always liked the original amp designed by Mr. Leach, but thought that
all the wiring was a bit of a bad solution, because it is easy to make a
wiring mistake and maybe destroy expensive transistors in the process.
This is the reason that I made my own layout to fit my need for a
powerful and good sounding amplifier. I like the flat pack transistors,
so I decided to use these types in my design. There are some issues
regarding the safe operating area for these types compared to the
original TO-3 types. This is the reason that I choose 5 parallel
transistors in stead of the original two. The higher transistor count
enables me to run the amplifier with +-68V rails and into a lower
The PCB panel for the 10 transistor version looks like
The main PCB is 25.4 cm long and
10.8 cm wide
I'm pretty happy with the
result. The board is 2.4 mm thick and the copper is 105µm (3 oz) I have
had a prototype of this board playing for the last year or so in my
living room. The amps has five parallel output transistors and is very
powerful putting out around 180W into 8 ohms.
The two small PCBs in the bottom of the PCB panel
are both for the thermal tracking diodes. The bigger PCB is for 1N4007
types and the smaller is for SMT (Surface Mount Technology)
The next picture is the bottom
view of the board. The two rail lines run nice and fat down to each of
the output transistors. Each transistor has a small cap on the collector
pin to maximize dynamics and improve stability. There is room for two
large 10.000µF caps on the board to decouple the rails, along with fuses
for protection and zobel network for stability.
Underneath here is a picture of me mounting some of
the components on the first board. My fiancée Janni took the picture.
Next after mounting all the parts it was time for a
test. Underneath here is a picture of the test setup in my basement. The
black thing in the background is a 800VA 2 x 45V AC transformer. Nothing
was smoking, so I connected the 8 ohm dummy load seen to the right of
Initial testing showed good stability and an
excellent power output level of around 150W RMS into an 8 Ohm load and
275W RMS into a 4 ohm load. The rail voltage dropped from +-63V unloaded
to +- 56V under the 8 ohm load and to +- 53V during the 4 ohm test. I
plan to eventually run the amp at around +-70V unloaded.
Square wave response is ok, at 1 kHz into 8 ohm
there is a small overshoot, but nothing alarming. The oscilloscope is
set at 20V/DIV.