Phono Preamp Tube Rush

I then extrapolated from that number to the result corresponding to the 12 ohm resistance of my cartridge, over a 20 kHz bandwidth:

@almarg Most LOMC cartridges have bandwidth well past 20KHz- most have no worries going to 40KHz.

if executed in the balanced domain you get a maximum of 6dB less noise

Actually, that's not true. With balanced amplification you get 6dB more signal gain, but you also get 6dB more differential noise gain. Overall SNR is the same.  

IME you get the same gain- not 6dB more. We might have to straighten out our terms here- it might be that we are talking about the same thing.

Most LOMC cartridges have bandwidth well past 20KHz- most have no worries going to 40KHz.

Hi Ralph,

Yes, I realize that. But what I was attempting to calculate was an approximate figure of the amount of Johnson noise (aka thermal noise) attributable to the cartridge’s resistance **that might be audible** while a record is not being played.

@hagtech, thanks for confirming that the ESR (equivalent series resistance) of a cartridge can be used for purposes of that calculation.

Best,
-- Al

Almarg and Hagtech, Do I conclude from the talk about Johnson noise that (1) MM cartridges with their inherently much higher internal resistance and much much greater inductance will be noisier than MC types?  (But of course the noise that accompanies the greater gain needed for a typical MC vs an MM would swamp Johnson noise in most cases.)  And (2) what about all those high value Rs often used in an RIAA network, again, IF one is obsessive about Johnson noise?
I've never measured the R across the secondary of a typical SUT but I would never have imagined it could be as high as 3900 ohms.  Surprising.  I am not an SUT kind of guy, either.

Do I conclude from the talk about Johnson noise that (1) MM cartridges with their inherently much higher internal resistance and much much greater inductance will be noisier than MC types?

What matters, of course, is signal-to-noise ratio, and of course the greater amount of Johnson noise that would be generated by the greater amount of resistance in a MM cartridge occurs in conjunction with a greater amount of signal.

And (2) what about all those high value Rs often used in an RIAA network, again, IF one is obsessive about Johnson noise?

I’m not especially familiar with the internal designs of phono stages, but I would expect that quality designs typically apply a significant amount of gain "ahead" of the RIAA equalization circuitry. Resulting in much better S/N ratio relative to Johnson noise generated by that circuitry than if the output of the cartridge was applied directly to it.

Also, keep in mind that Johnson noise that can be generated by a high value resistor may or may not matter depending on the impedances and configuration of the associated circuitry. For example, as Hagtech and the Sound Devices paper I linked to both indicated, shorting a high impedance input will essentially nullify the effects of Johnson noise that would otherwise be introduced by that high impedance.

Finally, of course, the RIAA network rolls off everything above 1 kHz to some degree, which encompasses 95% of the spectrum that is nominally audible.

In any event, the only reason I raised the issue of Johnson noise was in response to a question about whether any tube-based phono stages are quiet enough for hiss to be inaudible when one’s ear is placed against the speaker, while a record is not being played. And my response was to the effect that for typical combinations of system gains, sensitivities, and volume control settings theoretical limitations might come into play which could make that impossible, or close to it.

Best regards,
-- Al

If Johnson noise is a problem you can narrow the bandwidth of your cartridge to 100 Hz (stick chewing gum behind the cantilever I guess)
or dunk your cartridge in liquid nitrogen before you play a side:) 
Seriously, this is all academic because there is not a darn thing (reasonably) that you can do about it and I hate math.