Question about bypass capacitor values in TL XO

On3s&Z3r0s

Can anyone explain how the 5.8uF bypass cap works in the TL mods that involve replacing the 194 tweeters with the 198? From reading various threads (and especially this one where Matt Polk is quoted in #52) it sounds like the cap is boosting high frequencies to compensate for roll-off in the tweeter. My question is what specifically happens if the value deviates from the 5.8uF significantly higher or lower.

The reason I ask is bought some caps to do a TL since Sonic Craft is having a sale. Apparently, I wasn't the first one to think of doing that because I was told that 5.8uF customs had already been backordered and 5.6uF was sold out as well. The guy I spoke to sold me on 6.0uF caps.

I know people have used 5.6uF caps but haven't ever heard of anyone substituting 6.0uF, so I'm wondering what the effect is likely to be.

ben62670

The 5.8uf in parallel with the 2.7r changes the slope. So yes it does boost the tweeters as the frequencies go up. 6uf should be fine. I am surprised that they don't have any custom value caps???

On3s&Z3r0s

ben62670 wrote: »

The 5.8uf in parallel with the 2.7r changes the slope. So yes it does boost the tweeters as the frequencies go up. 6uf should be fine. I am surprised that they don't have any custom value caps???

The guy I spoke to (didn't get his name) said they were already having some 5.8 made but that they were spoken for and that they were out of the 5.6 with a lead time of 2-5 weeks. Good to hear that the 6.0 should work, thanks! I'll check the capacitance when I get em and hope they're not off of 5.8 by more than 5-10%.

I am still a little curious generally about how the value of the cap figures in. Does it follow the normal rule that the bigger the value the narrower the range of frequencies you are filtering out? So in this case a marginally larger range of frequencies would get passed through/boosted? It seems like this might be doing something different than a normal cap in a 2nd order XO, so that reasoning might not apply.

I really need to find a book on this stuff.

ben62670

The cap sets the frequency where the slope begins.

On3s&Z3r0s

ben62670 wrote: »

The cap sets the frequency where the slope begins.

LOL... Ben, are you messing with me or am I just that dense... please don't answer that question. But please do answer this one: so does a bigger value in microfarads mean that the slope begins at a higher frequency or a lower one?

As a somewhat funny aside, I'm not any better at asking questions over the phone than I am typing into a text box. When I was ordering the caps and asking about the availability of the 5.8/5.6, the guy told me he didn't have any but he could look for some low 6's.

I said, "What's the wait?"

He said, "They're all about the same, maybe a couple ounces."

ben62670

The bigger the cap the lower the slope will start.
LOL funny.

On3s&Z3r0s

Cool. Thanks!

FTGV

On3s&Z3r0s wrote: »

I am still a little curious generally about how the value of the cap figures in. Does it follow the normal rule that the bigger the value the narrower the range of frequencies you are filtering out? So in this case a marginally larger range of frequencies would get passed through/boosted? It seems like this might be doing something different than a normal cap in a 2nd order XO, so that reasoning might not apply.

.

The 5.8uf and 2.7r form a contour network.The resistor by itself will attenuate all frequencies equally.With the addition of a paralleled cap the amount of attenuation at higher frequencies can be reduced.(Think of it as basically the cap gradually reducing the resistors ability to attenuate as frequency rises, the actual range determined by the caps value.)The result is the very high end of the tweeter recieves less attenuation. If that makes sense.

jcandy

On3s&Z3r0s wrote: »

The guy I spoke to (didn't get his name) said they were already having some 5.8 made but that they were spoken for and that they were out of the 5.6 with a lead time of 2-5 weeks. Good to hear that the 6.0 should work, thanks! I'll check the capacitance when I get em and hope they're not off of 5.8 by more than 5-10%.

I am still a little curious generally about how the value of the cap figures in. Does it follow the normal rule that the bigger the value the narrower the range of frequencies you are filtering out? So in this case a marginally larger range of frequencies would get passed through/boosted? It seems like this might be doing something different than a normal cap in a 2nd order XO, so that reasoning might not apply.

I really need to find a book on this stuff.

You can use a 4.3 and 1.5 in parallel:
http://meniscusaudio.com/43uf-400v-p-308.html
http://meniscusaudio.com/15uf-400v-p-301.html

Book:
http://www.parts-express.com/pe/showdetl.cfm?partnumber=500-035&source=googleps

Here are curves of the magnitude of the impedance for an RC contour element with R=2.7 Ohms, and C varied between 5.6 and 6.0uF. Since the impedance drops as frequency increases, the voltage supplied to the tweeter will increase as frequency increases. Therefore, the figure implies that higher capacitances will give a "brighter" presentation.