Very interesting reading...

[kc2ifr]

Wonder what u "techies" think of this.
I think the amp described is for SSB only........but not sure.
DL and read the original article. Interesting stuff....
http://www.ifwtech.co.uk/g3sek/misc/g2daf.htm

[Steve - WB3HUZ]

You might want to do some more reading on that design. It is known to create lots of IMD.

[W2VW]

If you check the AMPS reflector archives there is a ton of good information on this setup. There is also 500 Lbs of the other stuff...

[kc2ifr]

Wow......I did check the amps reflector on this amp design and it is/was a HOT topic. Mr. Measures HATES the damn thing (he hates about everything!). Others seem to like it. It all makes for some interesting reading.
And Steve......Im not building one or advocating the design.....I just thought the idea is interesting. Its been around for a long time......but I just ran across it.
Bill

[W3SLK]

If I recall the big thing was that screen would not keep up with the rest of the circuit thus creating the poor IMD and distortion. Ian, G3SEK and Rich, AG6K battle back and forth. I find myself most of the time aligned with Rich, even though he does harp about VHF parasitics.

[K1JJ]

Interesting design.  Always good reading stuff from people trying different experiments. That's how we come up with good ideas.

Normally, a tetrode is designed for its best linearity with a rock solid screen voltage supply and recommended screen current with zero grid current.. and idling at about 1/2 it's rated diss. What this design is saying is that they can make it better by varying the screen voltage in sync with audio modulation. They mention -45db 3rd order with that circuit ["on the air checks"] when that tube normally does -28db at best.  I doubt that kind of improvement [probably none or worse] or everybody would be doing it by now.

What they are trying to do is raise the idle current during transmission at a slow time constant keyed by audio. I can see the efficiency aspect there, but not the big IMD improvement they claim compared to a stage with fixed idle and no screen feedback..

In my opinion, either regulate the screen rock solid like designed or better yet, go with a power grid triode in GG that was specifically designed for linear service. Tetrodes were designed for class C FM,  AM, and pulse service.  At this point I run only triodes in linear and either triodes or tetrodes/pentodes in class C.

There are beautiful triodes out there now for cheap. The Russian GS-35B for $80, new. [a 3CX-1500 more or less]  Also, for the extreme, the YC-156 [3cx-5000] pulls for about $175.  Both have great IMD and simplicity.

But, if motivated, someone could easily try this circuit on any tetrode linear and vary the level of "positive feedback" to the screen. Use a good spec analyzer and it will be easy to tell if there is any improvement. Try to dial it into the sweet spot. Let us know what you find.


T

[WA1GFZ]

The closer you bias a tube to class A the better your IMD. Little gimmics with the wrong time constant make a mixer with sidebands, kind of like controlled carrier screen mudulation. I have never seen any bad information come out of Measures.
Take a big tube with plenty of bias and a big blower. Need to strap talk louder and have plenty of reserve drive.

[Steve - WB3HUZ]

The three-tone IMD test on this design shows it producing nasty IMD. Measures isn't the only one who dislikes it. So does Rauch. When Measures and Rauch agree.... well, 'nuff said!

[W3SLK]

Steve said:

The three-tone IMD test on this design shows it producing nasty IMD. Measures isn't the only one who dislikes it. So does Rauch. When Measures and Rauch agree.... well, 'nuff said!

Aint that the truth! The way I thought the DAF was to work was along the same lines as modulating the screen, somewhat similiar to how we do it with tertiary windings etc. But like Frank said, due to the T/C the screen winds up lagging the the RF input. Ian and Colin, K7FM swear by it (while other's swear at it!). I think that the both of them are contrarians just to piss off Rich Measures.