Mpp

[Joachim Gerhard]

Michael, this may work somewhat better.

[Joachim Gerhard]

Problem here is that this circuit can only swing 1,54V. One possibility is to raise the gain, another is to double or quadruple the Fets. Also a BJT input comes to mind.

[Joachim Gerhard]

This is the BJT version, although it does not have a very high input impedance.

[Audiofanatic]

Hi Joachim,


I'm speechless, WOW !
I love stroomspiegels, (Current Mirrors) I don't know what to say, it's so great to see you posting these circuits.
The only thing that comes to mind is, why didn't I think of this before??

Well done Joachim, I'm very happy with this and will let you know how it performs when it's ready.

Danke

atb,

Audiofanatic

[Joachim Gerhard]

Jose`, the line stage with current mirror is not finished yet. The problem is that for a certain amount of gain we need a certain loading resistor. If the output impedance of the mirror is high the resistor value is low or we end up with too much gain, say 220 Ohm in my example is about right. We have only the current that flows in the input stage. Take the Starless input stage for example. We have 2 2SK170BL on Idss in there. That is 16mA. 16mA into 220 Ohm is 3.52V. That is just enough for a line stage but i would prefer if we are able to swing plus-minus 12V that is over the Fets to get a better dynamic range. One way is to raise the current in the mirror but that is expensive because we need a lot of 2SK170BL. After thinking hard and long ( this is for you Michael ! ), a better way would be to lower the output impedance of the mirror. That way we can raise the loading resistor say to 1500 Ohm for full swing. To lower the output resistance we can lower the differential feedback resistors ( now 50 kOhm ) or we can go back to simpler mirrors like the Widlar or three transistor mirror. Anyway, we will solve that problem When you ask me how i came to the idea to use mirrors i think about that since over 20 years and in the beginning my friends did jokes about me : " That does not work, i tried it" and so it goes. Now you say : " Why didn´t i think about this before". "The times they are a changing " Bob Dylon and later Willy Brand.

[MiiB]

I agree that there must be sufficient voltage headroom, 4-5 times the signal swing would be preferred.. Question is do we actually need gain in a line stage....??
What is the role of a pre-amp... switching and attenuation...and impedance matching...??

[Joachim Gerhard]

I think we need a little gain, otherwise a buffer would do and i am already very pleased with my Buffalo Bill modified diamond. Yes, it is impedance matching and driving a potentially high capacitance cable. Curiously the Forsell sounds better with the Buffalo Bill in the chain and it has already a very elaborate output stage. With a normal CD player with a wimpy analog output stage the effect is even more drastic. 6dB is enough though in most cases. Another issue is that i loose some gain in my balance and volume control.
When i put in 6V, 10V comes out so i would like to compensate for that.
I know that a transimpedance line stage is not easy to design. Therefore i presented the alternative circuits with the Hawksford cascode.
The picture made my day.

[MiiB]

Nice tonearm....

[jackinnj]

Joachim -- when do you sleep?

[Audiofanatic]

Originally Posted by jackinnj

Joachim -- when do you sleep?

Jack,

It's simple, my friend Joachim sleeps when we don't see him posting on the forum.
I already ask him this a few times and his answer was, (in short) a few hours a day spread over 24 hours.

We are a bit alike Joachim and I

Regards,

Audiofanatic