[porthillsbomber]

Hi guys,
 
I'm putting together a cmoy and diyMod and have a couple of questions regarding the dc blocking caps. Looking at the two options:
 
A. You have the dc blocking caps inside the ipod or somewhere before the volume pot and remove the ones from the cmoy.
 
B. You do not put any dc blocking caps in the ipod and leave the ones in the cmoy.
 
As I understand it this is what would happen in each case.
 
A. Your cap has to be ~10x as large as in case B (for same corner frequency) because the load is effectively the 10k pot.
 
B. You cap is smaller because the load is the 100k R2 but there will be a dc voltage on the volume pot wasting the dac's power.
 
Is this correct? What is the best option?
 
Cheers,
James

 

[Avro_Arrow]

Cap in the CMoy is the better option.
You will not always be plugging your CMoy into the same iPod.
The 10k pot is not going to waste hardly any power (if there is indeed a DC offset preset).

 

[amb]

Think of it this way, if your diyMod has no caps, it will output a large DC offset which, if you or someone plug it into an amp with no input caps, could cause headphone damage.  I believe in fixing the problem at its source, so put the caps in the diyMod.  Even if you disagree with my opinion and choose to put them in the cmoy, they should be before the pot.  The problem is not with the pot "wasting power" (which it does, only a little), but DC on the pot will make it sound scratchy when you turn the knob.

 

[porthillsbomber]

Ok, I'll probably do that then. I don't intend to use it with anything else as I've got another amp (a mini^3 thanks amb!), I actually want to build them into 1 unit.

 

[qusp]

if you want to build them into the one unit, then why the question? isnt it effectively the same thing? ie cap at the output if the ipod (last component in the ipod) in the same enclosure is the same as the first cap before the volume pot..isnt it?? or am I missing something here?

 

[akgfan]

How is optimum capacitor for say 10k or 100k pot? Off course, if you don't know output cap on your source and you want it before pot.

 

[nikongod]

 
Quote:

How is optimum capacitor for say 10k or 100k pot? Off course, if you don't know output cap on your source and you want it before pot.

You calculate it the same way as for a cap after the pot.

 

[akgfan]

OK. But if I can put there say 0,1u to 1u, which is better? Less or more? The more I read about hifi mambo jumbo stuff, I'm more confused.

 

[qusp]

use the lowest you can get away with plus a bit more, no mumbo jumbo here, its the same for any component

 

[porthillsbomber]

Quote:

if you want to build them into the one unit, then why the question? isnt it effectively the same thing? ie cap at the output if the ipod (last component in the ipod) in the same enclosure is the same as the first cap before the volume pot..isnt it?? or am I missing something here?

No, here's a poorly drawn schematic of the two cases.
 

 So as I see it if you want a corner frequency of say 2Hz and the pot is 10k and R2 is 100k then for case A you need ~8uF and for case B you need 0.8uF. What would be best if you were limited by physical size, A with an electrolytic or B with film?

 

[digger945]

In the "B" circuit, wouldn't the pot vary the resistance before the cap, creating a variable RC timing circuit with the time constant going up as volume goes down?

 

[porthillsbomber]

Quote:

In the "B" circuit, wouldn't the pot vary the resistance before the cap, creating a variable RC timing circuit with the time constant going up as volume goes down?

Not sure that's why I asked but to me it would seem with B you still have a plain old RC circuit but the input signal is attenuated. With A though you effectively have 2 resistors in your RC circuit, R2 and the bottom half of the pot going to ground and the top half of the pot in series with the cap.
 
Note I would have A if I had a diyMod with caps hooked up to say a mini^3 and I would have B with a diyMod without caps hooked up to a standard cmoy.

 

[qusp]

well my point was, I would wire it as A regardless, if they are joined together, I would simply have the cap in the joining wire to the pot and lose the input cap. wired as B you would get the output DC from the ipod effecting the pot

 

[porthillsbomber]

Ok looks like I'll go with A then. Why though is the cmoy B? Is it because the cap is only there as a precaution and B allows it to be smaller?

 

[nikongod]

 
Quote:

In the "B" circuit, wouldn't the pot vary the resistance before the cap, creating a variable RC timing circuit with the time constant going up as volume goes down?

Tricky...
The answer is that it does, but if you follow the rule of thumb that the "bleeder"/input resistor is 10*the nominal value of the pot the error is the same as using 5% caps. If you need to hit a very specific target frequency circuit B is the wrong way to do it but nobody really cares whether the -3db point is 2hz or 2.1hz in a DC blocker.
 
Quote:

Ok looks like I'll go with A then. Why though is the cmoy B? Is it because the cap is only there as a precaution and B allows it to be smaller?

Option A is the only way to do it if your source has DC offset on its output, and less importantly for most people makes the use of bipolar input op amps tricky. Option B lets you use smaller value caps which are easier to find in better quality and physical package sizes, less importantly for most: it makes using chips with bipolar inputs a bunch easier.
 
If your source has DC offset on its output thats a funny joke. On you.* 
 
*Excludes a few DAC's and amps designed to work together as a system and sort out DC offset in slightly unconventional ways in the endless pursuit of better audio.