A 6SN7 direct coupled preamp & regulated supply for everyone's perusal.

[Miles Prower]

Originally posted by G
Hi Miles. Thank you very much for posting a reply. Here is the story about this circuit. I don't think I really need all that gain but maybe you can tell me after I tell you where I am coming from. I have been planning to build a matching (chassis wise) preamp and amplifier for a while. I actually decided on the amp first. It is a 12B4A gain stage and a 12B4A output stage.

If you don't need the gain, you can ditch some, and improve the linearity by doing a plate follower. First, get rid of the cathode bias resistor and replace it with a green LED. The Vf is quite close to the required grid bias, and the LED has very little AC resistance in the forward direction.

Next, apply the NFB to reduce the Avcl to about four (doesn't look like it'll need much more). If there's a problem of DC offsets from the source, you can add a DC blocking capacitor. For this, move the volume control to the output. The A Number One problem with this is that variations in source impedance, as you'd get with a volume control, will cause the closed loop gain to vary since that impedance is in series with Rs. Given an Avol= 15, make Rf= 5.1Rs. That should give Avcl= ~4.0. That should improve your THD to ~1.0%.

Plate Follower

[G]

Originally posted by Miles Prower


If you don't need the gain, you can ditch some, and improve the linearity by doing a plate follower. First, get rid of the cathode bias resistor and replace it with a green LED. The Vf is quite close to the required grid bias, and the LED has very little AC resistance in the forward direction.

Next, apply the NFB to reduce the Avcl to about four (doesn't look like it'll need much more). If there's a problem of DC offsets from the source, you can add a DC blocking capacitor. For this, move the volume control to the output. The A Number One problem with this is that variations in source impedance, as you'd get with a volume control, will cause the closed loop gain to vary since that impedance is in series with Rs. Given an Avol= 15, make Rf= 5.1Rs. That should give Avcl= ~4.0. That should improve your THD to ~1.0%.

Plate Follower

How would this effect the output impedance?

[Miles Prower]

^^^^

Since this is a gNFB scheme, the NFB will reduce the Zo. Being a current feedback method, it will also reduce Zi. The latter isn't such a big deal since the open loop Zi can be made so high to begin with, way in excess of what you'd get with BJTs.

[chacha]

Hey everybody. Is this cirquit reversing the phase on the output?