EME 23cm

Equipments

LNA : 0.22 db NF 38db gain (G4DDK)

Transverter: TR 1296 H by Khune (20w)

Transceiver Elecraft K3 + P3

SDR receiver to monitor the band activity (connected in parallel on the 28Mhz IF) : The little giant Perseus

With a 40m line to the dish the power delivered to the feeder is only 8w and self echo of 10db above noise level(first testing level)

29th/Sep/2011 Upgrade

One further step, power increased to 250w for indoor testing which means about 100w on the feed (self echo of 20db)

16th/Nov/2011 Upgrade

Power amplifier moved to the dish great improvement in self echo reaching 25db over the noise

250w SSPA installation

14th/Jul/2012 Upgrade

250W PA removed and installed instead a 1Kw mosfet unit, the currently available driving power and the line attenuation
limit the output to 500w at feed

22nd/December/2012 Reference frequency installation

A Rubidium lamp locked oscillator (surplus item) provides a 10Mhz reference frequency, the K3 transceiver and the Kuhne transverter are frequency disciplined by the external 10Mhz source.

May/2013 Power Improved

By adding a small amplifier between transverter and PA the power at feed has been increased to about 800w (CW or SSB), using 400w when in digital mode.

Aug/2015 Power Monitor


Due to several LDMOS failures (and not knowing why) a few preacautions have been taken at least to keep the situation under control.

Power supply spike limiter

In parallel with the 48V power supply a power mosfet triggered by a 63V zener will short any spike above the zener value;

Circulator

A one Kw circulator connected at the PA output should protect from any load mismatching. The third port of the ciculator goes via a 3 db lossy line (25m of 1/4") to a 450W dummyload with a DC monitod output.

Power Monitor


An indoor (Arduino based) monitor receives signals from:
-PA output
-Driving power
-Reflected power
The real time situation in shown on a small TFT display:


The picture could have been much better.

There is a provision of showing also the DC 50V current but not implemented yet.
Between driver and PA there are more than 3db of line loss and hence the apparently low power gain of the final stage.