Hi Dick,
First of all the FET's used.
The original design uses 2 times 34NB20 FET's Vds200V/Id 34A, Rds(on)
<0,075 Ohm
> I don't have them, but I can replace them by IRFP260N or IRFP360
The figures for a IRF360 are: 400V/23A Rds 0.2Ohm
> I'll plan to run the stuff on 50Volt.
The IRF360 has an higher Vds but also a litle higher Rds, I wonder if
the Rds has an effect on the output power.
Assume you run the FETs at 15A (and 50% duty cycle) then the "switched
state" dissipation of the 34NB20 will be less than 8 Watt while the
IRF360 will dissipate 23 Watt.
But much more important is the "dynamical" dissipation : when the
FET's are switched you have both rather high current and rather high
voltage, thus rather high dissipation. The faster you can switch the
FET's the lower the losses, both the ON/OFF times and the gate-source
capasitance will affect this.
Second and more important question is about the output transformer.
I don't have the original 3C90 core, so I'm planning to use the core
of an old computer monitor for this.
I'm using a "double-U' core from an old TV set, it doesn't get hot at
500 W output.
Did any of you do this before on the G0MRF's design, and if so I
would like to know if there was any difference in number of turns
used.
I guess the amount of power greatly depends on this output
transformer so that makes it to one of the important parts of the
project........
If you're running a few 100 W from a 50V power supply the output
impedance of the PA is very low (few Ohm), so you will need only a few
turns on the primary (FET) side.
I'm using a variation of the G3YXM design, I believe it is pretty much
the same as the G0MRF design. Primary winding is 2 x 3 windings,
secondary (to LPF) has 25 windings with a lot of taps. By selecting
the tap I can easily vary the output power from 50 W to 500 W.
Maybe the number of turns might also be dependend in some kind of way
to the type of FET's used.
As far as I know : no
73, Rik ON7YD
|
