Formally, the AuriCap site conveyed that bypassing in the signal path is wrong
http://www.jacmusic.com/auricap/htm/auricap_application_notes.htm I obviously disagree. I have tried this and openly disagree with their statement.
In contrast George Short ( NorthCreekMusic ) conveys affirmation for bypassing in loudspeaker crossovers.
I am in the middle on these issues. I fully understand time constant issues as addressed by Horowitz in the Art of Electronics. This issue is very simple. Beyond this... can I hear it.? In loudspeaker crossovers using quality capacitors the answer is "no". In high impedance circuits (i.e. bypassing) and when bypassing power supply electrolitcs the answer is affirmatively "yes".
What do you think of axon capacitors used in xovers?
I think they sound on par with Solen capacitors, but I prefer the stiff/solid lead of the Axon. Most of the Axon capacitors seem to have a stiff/solid lead. Most of the Solen capacitors have a stranded lead with insulation. I think either of these capacitors are better than Bennic capacitors. I think Axon and Solen capacitors are a very wise budget capacitor and found in commercial products up to $15k. I'd easily use an Axon capacitor in a parallel circuit behind a woofer.