MC preamp input stage (BJT based) questions

[jason12]

Hi,

I am about to build an MC phono preamp with BJTs in the input stage. The circuit will be similar to this one:
BC550 BC560 Very low noise RIAA

The design parameters are:
1. MC cartridge resistance 10 Ohm;
2. No AC coupling in the signal path;
3. Noise as low as possible.
4. Power supply +/- 15VDC

Here are the questions I have:

1. Which BJTs are better for low noise performance - BC550C/BC560C or 2SA790GR/2SA2240GR ? By comparing the data-sheets I am leaning towards BC550/BC560. It seems that at low source resistance they have lower noise figure (2db) than 2SA970/2SA2240 (6db).

2. What collector current would you recommend (per transistor in case of paralleling).

Thanks

[Joachim Gerhard]

This is not a super low noise circuit at all. Look at R18 for example.

[jason12]

Actually the circuit will look more like the one by Pigletsdad:

Discrete phono preamp designs?

with paralleled BJTs, CCS and class A push-pull output.

What is wrong with R18? Will it add to the noise? I was under impression that the resistors in emitter and base would do most damage.

[Joachim Gerhard]

R18 is effectively in series with the cartridge. See a detailed analysis of noise in MC stages on syn08s webpage or read my rather long thread MPP for a detailed discussion on
all kinds of phonostages. Even better buy the rather expensive book "The Sound Of Silence" by Burkhard Vogel. It is imposible to give a quick rundown on all the issues involved. Piglet Dads circuit is easy to make and low noise, so this is a good start but in no way the end of the road.

[bocka]

The BC550/560 aren't a good choice for low-impedance / low-noise applications. Unfortunately the noise vc ic data is ripped off of the datasheets. For a source impedance of 500 ohms you can expect a noise figure of about 3dB @ ic = 300uA. This seems to be the best what you can get from the single BC550/560. 2SC2240 is better in this way. You can obtain a noise figure of about 6 dB @ ic = 4ma for 10 ohms source impedance.

Joachim is right. R18 is in series with the cartrige an adds some additional noise.You should lower this resistor to about 5 ohms and recalculate the riaa. In this way I'd expect a noise figure of about 4 db for a single complementary diff input stage.

Zetex/Diodes has also a nice application note about low-noise design:

http://www.diodes.com/_files/design_...zetex/dn11.pdf

[rayfutrell]

You may still be able to download the sound of silence here re "The sound of silence"

Ray