[time-nuts] homebrew 13 dBm distribution amplifier based on NIST design 5 to 100 MHz

[dk4xp at hoffmann-hochfrequenz.de]

I've done a redesign of the distribution amplifier without
transformers or other ferrite (apart of dc filters). This required 
2 transistor chains in parallel or 96 mA at 24V / Channel.
 
It runs quite hot but looks like it could work from KHz to 600 MHz+. 
Harmonics are down 40 dB at 250 MHz/13 dBm including SMPD signal generator
and the 54846B scope digitizers.
AF decoupling is in again as well as Bruce's base divider caps.

http://www.hoffmann-hochfrequenz.de/downloads/IMG_0499__Q73.jpg

I had quite a lot of stability problems in the last stage; 
it liked to oscillate at 1.5  to 3 GHz. I had to use 33R
base stoppers for the pnp transistors and tried a week's evenings
with Genesys to get rid of them to no avail.

I even tried 0603 base decoupling caps right _through_ the board
from the base pad to the ground plane 0.5mm / 20 mil away.
 
There is still a lot to do optimizing the voltage drops
per stage and characterizing that thing. Any idea of more
harmless pnp wideband transistors with Ptot = 1W?
Maybe I could run this with -24V and use more NPNs and the
pnp only in the CC stage.

The efficiency might be improved with a 10 uH choke from the 
final collectors to gnd. That could be accomplished without
ferrite or at least with a large air gap.

regards, Gerhard dk4xp