An interesting paper - concept transferrable for photo preamps?

[NickKUK]

High-performance fully differential photodiode amplifier for miniature fiber-optic gyroscopes

A photo diode is going to be low current, so that got me thinking if the paper's approach would also be useable for moving coil amplification. You could go one step further with a fibre optic gyroscope stylus (but that would be silly, no?)

 

[Bonsai]

Look at mediatech’s design on this forum which uses opamps and low noise transistors - some circuits are in the last 10 or so pages of the MPP thread.

Diff amps offer superb suppression of mains induced hum, but there is at best a 6 dB thermal noise penalty.

 

[hweb]

Good luck finding a gyro that will work out to a few kHz. If you do, you'll see that it's broadband noise is awful. On the bright side we don't need response down to DC, so we can eliminate bias drift of the gyro via a high pass filter.

 

[spladski]

Where does the thermal noise penalty get introduced?

 

[aboos]

You have two amplifiers adding noise in a fully differential and balanced amplifier

 

[Bonsai]

Yes.

If you go single ended your get the lowest thermal noise.

 

[alayn91]

Hello,
It does already exist !
One exemple: Retro - Balanced RIAA Stage

 

[PRR]

....A photo diode is going to be low current, so that got me thinking if the paper's approach would also be useable for moving coil amplification....

While the output is "low current" (compared to what??) an MC cart has more available current than a MM cart.

And most of that paper wrestles with the dominant and relatively heavy (re: data frequencies) capacitance of a photo diode. Which is again more typical of our MM interface than MC.

 

[spladski]

What about the noise that becomes common mode in a diff amp?

 

[billshurv]

Look at mediatech’s design on this forum which uses opamps and low noise transistors - some circuits are in the last 10 or so pages of the MPP thread.

I have a set of those boards. Really should fire them up now the nights are drawing in.

Diff amps offer superb suppression of mains induced hum, but there is at best a 6 dB thermal noise penalty.

Personally I think the benefits outweight the losses*. There is of course the oddity of an inverting input INA where you don't have the 6dB noise penalty, but you open yourself up to a whole new world of problems.



*And at the end of the day, it's only a record so it's all just a bit of fun

 

[Mark Tillotson]

Diff amps offer superb suppression of mains induced hum, but there is at best a 6 dB thermal noise penalty.

Its 3dB not 6, and its not thermal noise as such but device input noise (voltage and current). Its because you have two identical but independent devices whose input noises are in series, so you have double the noise power which is 3dB extra.

 

[Bonsai]

Agree it’s 3 dB because the noise sources are uncorrelated and not 6 dB - my bad.

Thermal noise (aka Johnson or Johnson-Nyquist noise) is the same as En and In that you refer to and is proportional to absolute temperature, hence calling it ‘thermal noise’ which is used often in technical and academic papers on noise.

If you cooled the circuit to absolute zero there would be no noise.

Johnson& - Wikipedia