had been only a little over a decade since the transistor was invented
when this article appeared in the August 1959 edition of Popular
Electronics. Transistors were still a mystery to most people, including
engineers, technicians, and hobbyists. Author James Butterfield
takes a unique approach in presenting the material by writing it
as a dialog between an instructor and a student. If you are also
new to transistors, this will be worth your while to read.
August 1959 Popular Electronics
Wax nostalgic about and learn from the history of early electronics. See articles
published October 1954 - April 1985. All copyrights are hereby acknowledged.
Understanding Transistor Circuits
A common-sense guide to using transistors
Canadian National Telegraphs
PART 1: Ohm's law and the
Pete introduces Johnny to the grounded emitter circuit.
Its "why's" and "how's" are investigated and the not-too-mysterious
matters of base and collector bias are discussed. The rule is, as
Pete explains it, "Let Ohm's Law be your guide."
PART 2: Design "Do's" and
Trying to design a one-transistor amplifier, Johnny
gets tangled up with input and output impedance problems. Pete straightens
him out with some important "do's" and "don'ts" in getting a signal
in and out of a transistor.
PART 3: Controlling, Coupling.
The transistorized approach to filters and tone and
volume controls is detailed by Pete with some practical suggestions
on the testing of transistor characteristics. Coupling problems
are discussed, and Pete draws up a chart that matches the coupling
capacitor to the collector resistor
Ohm's Law and the Transistor
should do it," grunted Pete, as he soldered a final connection.
"Now, let's tryout the amplifier."
"That's an amplifier?"
snorted Johnny. "You must have dreamed up that circuit during a
fit of indigestion! Anybody knows even the simplest transistor amplifiers
are bristling with resistors and oversize coupling capacitors. And
where's the output transformer for the speaker?"
need any." Pete clipped on the ground lead from the audio generator
and inserted a .1-µf. capacitor in series with the other lead. "Trouble
with most of you experimenters is that you just copy diagrams -
maybe change 'em a little here and there to see what happens - but
you never bother to find out what really makes things tick." He
touched the other end of the capacitor to his input lead, and the
speaker sang with a clear tone.
"Pete, I've got six books
on transistors, and I still don't know what I'm doing," complained
Johnny. "They are either so simplified that they don't say anything
- or they're crammed with equations from stem to stern,"
"Nothing wrong with equations." Pete was busy tidying up his
workbench. "But you don't always need 'em. I threw this amplifier
together without fancy calculations - just horse sense. Let's go
into the kitchen for coffee, and I'll try to fill you in on the
practical side of transistors."
Diodes Back to Back
off," said Pete settling in a chair, "You know that a transistor
is sort of like a pair of diodes, back to back, like this." He sketched
rapidly on a napkin.
"The way I've shown it here, the transistor
is a p-n-p unit. To get an n-p-n job, you flip both diodes end-for-end.
We'll dig deeper into this later-but for now, just remember when
positive voltage is on the p section and negative voltage on the
n section, the transistor conducts."
"I suppose the lead
in the middle is the base," said Johnny. "But which is the collector
and which is the emitter? Your diagram doesn't show any difference."
"Here's something that will probably surprise you, Johnny.
The collector and the emitter could be changed around and the transistor
would still work! But since the manufacturer normally builds the
collector bigger and stronger than the emitter - take his advice
and use the collector lead as the collector.
"The real difference
between the collector and emitter is in the way we treat them. The
emitter is biased to allow maximum current flow (forward bias) and
the collector, on the other hand, is back-biased for minimum current."
Johnny cut in. "Wouldn't a heavy current flow from the base through
the emitter mean that there would be darn little voltage drop between
the emitter and base?"
"Right, Johnny, there's seldom more
than a quarter of a volt difference between emitter and base, which
makes it easy to figure emitter current. Take this circuit for example:
"The collector isn't important just for the moment, so I
won't show any wiring to it. Since the emitter is at ground, you
can figure that the base is almost exactly at ground, too. This
"I get it," cried Johnny. "The battery puts exactly
three volts across the 1500-ohm resistor, so Ohm's law says that
the current must be ... ah ... two milliamps !"
chuckled Pete. "And since this part of the transistor is conducting
in the forward direction, the current will not be affected by the
collector. By the way, what type of transistor did I show in the
"Negative voltage on the base, and positive on
the emitter ... so the transistor must be a p-n-p!" exclaimed Johnny
"Now, here's point number
two. The collector must be biased in the non-conducting direction.
How would you do that?"
"Put negative voltage on it, I suppose,"
said Johnny thoughtfully. "But, wouldn't that mean that no current
would ever flow?"
would be true if the collector and base only were connected," replied
Pete. "But when the emitter is in the circuit with positive bias,
an interaction allows current to flow through the collector."
