Building a 145 MHz high power solid state amplifier

(pictures coming soon!)

After participating at some contests in the VHF range, "private" as DJ9KW or with our contest crew DR2W, I decided that 100 Watts from my IC-275H aren't enough to get the goods so I decided to build an amplifier.

So I had to choose between a tube amp and a solid-state amp project.

At the same time a friend started to build a GI-7B tube amplifier but after some weeks we got some problems we were unable to solve at that time so I decided to try my luck with a solid state PA (by the way the GI-7B amp works fine now, too!).
A week before I received the recent "VHF communications" - magazine where they explained an interesting design uning a Semelab FET "D1030UK" with max. power dissipiation of 600 Watts an 28 Volts Ub and a drain current of 20 Amps.
I still had a 24 - 36 Volts transformer, 20 Amps, bought at a flea market for approx. 10 bucks on stock, so I only had to order the semiconductor and a 19" rack case was available, too.

The power supply:

To supply the FET with power, I needed a stabilized +28V voltage source capable of delivering 20 Amps.
I decided to use a LM723 as voltage controller IC (although it's a "dinosaurus") and some power transistors to regulate the voltage.
You can view the schematic and download the schematic / pcb layout here. To open the layout data you need the freeware tool "Eagle" from CadSoft.

It is important to test the supply with a dummy load (e.g. car head lamps) to ensure it's not oscillating at any load condition. To be sure you could also add several 100 nF as stabilizing capacitors as shown in the schematic (perhaps that's a problem cause the LM723 is operated at its upper limit; I will try to add a zener diode to decrease the operating voltage of this IC when I have some free time).
Load regulation is great; no spikes when applying 100% load or disconnecting the load.

The amplifier itself:

The heart of the amp is as mentioned above the D1030UK FET from Semelab.
It is matched to 50 Ohms by a quarter wave transmission line with 25 Ohms (2x 50 Ohms parallel) at the in- and output.
On the input I've used 2x RG315 PTFE cable due to its roughness when accidently touching while soldering in the PA (the earlier used RG58 didn't look really nice after some hours of optimizing the PA ...); the output is transformed by 2x UT141 semi-rigid because it has a rather low loss, is small and allows a "hard mounting". I could have used Aircell 7 here, too, but this is so hard to bend and almost removed the pads on the pcb ... and remember the soldering-iron-thing from above ;.)

After the transmission line the FET is matched by a shunt and series capacitor and a series inductor. The input isn't so critical, here you can play around with some SMD C's; at the output I've used good ceramic capacitors and an air trimmer for fine tuning and to adjust the matching if there's a real antenna connected and not the lab dummy load to be able to tune to max. output.

The FET needs a positive gate voltage to set its idle current, this is done by a 7812 voltage regulator followed by a divider network with capacitors to prevent RF from entering the DC board and a 1k trimmer to adjust the quiescent current of the FET.
Perhaps I should add an active, temperature-compensated idle current controller because it strongly depends on the temp of the FET.

RX / TX switching:

The RF switches consist of the types shown on my project's site.
The input switching is performed by an OMROM relais as the FET needs about 10 Watts of drive power for 400 .. 500 Watts output power (view schematic here).
Additionally this pcb contains a relais which pulls the RX-line to ground while transmitting to prevent the amplifier from oscillating (isn't required but you never know ...).
I've also added a rf rectifier with a BAT41 shottky diode whose output voltage is proportional to the drive power, so the amplifier can be shut down when overdriven.
The pcb looks like this (only one Omron relais installed yet):

The output switching is performed by PIN Diodes as referred to at my High Power PIN RF-Switch for VHF.

"protection" circuits

SWR Protection:
If the output VSWR is greater than 1:3 the amplifier is shut down to protect the FET.

Current monitoring:
The current through the FET is monitored to detect oscillation (high current in RX state) and to avoid currents > 20 A

Voltage monitoring:
The supply voltage is monitored, so in case of a regulator malfunction (overvoltage e.g.) the power supply will be disconnected.

(c) DJ9KW, 25. 05. 2008

Input Power monitoring:
The drive power is monitored to shut the PA down in case somebody applies too much power to the input to avoid damage to the FET.