2007-02-24
4-5 Watts PEP
My order of 50 IRF510s from futurlec arrived in the mail, as did some cute red-anodised TO-9 heatsinks from Rockby and a quick stop at Jaycar on Thursday evening for TO-220 heatsinks meant I had all the components to build the beginnings of a TX power-chain.
Without a 24+ Volt supply completely thought out I decided to go with a more conservative 4 Watts to start with, having not worked with IRF510s before. A simple three-stage chain was assembled, starting with a 2N3904, through a 2N3053 and finishing with a single IRF510. The 2N3904 was loaded with about 800 ohms on the collector, the 2N3053 was designed to match a collector load of 200 Ohms into the 47 Ohm gate resistor I put in the IRF510 stage. The IRF510 stage was simply biased with a zener stabiliser and a 10 K pot, through the decoupled bottom of the 47 Ohm gate resistor. The drain transformer stepped down the 50 Ohm load to 12.5 Ohms to achieve the desired output power.
Apart from a little mistake where I connected the drain transformer incorrectly and loaded up the drain with 200 Ohms (resulting in much less than 1 Watt out!), the power chain went together easily and proved very robust. So far I haven't destroyed the first IRF510, despite abuse ranging from almost infinite load mismatches and even running the amp without a heat sink for about 1 minute.
The initial 2N3094 amplifier proved unnecessary when it came to matching up the power amplifier to the AM modulator, so I rebuilt the drive stage with just the 2N3053. This also gave me the opportunity to build in a drive-control pot which proved quite useful during testing.
I noticed that the output level was controllable from basically Zero to a 4.99 Watts CW just by manipulating the gate bias. Some investigation of gate-bias modulation might be in order! Although I have not investigated the linearity yet.
Adjusting the drive level and bias allowed quite good audio quality to be achieved, with an output power of about 4 Watts PEP. This is more than sufficient for an exciter driving a pair of IRF510s in push-pull at a higher supply voltage to achieve up to 40 Watts, maybe more.
Now all I need is a metal box to build it in.
2 comments.
Parent article: "2007 80m Homebrew Challenge".