Bypass Cap in XO

Status
Not open for further replies.
Why not just use a decent cap in the first place that doesn't need bypassing? It's not like PP/foil is exotic or horribly expensive.

If you have a high ESL bipolar electrolytic, I can see how a bypass could slightly change the transfer function, but a good PP cap? No.
 
Why not just use a decent cap in the first place that doesn't need bypassing? It's not like PP/foil is exotic or horribly expensive.

If you have a high ESL bipolar electrolytic, I can see how a bypass could slightly change the transfer function, but a good PP cap? No.

Agreed.

One could input all the cap parameters like ESR and capacitance and model XO in SPiCE with and without bypass cap to see if there is difference in frequency response. I think it would who up.

I think you may have missed the point. As I noted above, assuming we're talking about small value 'bypass' (not really but never mind) capacitors added to a large main value, the objective was not to make any significant change to the FR or bald electrical transfer function per se. When they first started to be widely used in the late 1970s / early '80s, there were two basic goals: firstly to improve consistency with fairly mediocre bipolar electrolytic caps (sometimes Mylar film types), and secondly to use the faster rise-time of the small cap to improve definition on musical transients. The latter is rather more contentious 😉 but you can follow the (slightly dodgy) reasoning. So if one wants to make a SPICE model, these variables will need to be included. Without, all it will do is demonstrate that they don't significantly affect something they weren't intended to affect.

With a quality MKP as the base cap? I wouldn't bother with the trad. small value bypass. I'd either leave it alone, or make about half the main value out of a PIO or film & foil.
 
Last edited:
Don't get me started on this!

Look, all this bypassing stuff is so dependent on the speaker/amp/rest of system that the only way to know if it will make a difference with YOUR speakers and YOUR ears is to try it for yourself. Just solder a small bypass cap across the main cap and see if you can hear an improvement over a wide range of music. If you can, leave it in. If you can't be sure leave it in or take it out, no matter. If it makes it sound worse take it out.

Costs next to nothing to try. Simples!
 
The diminishing returns post is reasonable. The other important thing as I see it is that a crossover won't stand in the way of good sound even with moderately priced components. A crossover begins with baffles and sources, not capacitors. There is simply too much else going on to think that capacitor perfection is the key to it. Even electrolytics can pass good sound.

In cases where different capacitors change the response, they aren't necessarily doing anything to improve the crossover proper.

I think the unspoken assumption in all these bypass cap threads is: everything else has been perfected except the XO. It needs a little 'tweaking' mindset prevails.....
 
Don't get me started on this!

Look, all this bypassing stuff is so dependent on the speaker/amp/rest of system that the only way to know if it will make a difference with YOUR speakers and YOUR ears is to try it for yourself.

Very true.

Just solder a small bypass cap across the main cap and see if you can hear an improvement over a wide range of music. If you can, leave it in. If you can't be sure leave it in or take it out, no matter. If it makes it sound worse take it out.

Costs next to nothing to try. Simples!

That would be the case assuming you're using inexpensive little MKPs or polystyrenes, the elephant in the room being that the OP was talking about Mundorf Supreme silver/oil as the bypass. Based on UK retail price, at today's approximate exchange rate, the smallest value available of that capacitor (0.01uF) would cost approximately $30 apiece, not including postage, which is not what I'd define as 'cost next to nothing', and a potentially expensive way of determining whether there's any advantages, disadvantages or the square root of jack difference in a given application. 😉
 
Status
Not open for further replies.