# Finding the bias current in transistor circuits.

Discussion in 'Electronic Basics' started by Kristine Hyvang, Jan 22, 2004.

1. ### Kristine HyvangGuest

I'm about to build a small power amplifier using bipolar transistors. I
think it is called a class AB amplifier with a push-pull concept, using the
transistors 2N1613.

My problem is: What should decide the bias current (the current running
through both transistors when input is zero), and how does that circuit look
like ?

Maybe there is another newsgroup for this discussion (I found a great one,
but it was in Dutch ?! lucky them !! )

2. ### Bob MastaGuest

It's been a while for me, but the basic idea is to keep the
bias current as low as possible to just eliminate crossover
distortion. You can adjust this with a scope, or even by
ear if you are careful. (Use a pure sine at about 400 Hz
so any distortion products will be in the most-audible

If you get it too high, you will go into thermal runaway
in very short order, so use caution.

The standard bias regulator that I recall was a simple
one-transistor regulator between the bases of the
positive and negative drivers. This regulator had a
small resistor in the emitter, and the base went to
the wiper of a pot between the collector and the
far end of the emitter resistor. It just forces a constant
voltage between the top and bottom.

The transistor must be in thermal contact with an
output device.

Simpler approaches that I've seen in cheap amps
are to just put in a couple of diodes, but you have
to err on the low side to stay away from thermal
runaway. I've also see ordinary resistor strings used
here.

In the "bad old days" some of these cheesier methods
were tolerable because the overall amp had huge
amounts of overall negative feedback to reduce the
distortion. We now know that tends to produce
Transient Intermodulation Distortion, so global
feedback is more restrained.

Bob Masta

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com

3. ### Robert C MonsenGuest

There is a short discussion of this in "Art of Electronics", chapter 2. The
idea is to choose a quiescent current that is a 'compromise between low
distortion and excessive quiescent dissipation'. They go on to mention that
using feedback can be used to reduce distortion.

The circuit is a push-pull output stage with small resistors between the
emitters and the output. The bases are kept at 3 or so diode drops apart.
The quiescent current is determined by a diode drop across those small
resistors. So, if they are, say, 1ohm each, then the quiescent current is
700mV/2 = 350mA. Thus, using different size resistors, you can pick the
quiescent current.

Regards,
Bob Monsen