> Date: Sun, 29 Dec 2013 18:59:40 +0100
> From: [email protected]
> To: [email protected]
> Subject: LF: MF PA
>
> Hi Wolf,
>
> What type of FETs are you using? The standard IRFPs? Many of them have
> internal Z diodes, so no addional RC network or Z idode is needed. The
> losses are there anyway, of course.
>
> In some older projects i also played with that critical working point,
> i.e. where one FET shuts down and the other opens. I've tried simple RC
> networks (series R, parallel C) with an 100 Ohm trimmer. That worked
> reasonably but usually it is not even necessary. My small MF PA is using
> a simple ICL7667 and 2x IRFZ48N (Reichelt...) without problems.
> My large PA generates up to 1.2 kW, which is also enough to reach 1 W
> ERP, https://dl.dropboxusercontent.com/u/19882028/MF/mod%20MF%20PA.jpg.
> It was designed for field experiments (Field day) where MUCH more
> inefficient antennas will be tested, e.g. in France, next year, if they
> have the allocation then...
>
> 73, GL, Stefan/DK7FC
>
> Am 29.12.2013 18:09, schrieb wolf_dl4yhf:
> > Hi Stefan,
> >
> >
> > > Hmm, P ~ 300W? :-)
> >
> > Not really... the design goal was driving the PA with the 13.8 V, "20
> > ampere" power supply. There actually is some stray inductance (despite
> > using 17 mm wide copper foil) so the load "seen" by the drains is a
> > bit higher than 12 V / 20 A = 0.6 Ohm. The max DC input current is now
> > about 15 A, but that's ok for me. Also, after adding the first lowpass
> > (which transforms from 12.5 to 50 Ohms, at very 'moderate' RF
> > currents), the waveform at the drains isn't such a nice square wave
> > but the HUF75343 can 'absorb' the very narrow spikes (80 ns long, and
> > over 55 V) when none of the two FETs conduct for a very short time.
> > According to the datasheet (where the FET are mistreated by driving an
> > inductive load, as a test) they should survive this, even without an
> > RC snubber network or an extra Zener diode.
> > I will find out soon - also if P ~ 150 W is enough to reach 1 W ERP.
> > From experience with the old PA (in which the FETs were biased into
> > 'almost linear operation', at the expense of efficiency), it should be
> > possible.
> >
> >
> > 73,
> > Wolf .
> >
> >
> >
> >
> >
> >> An alternative to the foil could be: 4x
> >> http://www.reichelt.de/index.html?ACTION=""
> >> 200/120
> >>
> >> 73, Stefan
> >
>