single-ended to differential?

[theAnonymous1]

Originally posted by Cauhtemoc
I was obviously refering to figure 13.

My apologies, I assumed you were mistakingly referring to the Vcom buffer because I had already mentioned the OPA1632 input buffer in post #4.

[mako1138]

Originally posted by theAnonymous1
That would be -0.3V to 5.3V (assuming Vcc is 5V), not -0.3V to 0.3V.

I assumed that he would have no intervening circuitry, thus the signal would be centered at ground. Though I guess he could just use a coupling capacitor, since the inputs are biased to Vcc/2. There'd be a low-frequency cutoff in that case.

[theAnonymous1]

Originally posted by mako1138
I assumed that he would have no intervening circuitry, thus the signal would be centered at ground. Though I guess he could just use a coupling capacitor, since the inputs are biased to Vcc/2. There'd be a low-frequency cutoff in that case.

Another good reason for using the OPA1632 since it's outputs can be biased to Vcc/2 by connecting Vocm to Vcom of the ADC. There would be no need for coupling caps in this case and no associated low end cutoff.

[Cauhtemoc]

Originally posted by theAnonymous1
My apologies, I assumed you were mistakingly referring to the Vcom buffer because I had already mentioned the OPA1632 input buffer in post #4.

My appologies as well, I managed to miss post #4.

Anyway, to original poster, the OPA1632 was made for exactly this type of thing. It's the easiest, simplest and best solution you will find without going to the extremes of complexity.

[mako1138]

Originally posted by theAnonymous1


Another good reason for using the OPA1632 since it's outputs can be biased to Vcc/2 by connecting Vocm to Vcom of the ADC. There would be no need for coupling caps in this case and no associated low end cutoff.

Yeah, certainly the most straightforward solution is to follow the datasheet. I just wanted to advise the original poster of the consequences if he decided to go ahead and ground an input, per his original proposal.

[gmarsh]

Little bit of caution if you're using figure 13.

If the source impedance of the + and - inputs aren't the same, ie. if the - input is grounded (0 ohms) and the + input has a non-zero source impedance, you'll get a DC offset on the output.

Poke an op-amp buffer on the + input to get around this. A single OPA2134 or similar should suffice for both channels.

[gentlevoice]

Thank you all for your suggestions and inputs ..... It clarified issues that I have to pay attention

To the end of this A/D project that I'm doing - any of you who knows how I can transfer the 2 channels of 24 bits/192 khz via USB to my laptop?

I'm looking for a straightforward chip solution that accepts the output from the PCM 4202 and converts it to USB in a way so that the laptop recognizes that a new USB device is present (with the characteristics necessary for audiosignal transmission).

My aim is to not have to do any programming ....

Any ideas?

Thanks Jesper