If you're talking about a shunt regulator to supply the B+, certainly. This circuit doesn't really care much (to use a pathetic fallacy) about the power supply as long as its 250V or so and reasonably clean. If it were me, I'd save the fancy shunt supply for a circuit that needs it and use a very simple reg here, but a fancy one won't hurt.
I'll revive my earlier post about theory of operation.
I'll revive my earlier post about theory of operation.
It's pretty simple. The CCS act as plate and cathode loads for the voltage amp and cathode follower, respectively. The LED biases the first stage at 2V cathode to grid, while holding the cathode at AC ground. At a constant current, the output of the first stage is mu times the AC component of Vgk (which is the input voltage). Power supply rejection is high because of the voltage divider effect of the impedance of the CCS (many thousands of megohms) and the plate resistance of the ECC88 (a few thousand ohms).
The second stage cathode follower is direct coupled to the first stage voltage amp. The grid is at the plate voltage of the first stage- in order for the tube to conduct the current forced by its CCS, the cathode voltage will be about 2 volts above that and "follow" the AC signal.
Gain of the circuit will be close to the mu of the input tube (30-ish, in this case). That's pretty high for a line amp, but you said you wanted it high. Output impedance will be 1/gm of the second stage, plus the series cathode stopper resistor used to discourage oscillation.
The second stage cathode follower is direct coupled to the first stage voltage amp. The grid is at the plate voltage of the first stage- in order for the tube to conduct the current forced by its CCS, the cathode voltage will be about 2 volts above that and "follow" the AC signal.
Gain of the circuit will be close to the mu of the input tube (30-ish, in this case). That's pretty high for a line amp, but you said you wanted it high. Output impedance will be 1/gm of the second stage, plus the series cathode stopper resistor used to discourage oscillation.
You can use PSUDII to determine the effect of an RC filter- the circuit draws 20mA constant current per channel. Or you can use a simple reg like the Maida or the two transistor regulator I used in my phono stage. Or you can use the Salas high performance regulator; it's total overkill for this circuit, but if you already have it up and running, why not?
attaching the + end of the CCS to a power supply (24V or more) and the – end to a 100R dummy load resistor connected to ground. The 20mA CCS should be adjusted to get 2V across the dummy load, and (naturally) the 10mA CCS adjusted to get 1V
example for minus side, the same for positive side with voltage positive
example for minus side, the same for positive side with voltage positive
Attachments
That one appears to have a series resistor in it.
A9A Chicago Miniature Lamps
This is a better choice and 90% cheaper.
What happened to the CCS plate load for the first stage? Seriously, build it the way it was drawn- you're seemingly making random changes which will all guarantee non-operation.
What in the world does that minus voltage do (other than prevent the circuit from working and probably blowing the MOSFETs)? You don't need to do things like this. Just build the simple circuit. Once you have it up and running, then you can start making random changes to see what happens- it could be an educational experience.
To adjust CCS current (and it's totally noncritical, you don't need to sweat getting 10mA exactly), alter the source resistor.
To adjust CCS current (and it's totally noncritical, you don't need to sweat getting 10mA exactly), alter the source resistor.
Not so much- though rp and gm can vary, mu is usually pretty well-matched.
For lower gain, it's easier to use a lower mu tube than to try using feedback with this topology. With a second voltage amp, feedback would be easier (and a lot greater), but you'd definitely want to make the headroom a chunk larger.
For lower gain, it's easier to use a lower mu tube than to try using feedback with this topology. With a second voltage amp, feedback would be easier (and a lot greater), but you'd definitely want to make the headroom a chunk larger.
The DN2540 cascode CCS has such ridiculously good performance, I haven't bothered looking at anything else.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.