Thanks again Juma for your cascoded version.
I still haven't built it yet.
After some time to ponder all the possibilities. Do you feel this would be the best version?
Or is there anything else you might do differently?
Thanks again.
There is no best version - there are just contemporary solutions. If it were otherwise we would all listen to that one, the "best version". My "best" and your "best" would probably differ.
There are thousands of ways you could build a voltage gain stage - you should build this one or some other, see if you like it or not, learn something from that experience and go on from there...
Maybe I should rephrase the question. What I am trying to achieve is zero feedback, however I want to make full use of the output from my cd player (ie 2 V rms), hence the need for cascoding. The thing that has stopped me from building this is the +40V and -3.5V rails.
Is it possible to use zeners to get the -3.5V? What would the power supply look like?
That negative voltage is essential to this design - it gives us the high Id and low gain (Rd/Rs ratio) at the same, using only source degeneration.
The quality of negative supply is very important - every nastyness from it will be amplified by the gain stage. I recommend salas shunt reg, I used it here:
http://www.diyaudio.com/forums/pass-labs/146310-bf862-preamp.html
see post #28 in that thread
The quality of negative supply is very important - every nastyness from it will be amplified by the gain stage. I recommend salas shunt reg, I used it here:
http://www.diyaudio.com/forums/pass-labs/146310-bf862-preamp.html
see post #28 in that thread
Vdi_nenna,
If you don't want to mess up with the negative power supply, then you may try this as well (see the pic). I bet this schematics is mentioned here as well, but since these threads have grown a lot, I am lazy to point to the exact source, except that it comes from the serbian Diy audio forum DIYAudio.rs , and Juma must know more about it...
p.s. My comment was related to thanh1973, not Vdi_nenna. Sorry for mixing up. Trying to work and cheat an employer by having one eye at diyadio forum...
If you don't want to mess up with the negative power supply, then you may try this as well (see the pic). I bet this schematics is mentioned here as well, but since these threads have grown a lot, I am lazy to point to the exact source, except that it comes from the serbian Diy audio forum DIYAudio.rs , and Juma must know more about it...
p.s. My comment was related to thanh1973, not Vdi_nenna. Sorry for mixing up. Trying to work and cheat an employer by having one eye at diyadio forum...
Attachments
Last edited:
... and Juma must know more about it...
Yes, I made that one more than a year ago and it sounded really fine with F5.
Notice that J310 (jes it's Vishay/ON Semi/Fairchild j310 exactly, not a 2sj...) is quite a different JFET from 2sk170 (large Vpinch_off) and it demands different application.
With R5=220R Id will be set at about 10mA (that's where we want it) and R3 will set the gain and output swing - you may try values up to 1k2.
Of course, power supply should be cleanest possible (shunt reg. recomended) because this preamp has very, very low PSRR.
With PS voltage of 24V you may expect about 6Vpeak of undistorted signal at the output.
You can skip it or you can put the cap multiplier before the shunt reg - it's cheap and easy to try how will it sound, especially if you have high enough voltage in the PS (30 to 40V).
Zout s determined mainly by R3, so if you want to lower the Zout to about 50R, adding a buffer on the output will enable you to drive low Z loads (5k and higher). It would look like this (I prefer BF862 instead of 2sk170):
