HFU HF Underground
Technical Topics => The RF Workbench => Topic started by: digitalmod on January 22, 2018, 2234 UTC
-
This area has changed so much that basic miracles seem to be possible. This young RF engineer tells in far too little space how to design a successful Class E RF amplifier.
He is talking here about 1000 Mhz or 1 Ghz Class E power amps. Like wow.
And the rumors have it 40 meters is a high frequency for such designs. like DUH!!
The peripheral CL networks , choice of device, operating frequency, design parameters for SAFE regions of operation are all covered. But, for me twenty years away from this stuff, my eyes glaze over.
This does show the myth that there are no devices that can be designed for efficient class E RF operation at frequencies we all enjoy, namely shortwave.
Hope this doesn't put you to sleep. I certainly am just getting a feel for what I need to consider in designs.
When he talks of conjugate match, simply says tune out reactance. When he says PULL, that's a way of phase shifting nth harmonics so they are in phase with fundamental and their power adds to the fundamental !
https://www.youtube.com/watch?v=iABwHeZ3_Jw
-
I fast forwarded through the video but as an ADS user, I'm pretty sure that he hasn't included full parasitics, based upon his schematic at the end (which I post a link to below). He has some transistor parasitics and the microstrip model includes some as well but it isn't complete, e.g., bondwire parasitics. That will have some effect.
https://youtu.be/iABwHeZ3_Jw?t=727 (https://youtu.be/iABwHeZ3_Jw?t=727)
In addition, I suggest that you want to look here for info from a guy that has built a lot of QRP Class D, E and F stuff.
http://www.aoc.nrao.edu/~pharden/hobby/Hobby.shtml (http://www.aoc.nrao.edu/~pharden/hobby/Hobby.shtml)
I quote from MOSFET Class D/E/F amplifiers - Part 2, page 4, bottom of the second column and leading onto the third column:
The second point with the above equations is how the switching losses are frequency dependent, due to the term "2fo." This is why most Class E circuits on the internet are only for 160M or 80M, as even a sloppy job of designing the circuit and using a sinewave drive (equations) will still yield high efficiency...This is also why those scaling these amplifiers for 20M have had disappointing results, as the switching power losses double as you double the operating frequency.
-
:-X
Thanks for additional readings. Its a nice way for me to revive my electronics engineering, put in the closet about 12 years ago when I retired. Realized this area is really where I most enjoy doing things.
The idea of class E rather confounded me, because from old days, class C is easy to see as an energy pump into a swing (the LC tank). This idea of class E does have significant differences.
I'm not a fan of junk yard designs that I've seen hams use. I guess if I get to having something, its either somebody else vending theirs (that looks not likely) . And as some wanna be designers say, they could do some, but who would fork out lots of cash for same, as they are of course limited in their use. Surely even 500 dollars will turn pirates off entirely. 8)
It would seem we are always looking at somehow getting around the physical reality, capacity does take finite time to charge.
However, at the common pirate frequencies, even to 10 MHz (please do not jam WWV) haha, the device physics of FETS available today probably make it relatively easy for a 200 watt AM class E transmitter. I say this purely on intuition not on calculation or research. It also seems duty cycle twisting to a narrow domain at the input can very much increase efficiency. The one thing that seems clear here , a successful amplifier would produce RF voltages many times greater than Vdss !!
That must be calculated into designs and especially selection of suitable devices. Sloppy designs result in blown up components.
::)
-
8) Being out of the RF loop for at least 13 years, I did a bit of casual investigation and found very compact class E Fet amps, dating from 2006, boosting 60 watts average and over 400 watts peak output. Designed for microwave use. And yes with Ku band etc.. today I have a 5 watt radio uplinkling a geosat with 2 watts at 26,000 mhz.
So, the technology has moved along, but the complexity of design and adjustments are like wise making the old timers scratch their heads and say, Nagh that can't be so.. However as Houdini said "believe" :'(