"Now you're getting complicated," complained Johnny. "You
sound like a textbook."
"Okay, let's put it this way. Whatever
current flows through the base will produce an amplified current
through the collector. You know how to find the base current - we
just did it by Ohm's law. So you take the base current (Ib),
multiply it by your amplification factor (Hfe) - and
there's Ic, your collector current."
a minute," argued Johnny. "Suppose I put a higher voltage on the
collector. Wouldn't that make more current flow?"
all! Ignoring the fine points, Ic will always equal Ib
x Hfe. The latest G.E. transistor manual, by the way,
lists Hfe for over four hundred transistors of all makes.
"The experimenters' transistors, such as the 2N107 and the
CK722, have gains of about 20. Commercial grade transistors can
run gains as high as 100 or more.
"Let's take a look at
another circuit," Pete continued, sketching rapidly. "See what you
make of this."
me see," murmured Johnny. "With the emitter grounded, the base has
to be at almost zero volts - which will cause the battery's three
volts to drop across the 33,000-ohm resistor. That would give about
0.1 ma. flowing through the base. If the amplification of the transistor
is 20, that means the collector will take almost two milliamps.
But, doesn't the size of resistor R make any difference to the collector
current at all?"
"Nope," said Pete cheerfully, "with one
exception. What would happen if that resistor were, say, 2000 ohms?"
"Well ... Ohm's law says that two milliamps times 2000 ohms
makes a drop of four volts. Oh, I see what you're getting at. If
your collector resistor is too big, it won't pass any voltage ...
"And once you lose the voltage, you lose your transistor
action," finished Pete. "A resistance of about 800 ohms would be
dandy for this circuit."
"That would give a collector voltage
of - let's see - three volts minus a 1.6-volt drop across the resistor
leaves a little under a volt and a half for the collector. Isn't
that too low? You'd never be able to put out a decent signal voltage
with a supply like that."
"Depends on what you're using
the amplifier for. Usually, you don't need voltage - all you want
Beating the Heat
"Another thing, Pete, what about heat effects? Are transistors as
heat-sensitive as some of my books say? Does the gain change, or
"No, the gain of a transistor usually stays
fairly steady; but when the thermometer is high, a little leakage
current sneaks through. Don't worry about it, though. You'll seldom
have trouble if you stay well within the maximum ratings of the
"These 'max' ratings - anything complicated
"Gosh, no! Maximum collector-to-base
voltage, maximum collector current, and maximum collector dissipation
- that's all you'll have to worry about. They mean just what they
say. Don't put too much voltage across a transistor; don't put too
much current through it; and don't heat it up too much with a combination
of the two.
"Now - you've made a start on logical
transistor design. The fine points will come later. Whomp yourself
up a few circuits. You'll find that theory and practice go hand
in hand. Then when you think you've got the basic idea under your
hat, we'll dig into this thing a little deeper."
Design "Do's" and "Don'ts"
I've been playing around with transistors like you told me, and
I've got problems."
Pete carefully set down his soldering
gun and turned toward his visitor. "What seems to be the trouble,
"Well ... I wanted to boost the output of the radio
I'm using as a tuner. I laid out a circuit the way you showed me,
but it doesn't seem to work right. The volume is just as low as
before - and what's worse, the tone is mushy and distorted. I tried
a new transistor with no better results. Pete, tell me, where did
Pete grinned. "I think I know what the trouble
is, Johnny. Your circuit is probably good for a vacuum tube. You'll
have to learn to think in different terms for transistor amplifiers.
"First of all," said Pete, reaching for a pencil and scratch
pad, "do you remember how I told you that a transistor was really
a pair of diodes back to back?"
"Yes. I got that part okay."
Now let's go back to fundamentals again and attack the problem from
a different position.
"Remember that the base-emitter 'diode'
must be hooked up to conduct in the forward direction. Let's look
at the circuit of an ordinary transistor amplifier.
I told you last time, in a p-n-p transistor negative voltage is
applied to the base and positive voltage to the emitter, and conduction
takes place in the diode formed by the base and emitter. I'll sketch
this part of the circuit in detail for you.
"When a diode
conducts, it's just as if a switch has been closed. Now, when this
base-to-emitter diode is conducting, all your a.c. signal goes directly
through the diode to ground. For most purposes, the input impedance
is so low that you can consider it zero."
"Just a minute,"
interrupted Johnny. "If the input is a short circuit, how can you
get any amplification? There won't be any voltage left at the input
if you short it all out."
