battery box crosstalk

[udovdh]

Hello,

I hope it's OK to ask here about how to lessen crosstalk in my batterbox.
It's the simple setup: 2 x capacitor and 2 x resistor to power the (2 wire) mics and over the battery a 10 uF electrolytic in parallel with a 100 nF film to help the battery and to avoid crosstalk since both mics are connected there.
But signal still goes from mic a over the powering resistor, via the other powering resistor to mic b. From my measurements crosstalk is wayy too high (should be down 80+ dB?).

Anybody got an idea about how to fix this? More capacitance over the battery? (better quality?)

Please post!

[dklein]

I believe that you want to increase the resistor values.  Not sure what you're using but how about trying 10k?

[udovdh]

I believe that you want to increase the resistor values.  Not sure what you're using but how about trying 10k?

Hehe...
I was at 22K, but lower is the spec of the manufacturer...
Any tricks I could use? (diodes?)

[dklein]

which mics?  are they getting enough juice with 22k? maybe that's killing the output level and consequently raising the crosstalk relative to the signal?

[udovdh]

which mics?  are they getting enough juice with 22k? maybe that's killing the output level and consequently raising the crosstalk relative to the signal?

Everything was fine at 22K. No distortion.
Stuff is even better (highs and such) at lower resistance, exactly like the manufacturer specifies.

[dklein]

Everything was fine at 22K. No distortion.
Stuff is even better (highs and such) at lower resistance, exactly like the manufacturer specifies.

you are using higher resistance 
I would think 22k would yield low signal levels.  I believe setting bias voltage correctly is a balancing act between dynamic range and output level.  Regardless, I don't know which capsules you're using but you seem quite sure that what you're doing is good so we'll just leave it at that. 

[udovdh]

So, in short, when implementing the manufacturer powering requirements I get better mic performance but higher crosstalk.
How do I get rid of it?
How bad is the idea of diodes in the lines towards the mic capsules? (bfeore of after the resistor)
Other solutions?

[udovdh]

Everything was fine at 22K. No distortion.
Stuff is even better (highs and such) at lower resistance, exactly like the manufacturer specifies.

you are using higher resistance 

Lower than 22K.

I would think 22k would yield low signal levels.  I believe setting bias voltage correctly is a balancing act between dynamic range and output level.  Regardless, I don't know which capsules you're using but you seem quite sure that what you're doing is good so we'll just leave it at that. 

Here lower resistor values decrease amplification.

[dklein]

you might want to pm poorlyconditioned on this board.  He's been playing around more than I have.

[poorlyconditioned]

Everything was fine at 22K. No distortion.
Stuff is even better (highs and such) at lower resistance, exactly like the manufacturer specifies.

you are using higher resistance 

Lower than 22K.

I would think 22k would yield low signal levels.  I believe setting bias voltage correctly is a balancing act between dynamic range and output level.  Regardless, I don't know which capsules you're using but you seem quite sure that what you're doing is good so we'll just leave it at that. 

Here lower resistor values decrease amplification.

Hey, I heard my name.

Anyway, you gotta say what capsules you're using.  Is it a standard batt. box (or my 3-wire version)?

I'm almost sure you've gotta go down to 10K.  Do that first, then let's talk...

  Richard

[udovdh]

Anyway, you gotta say what capsules you're using.  Is it a standard batt. box (or my 3-wire version)?

I'm almost sure you've gotta go down to 10K.  Do that first, then let's talk...

It's 2-wire Sennheiser capsules. Amplification decreases with lower resistance values. Sound improves slightly when going down to the manufacturer recommended value.

[Bertho]

Use the resistor values that you like and increase the bypass capacitor. 
I would not mess with diodes.  As an alternative, split the feed to each side by for example a 100 ohm resistor going to the battery (100 ohm in series with each 10k) and then bypass each 100 ohm resistor separately.  Your bypass will be more effective by it self and you will have dual filters going from one side to the other.  Watch out for ground loops so use single point grounding.
Bertho

[udovdh]

Use the resistor values that you like and increase the bypass capacitor. 
I would not mess with diodes.  As an alternative, split the feed to each side by for example a 100 ohm resistor going to the battery (100 ohm in series with each 10k) and then bypass each 100 ohm resistor separately.  Your bypass will be more effective by it self and you will have dual filters going from one side to the other.  Watch out for ground loops so use single point grounding.

I..e: 100 Ohm in series with the resistance taht is there, parallelled by a few uF?

[Karl]

A few thoughts and a few questions...

Are you saying that your crosstalk is -80dB?  Or you want it to be -80dB?

A diode would be bad--an audio signals waveforms have positive values as well as negative--a diode would get rid of the negative side of all waveforms.

I think that to get rid of the crosstalk completely you would prob have to electrically shield one channel from the other.  I don't know if that's possible when using a single 9V batt for two channels.  Anyways, to shield you have to enclose all components of one channel inside of a conductor that is grounded.

Are you sure your crosstalk is happening ONLY inside your batt box and not other places also along the signal path?

In a live recording situation, -80db of crosstalk will not hurt you.  You will have a hard time noticing that it's affecting your recording at all.

[udovdh]

A few thoughts and a few questions...

Are you saying that your crosstalk is -80dB?  Or you want it to be -80dB?

I want it at -80.

A diode would be bad--an audio signals waveforms have positive values as well as negative--a diode would get rid of the negative side of all waveforms.

Two diodes. One to each mic resistor.
Crosstalk signal would see both.

Are you sure your crosstalk is happening ONLY inside your batt box and not other places also along the signal path?

Quite sure it is the box.
I have quality cables, good recording equipment, good plugs and soldering.
I also see the problem when one mic is disconnected from the battery box.