Simple phono stage (MC)

[Onvinyl]

Hi,
in my neverending quest for a decent phono stage I made a sketch that was surprisingly serious. More or less a side try, I might now examine it a bit further.
I started much more complicated and scaled it down a few times.
Here's the first stage:

 

[Onvinyl]

It seriously fought my main prototype, but then faded away due to an seemingly overheated voltage reg Q. Doesn't look too dangerous, though.
(above Q2 is the wrong breed, grrr)

Well, nothing new under the sun, just a bit from here and there put together.

Second stage:

 

[scott wurcer]

I ran the thought experiment the other day of punting the source resistors and letting the currents share as they may. The results were interesting in that a rather wide selection of Idss's had little impact on the net gm and noise of the paralleled lot.

 

[Onvinyl]

Interesting! According to the sim, the potentially impressive distortion numbers are highly dependent of input device matching, escacially the higher order THD.
My hope was, that the source res. would do something for this.

I now simmed different models and different source resistances and that reveals that the numbers don't benefit that much from the resistors.

Do you think its safe to try without source resistors with an matched oktett with IDSS variations between 7.5 - 8.2 mA?

Pity that I don't have a distortion analyzer...

Rüdiger

 

[scott wurcer]

That's pretty matched in my book. I was investigatiing paralleling large numbers of devices (32). I wanted to buy a tube of 50 and just throw out the "flyers". I would think the N/P differences would dominate the distortion.

 

[Onvinyl]

I tried it without source resistors. I doesn't work, because the bias isn't stable, and I can't establish stable DC-conditions. As always, there could be other mistakes, but none easily spotable.
Rüdiger