The Super Charged DSW-20
Special to The ARS Sojourner
How to fill a box The construction is kind of a sandwich: DSW is on the bottom of the box and the amplifier on the top. The amplifier section is built on a sigle sided printed circuit board using an "ugly" method. The components are on the copper side and there has been carved suitable "islands" into copper surface. By laying the toroids on their sides and choosing a flat heat sink for 2SC1945 it was possible to squeeze the thickness of the amplifier to 20 mm.
The outer measures of the box are b.11.2 cm (4.5"), d.10 cm (4") and h.4.3 cm (1.7") not including the knobs, connectors and rubber feet.
Talking about the knobs, that "over sized" freq adjusting knob suits to me. I tried with a smaller one but this rig needs a bit different adjusting; tune - stop and listen - tune - stop and listen... If you tune continuously you only hear birds and not recognizable CW. With bigger knob it goes better.
The amplifier itself is more or less ordinary, 2SC1945 is a high power gain device and driven with DSW it grinds out (class C) in my case more than 15 W at 13.8 V supply. Limiting supply voltage or limiting drive the output drops down to 5 W or less. Quiescent current when receiving is 40 mA for this combination, going up to and over 2.5 Amps when transmitting. This transistor seems to be a very good companion to DSW but 2SC1969 functioned OK too (not so much gain). 1945 had the benefit of tab connected to emitter and in my case directly to chassis and cooling fin. The transistor does not warm much but I used anyway a heat sink because there was enough space to do that. Same thing with ventilation holes - I made them just to be on the safe side.
The amplifier I used is not "production friendly". I made three of them and noticed that every one needed slightly different component values for optimum performance. If you are going to try this one, be prepared to play with different component values and to use some measuring equipment. I would say that oscilloscope is very useful, VOM of course, dummy load and swr indicator. Laboratory power supply with current limiter wount hurt either.
Follow this link to the schematic.
T1 = FT50-43 (Amidon) 2 x 6 turns bifilar 0.3 mm enamelled Cu
T2 = 2 pieces FT50-43 (glued together - easier to wind) 3 x 6 turns 0.7 mm Cu
L1 = T50-6 (Amidon) 10 turns 0.7 mm Cu
FB = Ferrite Bead
All coils wound evenly on toroids
RFC in output is for pulling output DC potential to gnd.
Transistor bolted to chassis or heat sinkAdjust 70 pF trimmer to max received signal (resonates with 4.7 uH coil in RX)
Power can be adjusted with changing 39 ohms resistor / drive level / and supply voltage. In my case DSW worked still with 5.5 V but output was then almost zero
Capacitors in tank circuit min 100 V
Be prepared that you must tinker with 680 pF capacitor. Its value depends on the layout. Find the value which gives the best output to 50 ohms dummy.
If your load is not 50 ohms, 200 pF caps may need adjusting too.
Modifications to DSW:
I removed the TR-switch capacitor C34 (27 p) and connected the diodes to the new output tank circuit using a 100p trimmer capacitor. Now I am able to adjust the series resonance to the peak. -
I short circuited D8 (1N5818) and got more voltage to the circuit.
I put a heavy duty diode in parallel with DSW and a fuse in series. If I will some day use a wrong polarity I think that the fuse will blow and not DSW.
Added a hi-intensity LED (draws 1 mA) to the IRT switch. I am absent minded and need a visual sign that IRT is on. I hope that is saves me from some confusion.
Photographs of my modified DSW-20:
Photo 1; Photo 2; Photo 3; Photo 4; Photo 5; Photo 6
73, Matt, OH2CF