capacitor voltage ratings

tajanes

I’m looking at trying a notch-filter crossover (to use on the mid to smooth out the peaking prior to its roll-off, and then come in with a filter for the tweeter to fit net measurements) and would appreciate input as to what voltage rating I should be looking at for the capacitors.  Seems to be quite the range; 250, 400, and up.  I’d be anticipating powering with 60w tube amp, but would like to know if down the road I used a 150w amp what rating would that require?  

 Thanks,

Andy

dynamo

For one thing, a 250 V Is more than sufficient for any normal speaker crossover design. It doesn’t hurt to go bigger but probably isn’t necessary. Even the 100 V non-polar electrolytic caps are fine again for most normal applications.

Don’t quote me on this, but if I remember correctly, 100 W at eight ohms is roughly 28 volts. 

You are good with 250v caps even with your 150w amp. Where voltage becomes more of a concern are the capacitors inside the tube amp, which need to handle several hundreds of volts. That’s why I keep my fingers out of tube amps! 

tajanes

and guess I should have inquired as to best resistors and wattage as well, 25wZisters or 15w sandcasts, or ???

PWRRYD

This is an over simplified reply (because speakers and their crossovers are not simple resistive loads) but 150 Watts into an 8 ohm load is only 34.64 Vrms.  1250 Watts rms is 100 Vrms.  

P = (V x V) ÷ R

A much better way to predict power and/or voltage across a crossover component is to model the crossover and drivers in Xsim and set the voltage source to the appropriate output.  Then you can graph the voltage across any component versus frequency or power versus frenquency.  I rarely look at this for caps and coils but do check how much power a resistor may see.  Remember these calculations are for constant sinewave signals and actual music signals are far less demanding.

Some people associate a higher voltage rating of a capacitor with higher quality and better sound.  It requires a different construction (physically larger) which may or may not sound better/worse/same.

jr@mac

Be advised the resistors that use heat sinks are derated considerably when not mounted to a heat dissipating surface. They look cool, though. That being said, the industry adopted 10W for a reason - in most applications it will suffice. Double them up if you are concerned. 

dynamo

jr@mac said:

Be advised the resistors that use heat sinks are derated considerably when not mounted to a heat dissipating surface. They look cool, though. That being said, the industry adopted 10W for a reason - in most applications it will suffice. Double them up if you are concerned. 

That’s good to know, I always believed they dissipated their rated heat as is and it was made even better when screwed to a sink of sorts. Good info

Wolf

It usually states in the data sheets that the 25W are more like 5W without additional surface area, and the 50W are more like 10W. You'll notice they are approximately the same size panel mount to ceramic/sand-cast.

PWRRYD

And you had no clue about that until I enlightened you in Iowa after seeing you use them without heat sinks in your cap test/Attitudes boxes.

tajanes

thanks, so I'll probably go with the ceramic/sand-casts, and may consider a parallel set-up (noting need to double resistance value in this configuration)

Wolf

And you must have me confused with someone else, because there are no heatsink style resistors in the Attitudes cases.

PWRRYD said:

And you had no clue about that until I enlightened you in Iowa after seeing you use them without heat sinks in your cap test/Attitudes boxes.

PWRRYD

Must have been at a later date.

Wolf

Even so, I initially read it in the spec sheet. I don't believe it was you that told me.

PWRRYD

Lol, you're the guy that remembers every driver's part number from the last several decades but you can't remember "If you don't believe me check the data sheet".  Whatever 

jr@mac

Alright guys, settle down. 

PWRRYD

  

Ron_E

PWRRYD said:

"This is an over simplified reply (because speakers and their crossovers are not simple resistive loads) but 150 Watts into an 8 ohm load is only 34.64 Vrms.  1250 Watts rms is 100 Vrms.  

P = (V x V) ÷ R"

RMS voltage is multiplied by 2.828 to get peak to peak voltage which I assume is the voltage caps could see.

Ron

R-Carpenter

Not sure I've seen more than 20v between the terminals.

I think higher voltage rating is more of a sales gimmick, subconscious related to the performance. Higher voltage caps are also bigger.