[Madgravity34]

I need to purchase the last part of a project I am working on. I need to know if the value of output coupling caps are really that important when used in a portable setup. I was wondering if there would be a difference between using .47uf or 47uf caps? Can anyone help me out here?

 

[balou]

Yes, thats quite an important question, there's a difference

First of all, do you really need output coupling caps? if your dc offset voltage is low enough, you don't need them.

Second, values over 1uF usually means electrolytic caps, which are considered to be a rather bad thing in your signal path. using a 0.47uF or a 1uF metalized polyester or polypropylene cap (polypropylenes are bigger but better, dunno if you have a portable project), you have a better sound quality.

but, the third thing to consider: the output cap creates a high-pass filter. um... somebody else should answer that. I think when the output impedance together with the headphone reach a certain value, you really might need a big capacitor.

edit: here you can calculate the cutoff frequency. but you need to know R, too
http://www.sengpielaudio.com/calculator-RCpad.htm

 

[threepointone]

R is anywhere from high-20ish (IEMs) to 600 (HD650s). In general, this means somewhere on the order of probably 470uF to be really safe. BTW, we're not talking about a subtle difference in sound here, you're talking about losing everything below 150Hz with 47uF and everything below 15kHz with .47uF (you wouldn't here anything, essentially)

As balou mentioned, you should really consider whether or not your design needs output coupling caps first. If it does, you might also consider modifying it so that you don't need to use output coupling caps.

Also, remember there's a difference between output caps and input caps--input caps only need to be about in the .22-.47uF range for most designs.

 

[balou]

isn't there a way to heighten the output impedance? I thought there were other factors than just the impedance of the headphones in this calculation. finding an output cap for a 32 ohm headphone would indeed be very bad.

 

[mb3k]

Code:

[left]Headphone Cap Lowest Impedance (uF) Frequency (Hz) 300R 1000 0.5 300R 470 1.1 300R 220 2.4 300R 100 5.3 120R 1000 1.3 120R 470 2.8 120R 220 6.0 120R 100 13.3 75R 1000 2.1 75R 470 4.5 75R 220 9.6 75R 100 21.2 32R 1000 5.0 32R 470 10.6 32R 220 22.6 32R 100 49.7[/left]

 

[Madgravity34]

Hey guys, thanks for all the replies! I should have done this in the first post, but I wanted to give you some more details on this "project". I wanted to see if I can bypass the DAC on my 5th gen ipod on my own. What I need help with is deciding what size caps to use. I will be probably using black gates or some other high-end cap. In this case, would .47uF caps work? I plan on using this with a hornet and was wondering if these caps are even needed. Do you think the hornet can handle the standard offset?

 

[balou]

If you are not directly driving headphones with it, but a headphone amp, 0.47uF should be sufficient under most circumstances. Go for 1uF if you don't want to loose the tiniest bit of bass. And I think that they are really needed... if unsure, you could just do a reading with your multimeter. If the dc voltage is over 5mV or so, you probably need output caps.

I haven't heard blackgates myself, but I must say that I'm a bit skeptical about them. Electrolytics in the signal path? I'd rather go for metalized polyester, they also have the advantage of being readily available at most distributors.

Please post pics if you open up your ipod, I'd be very interested to see the innards. I thought about bypassing the DAC too, but I'm a bit afraid of it.