Unity Gain Stage

[dorrisant]

I am building a 2 channel thing... and I need to match phase with the other channel. Channel 1 is 3 stages.

On channel 2, I am using the vibrato channel from an AB763 circuit. This is two stages. What would you guys suggest as a 3rd stage? I don't want extra gain added. Trying to keep this as the "clean as possible" channel.
Maybe something along the lines of a cathode follower? Idk, the gain of the cf is a little less than unity, so I'm guessing that may be good enough. If so, how would you implement it?
Maybe it should just be a normal triode stage with reduced gain. That's why I'm asking... I want to know how some of you have addressed this issue in the past. Suggestions?

[sluckey]

A CF is no good because there is no phase inversion. But a cathodyne will work well. Gain is about 1 and the signal on the plate is inverted.

[dorrisant]

Good point Steve... no phase inversion.

[xtian]

You can also use a MOSFET: http://www.geofex.com/Article_Folders/m ... tfolly.htm

[Stevem]

Correct me if I am wrong, but if you will never blend to two channels together, then it does not matter that there out of phase if you use a CF, as it will then be just like stock BF or SF Fender with reverb!

In the schematic you posted the normal channel enters the PI with the same phase as it hit the first gain stage with, on the vibrato channel it enters the PI out of phase, which is why you can't jump channels on these.

[Firestorm]

The Vibrato channel in AB763 is three stages, not two. The reverb mix stage attenuates the dry signal by about 95% by running it through the voltage divider formed by the 3M3 with the 220k/470k shunt, but does invert it.

[thetragichero]

first thing that popped into my mind is the subject of this thread: https://ampgarage.com/forum/viewtopic.php?t=15618

[Ten Over]

Whereas a cathodyne will invert with near unity gain, it also has half as much headroom as a fully-bypassed gain stage with the same load. This may not be desirable for a channel striving for "clean as possible". Attenuating the signal and then running it into an inverting gain stage like a paraphase PI does or using an inverting gain stage with local NFB are a couple of possibilities.

[pdf64]

And a cathodyne’s high value, unbypassed cathode resistor will tend to degrade the signal to noise ratio.
The self balancing paraphase around V4b of a Vox AC50 MkIII may be a good option https://el34world.com/charts/Schematics ... ac50_2.pdf
http://aikenamps.com/index.php/designin ... amplifiers

[Firestorm]

Whereas a cathodyne will invert with near unity gain, it also has half as much headroom as a fully-bypassed gain stage with the same load. This may not be desirable for a channel striving for "clean as possible". Attenuating the signal and then running it into an inverting gain stage like a paraphase PI does or using an inverting gain stage with local NFB are a couple of possibilities.

If you wrap enough negative feedback around a common cathode gain stage you wind up with an "anode follower," which I think doesn't suffer the headroom limitations of the cathodyne. Merlin covers it in the preamp book. Also in O'Connor's TUT 1, iirc.

[Ten Over]

Here are three ideas for unity gain stages that invert.

Unity Gain with Inversion.png

[dorrisant]

Thank you to all for the suggestions!! Now I will go let the iron smoke.

[pdf64]

Here are three ideas for unity gain stages that invert.
Unity Gain with Inversion.png

With the feedback designs, what’s the rationale for leaving the cathode unbypassed?

[Ten Over]

With the feedback designs, what’s the rationale for leaving the cathode unbypassed?

I calculated a closed loop gain of 1.03 using an open loop gain of 32 and an open loop output impedance of 68K. I calculated a closed loop output impedance of 15.9K. Under these circumstances, I could see no reason to use another component in order to bypass the cathode resistor.

[pdf64]

An unbypassed cathode facilitates noise to enter the signal path. To avoid this occurring early into the signal path (ie where it causes the greatest degradation of the signal to noise ratio), it’s beneficial to fully bypass the cathode of an input stage.