11
Circuit Theory
Much of the circuitry in the M•800 is self explanatory from the schematics. This section will
explain the unique circuits and architecture. Samples in this section will refer to Channel-1 for
circuitry that is identical on both channels.
INPUT CIRCUITRY
The signal path begins at the 1/4" TRS and XLR input jacks. Following Channel-1’s input, signal
is fed to a unity gain differential op-amp, U6A. The signal is next sent to U6B which serves as both
HPF, and CD horn EQ. The output of U6B is sent to mode switch, SW2, and also to Summing
amplifier U8A. The output of SW2 feed both front panel volume controls, R2 and R25, which in
turn feed unity gain buffers, U1A & B, (all on the display board) before returning to the main
board. In stereo mode, the signals on the channel-1 and 2 inputs are routed through the front
panel level controls, directly to their respective power amplifiers. U8A’s summed signal is used to
drive both front panel level controls in dual mono mode, with these signals being routed to
both power amplifier inputs. In bridge mode, U8A’s summed signal is fed to the channel-1 level
control , and the output of the level control feeds both the channel-1 power amplifier, and unity
gain inverter U8B. The output of inverter U8B feeds the channel-2 power amplifier.
POWER AMPLIFIER CIRCUITRY
The M•800 use a class AB triple darlington output stage, with complementary output devices.
The output stage has a voltage gain of slightly less than 1 and extremely high current gain. The
high current “output” parts (Q1-Q3 & Q11 - Q13) pull current from the +/- 64V supplies. Drivers (Q4
& Q10), Pre-drivers (Q5 & Q9), as well as the voltage-amp, are powered from the +/- 74V
supplies. Powering theses stages from the /-74V rails results in lower output stage
saturation voltages (less dissipation on the heatsink, so the amp is more efficient and runs
cooler) and improved linearity (intrinsically lower distortion). In the event of a catastrophic
amplifier failure, the triple darlington output stage is peppered with several fusible resistors to
minimize damage.
Q6 and Q8 are the outputs of the second stage of voltage amplification, and can be
thought of as current sources. These current sources are prevented from turning both positive
and negative current amplifiers on hard by the bias network. The bias network consisting of V-BE
multiplier Q7, and Buffer Q30, adjusts the voltage across C14 to a point where the output stage
just begins to conduct current (adjusted by the technician for 30mV across J5 the bias test
points). This “bias” current is needed to eliminate the conduction dead-zone that would
otherwise exist close to zero volts. This dead-zone is also referred to as crossover distortion. This
bias voltage across C14 needs to decrease as the output stage temperature increases. This is
why V-BE multiplier transistor Q7 is mounted to the heatsink. Without this thermal tracking, the
output stage would conduct more and more current as it heated-up, resulting in eventual
amplifier failure. This undesirable condition is commonly referred to as thermal run away.
Given the very high current gain of the output stage, if asked to, this stage can deliver
enough current to the load to destroy itself. To protect against this, VI limiting is employed.
Simply stated: if the output stage try’s to supply unsafe amounts of power, Q32 and Q35 divert
drive current from the output of the voltage amp (Q6 and Q8) that was meant for the pre-
drivers. Near zero crossing, if the voltage drop across either emitter resistor (R14, R53) gets greater
than .6V, then Q32 or Q35 conduct, limiting the output stage current. As the output stage gets
closer to the supply rails it is capable of sinking more current, so the drop across the emitter
resistors is divided down by R26, R27, R28, D21 and R75, R74, R73, D26. Also included in this VI
limiter is energy sensing, which is a fancy way of saying that for a short time (I.E. a musical
continued.