Circlotron Amp - Output Stage Regulation

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hello all, I'm building this circlotron amp base on the Mr. Corrado's work presented here:

Circlotron AMPLIMOS one stage amplifiers

It works, sort of. I'd I made a couple silly mistakes causing blown output bransistors. But my question here is about noise. I get consider noise at the output with an unregulated output stage. It does seem that with it's floating power supplies tied directly to the the outputs that it would be more affected by supply noise.

The voltage gain stage has good regulation. The overall gain of the amp is 38dB and has no loop feedback. I plan on making a cap multiplier with a 300V, 120amp insulated gate BJT.

Any thoughts or suggestions about this?
 
Here is the schematic:
 

Attachments

  • Circlotron_Schematic_Rbias_SS_4-9-2017.png
    Circlotron_Schematic_Rbias_SS_4-9-2017.png
    56.7 KB · Views: 257
Let's simplify, no need for a double VAS/fase inverter.
I think you can DC-couple the output stage with bias and balance in the VAS.
And some protection for the output gates.
Mona
 

Attachments

  • Circlotron_Schematic_Rbias_SS_10-4-2017.png
    Circlotron_Schematic_Rbias_SS_10-4-2017.png
    73.5 KB · Views: 209
Wow , thank you ... That is nice. Good point. The resistor bias is a work around of a thermal stability problem. The bais voltage at VAS Output fluctuates too much, like a couple volts much, and the bias voltages between positive and negative phase change some too. Part of that is the current source warming up and part of it is temperature imbalance between devices I think. I can adjust with potentiometers in the VAS degeneration resistance but it bias voltage keeps changing by as much as .5 volts. That is too much and cause havic with output stage. And I plan on adding DC offset correction as well. The resistor bias works well for now though.

So just got to work out some bugs first ...
 
I forgot to place the ground on the right place at the zero adjust :guilty:
To make the bias current more temp-sensitive, some diodes (to put on the same temp as the output transistors) will help.
Mona
 

Attachments

  • Circlotron_Schematic_Rbias_SS_10a-4-2017.png
    Circlotron_Schematic_Rbias_SS_10a-4-2017.png
    74.4 KB · Views: 182
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.