Selecting a DC blocking capacitor

[WithTarragon]

I need to show my ignorance on this one. I am replacing the analog output section on a Behringer DCX crossover (opamps and cheap parts galore). This uses a AKM 4393 DAC (voltage) and each leg sits on about 2.5 VDC. It will be replaced by a DC blocking capacitor followed by a RC filter for anti-aliasing.

This will then feed a First Watt B1 buffer with an input load determined by the pot of 25K Ohm. This is the design but I am stuck on the how to best determine the best capacitor value for the DC blocking.

For instance, with a 0.68 uF cap, the
Fc would be 9.4 Hz (about an octave below the 20Hz that I am interested in)
Xc (at 20 Hz) would be 11.7 kOhm
RC time constant would be 0.017 seconds

While a a cap of 6.6 uF cap, the
Fc would now be 0.965 Hz (well away from the audio band)
Xc (at 20 Hz) would be 1.2 kOhm
RC time constant would be 0.165 seconds


I am struggling on how to prioritize these 3 results.
1) Is there less DC attenuation when the Fc is lower (in which case should I go with the smaller cap value)?
2) At 20 Hz (period of 0.05 seconds), do a go with the larger cap value, so the RC is greater than the period of the lowest frequency?
3) Do I go with the larger cap value, since the Xc is smaller?

In other words, what is a designer weighing when they pick the value of the coupling cap?

Thanks

 

[Iain McNeill]

1: the capacitor will always be open circuit to DC (unless it's bad) and so will always provide infinite DC attenuation regardless of capacitance value.

2&3: a single RC filter gives a 6dB/octave filter In reality, the filter is 3dB down at the 1/2piRC frequency and then morphs into 6dB/octave below that. To have negligible attenuation (<1dB) at 20Hz you need to select an Fc at least an octave below. Your 680nF cap should be fine.

 

[orpheus]

Question about DC blocking in a capacitor, related to the original post.

I have a modded ipod 4G, where I took the output from the wolfson DAC through a Vishay 1813 2.2uf cap for each channel (they can easily fit inside the ipod case with a flash drive mod). The Wolfson outputs 1.5 volts of DC. At turn on, I get a high DC offset, even through the capacitor, which slowly decreases over time. It is often at about .6 volts on each channel. I stupidly had my headphone amp turned all the way up and blew out the opamps. After rebuilding, I now I just wait a bit before I plug in after turn on, and make sure I have the amp off at turn off.

If a capacitor functioning correctly always blocks DC, are my caps malfunctioning? Is this normal behavior for this scenario? Any suggestions?

Thank you for your time. I apologize for the thread hijack, I thought this question is perhaps related. I can depart from your thread if you would like.

Aaron.

 

[Mooly]

You will see decreasing "DC" at the output of a cap depending on the impedances it works into. Your caps will be fine... you are not understanding how it works.

It's impossible to add any more without seeing the actual circuits involved.