Frans, I have a comment for you.
If you remove the servo at the input it still works jut fine as you are correcting both halves separately and the the high and low side at the same time.
Simulation proves this theory.
Correct me if I am wrong but you would use the servo at the input in case any input offset would be induced.
However cartridge won't induce any do offset thus it looks like we wouldn't need that at least.
Let me know if what I am saying makes sense to you or not.
If you remove the servo at the input it still works jut fine as you are correcting both halves separately and the the high and low side at the same time.
Simulation proves this theory.
Correct me if I am wrong but you would use the servo at the input in case any input offset would be induced.
However cartridge won't induce any do offset thus it looks like we wouldn't need that at least.
Let me know if what I am saying makes sense to you or not.
In my building with monolithics I don't even have 1mV across input thus no need for servo there.
The servo at the input ONLY removes the input differential (if any) and thus makes sure that no DC can flow in the cartridge. It has nothing to do with the output offset(s).
Still the servo is there to prevent any input offset what so ever 🙂 It can always be build and then we can test what happened when switched of/on and also we can see what the sonic influence is of such a servo.
That's it, if you do worry about those things 🙂 then there is no harm in having the servo on board (actually it will help take away your concerns). And there is no 'servo-must-be-stuffed-police', so if you do not like to have the servo then just don't stuff the thing to the board. Only this reason should be good for having the input-differential-compensating-servo on board 🙂
But as long as the BJT's are there ... (I need to quit defending this anti-differential-input-to-cartridge-currents-and-voltages-servo 🙂).
I fully quote that. Without his contribute I would have never been able to finalize my circuit.
Thanks a lot Frans!
Now commenting on the input differential offset, I can again re-itearate that there is no need for that.
If BJTs are matched or monolithics are used, the input differential offset is extremely low which shouldn't constitute a problem.
I will test the circuit this week and start making some measurements on it.
Trrue.
Unfortunately I have tried that and however simulator tells me that you can get the 61dB gain mark, it is pretty far away from it.
Maybe the models I am using are not so realistic, dunno!!
I would have loved to use J74/K170, but I gave up on them for high gain circuits.
Beside, and I don't know if this was realistic either, but noise floor is significantly higher than with BJTs.
- Status
- Not open for further replies.