[Top] [All Lists]

Re: [Amps] Designing the Cleanest Linear with RF Negative Feedback

To: "Steve Thompson" <>, <>
Subject: Re: [Amps] Designing the Cleanest Linear with RF Negative Feedback
From: "Tom Cathey" <>
Date: Thu, 22 Dec 2005 14:17:48 -0500
List-post: <>
Please disregard my question below:

"Would there be a major disadvantage to running the whole tetrode chain in 
GG with standard voltages on the grid and screen?  It seems it would be much 
easier to tame and also have GG neg feedback and pass thru power to the 
output.  Maybe link coupling between stages work work."

I realized that a tetrode run with voltages on G1 and G2, cathode driven 
has the same gain as one that is run GG. So why bother?  I think hi mu 
triodes in a class A GG chain are simpler and would do the same job  - and 

I can see the problem with a Class A chain, however.  Putting the DRIVER 
chain in class A at 10% efficiency is super clean and the power involved is 
not too bad. BUT, the final running in class A, to be a good IMD match to 
the driver, will require a 3CX-10,000 or greater to put out a clean 1500w 
class A. If a smaller final tube is run AB1, the final IMD goes way down and 
almost defeats the effort of the clean driver chain.

The only solution I see is to have a loop of neg feedback going ALL the way 
back from the final to the early stage, speading the FINAL's need for NFB 
(gain reduction) out amonst many stages. Otherwise, if NFB is used ONLY from 
the final to the preceeding driver, there is low gain in these two stages 
and a need for ANOTHER big pre-driver. Like chasing your tail.

So, bottom line is: Use a GG driver chain in class A, with the FINAL in AB1, 
but with negative feedback around the complete loop. More gain can be added 
in the early stages if need be, not in the high level stages. But I 
understand NFB around more than two stages is a stabilty problem 

Again, the idea for using all GG triodes vs: tetrodes is for 10-40M 

I am wide open to correction if these are poor ideas...:-)

Tom, K1JJ

Amps mailing list

<Prev in Thread] Current Thread [Next in Thread>