|
Evan,
?
Had always assumed that topology of those filters would present a high impedance to all harmonics.
I've attached an LTSpice simulation for the output LPF's on 80m
as used on the QCX, QCX_50W_Amp, the QMX, and the uBitx.
Does an AC Analysis on all of them from 1mhz to 20mhz.
Will archive those two (very small) files under KE7ER in the files section.
?
You can click on the output node to see the voltage across the 50 ohm resistor,?
or click on the voltage source to see the current going into the LPF, which is inversely proportional to impedance.
Things look considerably more complicated than I was expecting.
?
Looks like somewhere around the second harmonic, it's fairly high impedance looking into the LPF.
For the higher harmonics, that impedance is generally no more than the it is for the fundamental.
A push-pull amp should mostly take care of the second harmonic, and giving the odd harmonics an easy path to ground
suggests to me that these are most suited for a current mode amp.
So perhaps the QMX would do better with two L502's, so the two phases of BS170's can have their drains uncoupled.
This is what the uBitx final does.
The QCX 50W amp uses two windings on T2 to supply power to the FET's, that's equivalent to feeding power at
the center tap of the output transformer primary, and best suited for a voltage mode amp.
?
I have only recently found that these amps can be classified as voltage mode or current mode,
and so may be a bit to fixated on it.? ?It seems the distinction is not always as clear cut as I had expected.
All of these amps seem to work well enough, though I still wonder if the QMX could be more robust.
Could be that the best path is to just replace the BS170's with TN0110's when they blow..
?
Jerry, KE7ER
?
?
On Mon, Mar 24, 2025 at 02:36 PM, Evan Hand wrote:
|
