Tom,
The PA is not the issue.? I have three designs using the IRF510 to as high as 50mhz
all wideband.? One even uses 8 of the in 4x4 push-pull for 225W at 6M.? with 6W in?
225 out is not what I'd call weak.
The ubitx has what I call a weak design using devices at gains unattainable at
higher frequencies like 20M and higher.? The problem is simple you need 47db
of gain from three stages to drive the finals.? The 3904s can maybe make
33db of gain at 10M and easily 50db of gain at 80m.? If you design the gain for
what they can do for flat 3 to 30 you get maybe 33db.? So you need more gain.
That means more stages and the ubitx has a shortage then.
Basically to get 10W from a IRF510 you need about 350mw at 80m and maybe 450 at 10M.
Put that in to a competent design you get 10 to 14W from IRF510s at 13V.? Do that at 24V
and put in about 800mw and 25-35W happens.??
But you need an amp that has the same gain at 80m and can produce that gain at 10M.
People make snide comments about the IRF510 being a switch and the 2n3904 is exactly?
the same thing, a cheap low current switch.
Since we have three stages and need 47db of gain that means the average gain
is 16db (round number is 48).? For that you need a device with a FT of not less
than 1200mhz (certainly not 3904s at 300FT).? TheDrive stange needs to do
1W for comfort. a pair of decent parts like 2n3866 or 2n5109(and others)
can do that easily.? That's not 4 but two.? But gain at high power is harder
so at best we can get 13db from them.? So the prior two stages need to produce
about 35db.? Best again is a single 5109 for the pre-driver and we get more
gain at lower power say 15db.? That leave the q90 stage needed to produce?
a few milliwatts and 20db of gain.? A 2n5179, BFR106, MAR6, and others
easily can do that.? Of course all the transformers need to be redone and
the output feed and transformer is pretty poor.? The board layout complicates
matters being much too tight.
allison
