
13
To further insure stability of the amplifier, the output node is de-coupled from the load via L5,
R187-R189. The values of these parts have also been selected to ensure good transient response
(i.e. square waves look like square waves, so the amplifier doesn’t sound unnaturally bright). The
output stage also sees a defined load at high frequencies, made up of R181 and C75. D17 and
D18 protect the output stage from situations when energy from the speaker (counter-EMF in the
speaker’s voice coil) might try and drive the output note above the +64V supply or below the
-64V supply. These parts are referred to as flyback diodes.
All things above, meld together to make for a better sounding amplifier when not driven
into clipping. When driven into clipping, the above mentioned Baker clamps, and the smaller
amount of negative feedback, also allows the amplifier to sound better than the competition.
After coming out of clipping the amp immediately returns to normal operation, faithfully
following the musical input signal. Other amps, of more conventional design, take a finite
amount of time to “catch up” before they begin to follow the input signal again. Mackie refers
to this phenomenon as “latching”. The “FR-Fast Recovery series” isn’t just marketing buzzwords: It
really does result in a more reliable, and better sounding product, with “good” performance
numbers.
The power amp’s limiter circuit is implemented with an LED / LDR (Light dependant resistor)
opto-coupler, and the Baker clamps. Referring to the positive Baker clamp: when the amplifier
clips, D1 and Q52 forward bias. This current originating at the collector of Q6 also flows out the
collector of Q52. The current flows through the LED in the opto. (U2A) which in turn decreases the
LDR’s resistance (U2B). Since R146 and U2B form a voltage divider, as U2B’s resistance decreases,
the drive to the power amp decreases (limited). Negative clipping activates the LED in the
same fashion through Q53. The limiter is defeated by shorting together LIM1(1) and LIM1(2) at SW3
which shorts across the U2A LED.
In the event that the power amplifier does fail, Crow-bars protect the speakers connected
to the output terminals. Referring to Page-3 of the 212 schematic: Q74 is turned on a few
hundred milliseconds after DC is present on the output line. When DC is present, R140 and R169
begin to charge C68. When the voltage across C68 gets to around 10V, silicon bilateral switch
Q72 turns on, dumping current through R183 to the Gate of Q74, turning on Q74, and shorting
the output line to ground.
THERMAL MANAGEMENT
The M•800 T-Design Heatsink/Fan cools output devices evenly, and does not collect dust on
the circuitry. The fan operates at two speeds, controlled by the amplifier. An LM35DZ, mounted in
the center of the heatsink, provides temperature information to the fan control and overtemp
circuits. There is also a thermal breaker in the power transformer.
FAN CONTROL CIRCUIT
The fan runs at two different speeds. When running fast, around 27V is across the fan. Voltage
is supplied to the fan by follower Q14 and the reference for this follower is derived from D59, D46
and R137. When running slow, around 15V is on the fan. In slow mode Q60 turns-on, acting as a
saturated switch, shorting out D59.
The fan speed increases in the following instances:
·
At turn on. To give the Fan an extra “goose” to get it spinning. At turn on Pin-11 of U5
(non inverting input) is lower that pin-10 (inverting). This makes the output (Pin-13) low
and Q60 turns off. The fan runs fast. A half second later, C82 charges enough to allow
pin-13 to go high, turning on Q60 and making the fan run slow.
continued.