Coupling cap before or after pot?

[sharpi31]

My amp has a 50Kohm attenuator before a 100Kohm load setting resistor.

Does my input coupling cap have to sit before the attenuator or can I put it between the attenuator and the 100Kohm load?

As I'm using a stepped attenuator rather than a pot I'm not so worried about small amounts of DC on the volume control.

Placing the cap after the attenuator would allow a smaller value to be used, for the same bass roll-off.

Any thoughts?

 

[Dan2]

as far as i know the input ac coupling cap is used so that dc does not get amplified, and create a large offset. so as long as the cap is present before the first amplifier stage it should be fine.

 

[sharpi31]

Assuming that there is no problem exposing the attenuator to small amounts of DC, and that a coupling cap after the attenuator achieves the goal of screening the amp from DC, which position is likely to sound better?

A cap before the attenuator passes a full line level signal, while a cap after the attenuator passes a (typically) much smaller attenuated signal.

In which position would you expect the cap to add more of a sonic signature?

Any change due to differences in signal level would have to be considered alongside changes due to the differences in cap size (the cap after the attenuator could be half the size of one before the attenuator).

As I write I feel that the only answer is to experiment! I'd still be grateful if any one has relevant experience they'd like to share.

Thanks.

 

[Frank Berry]

If there is any DC present before the switched attenuator, you will probably hear a "pop" when you change volume levels.
For that reason, I suggest that you put the coupling capacitor before the attenuator.

 

[DigitalJunkie]

I'm with Frank.

 

[Dan2]

Frank that does make sense

Any change due to differences in signal level would have to be considered alongside changes due to the differences in cap size (the cap after the attenuator could be half the size of one before the attenuator).

what do you mean by cap size?? the voltage or uF value?


a dc blocking cap is basically a high pass filter (think of DC as 0Hz), so a uF value that is too small will block low frequencies. i don't see how the value will change before or after the attenuator.

im not too sure about calculating the cap value, i know the amp impedence is in the formula, but wouldn't you calculate the cap value for worst case senario (full volume, lowest impedance)

please guys correct me if im wrong

 

[sharpi31]

If the cap sits before the attenuator it sees a 33Kohm impedance (50Kohm pot in parallel with 100Kohm load). If the cap sits after the pot it only sees the 100Kohm load. For the same bass roll-off point, the cap must be 3 times larger if positioned before the attenuator.

So a 3.3uF cap positioned before the attenuator (seeing 33Kohm) will have the same roll-off point as a 1.1uF cap after the attenuator (seeing 100Kohm).

Aside from the problem of hearing pops (which I agree is worth avoiding) which position will result in the least cap-induced distortion?

I'm wondering how differences in signal level, cap ESR, cap ESL, noise pick-up due to physical size will effect the sound.

If the answer is 'not enough to worry about' then that's fine with me (I'll stick the cap before the attenuator and live without pops!).

 

[Dan2]

oh its in parallel !!! my bad i was thinking of the attenuator being in series.

well if the caps aren't too pricey try both ways

 

[analog_sa]

Aside from the problem of hearing pops (which I agree is worth avoiding)

It's not just pops. Many a good pot have seen an early demise due to bad design allowing dc through the wiper.

 

[AndrewT]

what if you fit a source component that already has a DC blocking cap fitted to it's outputs?

 

[gootee]

what if you fit a source component that already has a DC blocking cap fitted to it's outputs?

In that case, if the input cap is before the step attenuator, you could end up with two caps in series, reducing the total apparent capacitance to (C1C2)/(C1+C2), giving a high-pass input filter (i.e. series capacitance and shunt attenuator resistance) response that has a higher -3dB frequency, resulting in the bass rolling off at a higher-than-anticipated frequency, possibly degrading the low-end performance of the system.

This potential problem could probably be avoided by using a large-enough capacitance value.

 

[AndrewT]

I adopted the thought of using one of the series caps @ 10uF PP (in the pre-amp output) and then that allows a smaller value cap to determine the HP filter.