Mr Teal,
The circuit you suggest sound very sound (!) as per gut feeling to me. I guess if one really crunch the numbers maybe slightly different gain settings might result in somewhat lower noise. Maybe a little more gain in the OPA2134 stage like 10-15 dB (3-6 times) and thus less in the LM4780. (Note that I have not used the LM4780 myself and thus have none of the parameters "in my head").
But it might be a distortion tradeoff though. While a lower gain setting of the power amp chip generally decreases distortion at high frequencies due to "more available bandwidth" that might not be the case here since you run the chip in NI configuration. Thus CMRR will also affect distortion, and if the LM4780 has poor CMRR (I don't know if it has) a larger signal on its input will increase distortion. Maybe some who has done measurements (or are they in the datasheet?) can chime in here.
But from a noise perspective you are on the right track. I can't remember the output driving capability of the 2310 but maybe you can get along without the OPA2134 altogether and perhaps reduce noise further.
But then, it is nice to have it there to take some gain off the power amp. It won't make a hell lot of noise anyway so the best bet is probably to leave it there. You can always experiment a little. 😀
Franz,
/Magnus
The circuit you suggest sound very sound (!) as per gut feeling to me. I guess if one really crunch the numbers maybe slightly different gain settings might result in somewhat lower noise. Maybe a little more gain in the OPA2134 stage like 10-15 dB (3-6 times) and thus less in the LM4780. (Note that I have not used the LM4780 myself and thus have none of the parameters "in my head").
But it might be a distortion tradeoff though. While a lower gain setting of the power amp chip generally decreases distortion at high frequencies due to "more available bandwidth" that might not be the case here since you run the chip in NI configuration. Thus CMRR will also affect distortion, and if the LM4780 has poor CMRR (I don't know if it has) a larger signal on its input will increase distortion. Maybe some who has done measurements (or are they in the datasheet?) can chime in here.
But from a noise perspective you are on the right track. I can't remember the output driving capability of the 2310 but maybe you can get along without the OPA2134 altogether and perhaps reduce noise further.
But then, it is nice to have it there to take some gain off the power amp. It won't make a hell lot of noise anyway so the best bet is probably to leave it there. You can always experiment a little. 😀
Franz,
If that was the case I never noticed. Anyway my favourite German expression is probably the "vielleicht sind wir auf dem falschen Dampfer". Sounds kinda funny when translated to Swedish... 😉
/Magnus