[Amps] 'heat spreader' sizing ??

[Steve Thompson]

> If you take a look at the data sheet, you will see it says:
> 
> "Typical pulsed performance at frequency of 225 MHz, a supply voltage of 50 V 
> and an IDq of 40 mA, a tp of 100 μs with δ of 20 %:" Then it shows 1200 W. 
> Everything on the data sheet refers to "pulses".
> 
> So this is designed for pulse use. With 100 uS pulses, and what I assume is a 
> 20 
> % duty cycle, it gives 1200 W, one would probably get about 1200/5 = 240 W 
> continuous. So the fact it's a 200 W amp continuous amp is not unreasonable.
> 
> Note is should be very linear for a 200 W transistor amp. The 1 dB compression 
> point of that design is 950 W.
> 
> I think for almost any ham amp, the 1 dB compression point would be less than 
> the rated output - here is is almost 5x the rated output.
> 
> That said, the transistors are pretty damm expensive for 200 W out.

The transistor is primarily intended for use with DAB or DVT 
signals, which are complex multicarrier affairs. The 100us/10% 
duty is chosen as a simpler but representative way showing what 
performance with 'real' broadcast signals will be, not as an 
indication of the maximum average power ratings.

To a first approximation, if you want a transistor that can give 
1kW pulse longer than 10s of us, you have to make a 1kW cw device. 
When running cw thermal effects will come into play, reducing gain 
and compression point a dB or two compared with pulsed operation. 
NXP app note  AN10800 describes using the BLF578 at 800W and 1kW 
cw in the FM broadcast band.

Steve