Unilab 10m FM conversion Transmitter PA mods - Driver Stage

I had a chance to rummage around in the junk box yesterday. If you're wondering why I would turn to a transistor I recover from a radio, instead of buying online, see my post at http://vk6tt.blogspot.com/2016/05/hf-transmitting-transistors-for-qrp.html

 Anyway, the first radio touched was a Philips 828 E band. They contain a BLY87 which would be a good candidate as a driver. However, they are stud mounted so that could be awkward.

The next radio was a RT85 for low band. They use a Mitsubishi 2SC1971 as a driver which is a nominal 6W VHF high band transistor. Not quite the 10W I was looking for but it forced me to rethink my approach.

The MRF247 final transistor has a gain of 8.5db at 175MHz. Adjusting for frequency at 6dB per octave that equates to 23.8dB at 30MHz. Much more than the 10dB I was allowing for. The full gain at 30MHz cannot be used if oscillations are to be avoided. However, if I shoot for 13dB of gain then the 2SC1971 would only be required to produce 5W, well within it's rating.

Adjusting the quoted gain for frequency suggests the 2SC1971 has 25dB of gain at 30MHz. Again, way too high. But the prospect of avoiding a pre-driver stage is a tease since a gain of 13dB from the driver stage dictates an input power of 250mW to achieve 100W from the PA strip. The exciter delivers 100mW. Quite close and maybe the gain per stage I have allowed is too conservative.

I'm going to try just a driver stage and dispense with the pre-driver.

I had to revise the input stage to the PA transistor for the increased collector load the driver stage required. After several iterations with the smith chart I concluded the higher collector load required meant the existing matching configuration could not be made to work.

My approach now is:

• C1, currently 100pF, is padded with a 10nF capacitor,
• removed the 2 turn coil L1,
• pad the input of the strip-line with an extra 860pF of capacitance using two capacitors, 470pF and 390pF.

That gives me potentially a good match between the base of the PA transistor and the collector of the driver stage. I will tackle this shortly and post a full recipe in due course.

73's
Richard