"Forget about voltages. Here's
the important thing. Since the transistor has a 'short-circuit'
input, it may not have much voltage at the input, but current flows
into the transistor. Current is what a transistor amplifies.
we have a transistor with a beta - that's current gain - of 20.
Then if I put one milliampere of signal into the base, I'll get
20 ma. at the collector. Write this in your notebook in capital
letters an inch high - A TRANSISTOR AMPLIFIES CURRENT."
"Pete, I notice you always put your signal into the base and take
your amplified output from the other collector. It seems to me I've
heard of other ways of connecting a transistor."
them, Johnny. There are very few jobs that can't be handled by the
grounded emitter. We'll talk about the other circuits when an occasion
arises that requires their use."
how do I go about using an output that's rated in milliamps instead
of volts? The current goes through the collector resistor, so it
can't be used as it stands."
"That's not exactly true, Johnny.
It depends on how you hook into the transistor output.
we put the output directly into another transistor stage," continued
Pete. "Since the input of this stage is a short circuit, all your
signal will be 'shorted' directly to this stage through the coupling
capacitor. In other words, all your output goes to the next stage.
"If both these transistors have a current amplification
of 40, we can do some fairly accurate figuring. An input of 0.5
ma. will produce 20-ma. output. All of this goes to the second stage
to be amplified, giving you a final output of 800 ma. - almost a
Johnny exclaimed. "That's a lot of current!"
"A little too
much current for comfort. To pass that amount of current, even in
a power transistor, you'd have to be careful of heat effects.
This doesn't just mean mounting your transistors to dissipate
heat - it means using special circuits to compensate for heat effects.
"For the time being, with the simple circuit we're using,
it would be wise to keep your output signal to a quarter of an ampere
or less. At this level, you can bring your d.c. bias down to about
a third of an amp collector current, which is well within safety
limits for a low collector voltage. If you use more than a couple
of volts on the collector, keep an eye on your power ratings. And
use a heat sink, of course."
I still don't see how we're going to use that final output current.
Do you connect a coupling capacitor to the speaker or what?"
"You could do that, but you'd waste a lot of power. The easiest
way to do it is to put the speaker directly in series with the collector,
like this. But if the collector load is too small, there won't be
enough voltage drop across it for appreciable power to be transferred
to the speaker. Ohm's law, W=I2 R, (where W is the wattage
delivered to the speaker, I is the collector current, and R is the
speaker impedance) explains the problem.
"Get a transformer
whose primary matches the collector load requirements and whose
secondary fits the speaker, and you're in business."
"Let's get back to your original
question on figuring your output. If you're using a transistor as
a preamplifier for a vacuum tube, you'll have to know your voltage
out. This is really very easy when you think about it, because the
input of the vacuum tube is usually high enough to ignore.
this diagram, the total transistor load is the 5000-ohm collector
resistor shunted by a 1/2megohm volume control. For all practical
purposes, our output load is 5000 ohms. If our design shows we have
a signal current of one milliampere on the collector, we use Ohm's
"Don't tell me!" exclaimed Johnny. "That will be
... ah ... five volts!"
"You're sharp this morning. May
I point out that if you want to use a transistor in this way - and
handle these levels - it would be common sense to put at least seven
volts or so on the collector?"
"I see what you mean,
Pete. And I suppose it would also be a good idea to have at least
one milliampere of d.c. current through the collector?"
"Right. Remember, of course, that a milliamp of a.c. will draw almost
1.5-ma. peak current, so you should allow a little extra. And that's
just for pure sine waves - when you're figuring in terms of average
music levels, leave lots of room for swing.
it's best to choose your d.c, bias currents keeping in mind the
a.c. signal voltages you want to handle," Pete went on. "And usually
it's best to have at least 1 ma. of collector current, even at low
levels. The reason for this is complex, having to do with impedances
and distortion. We'll save this topic for another time."
N-P-N or P-N-P
probably noticed that all the transistors I've shown have been p-n-p
types. As far as signal amplification goes, n-p-n transistors are
exactly the same. Your signal goes into the base and comes out from
the collector. Your connections for bias will be different, but
that won't affect your signal.
"If you use both types
together, one emitter goes to positive ground - that's the p-n-p
- and the other emitter will go to your negative voltage supply.
This is quite okay, since they are connected together signal-wise
through the filter capacitor, making both emitters 'grounded.'
"Your bias currents are easy to set up on any of these
stages. I've left the base and collector resistors unmarked, for
two reasons. First, the size of these resistors depends on the battery
voltage you are using. And, second, it will do you good to work
out the values yourself using the techniques I showed you before."
Getting the Signal In
discuss the booster you tried your hand on, now. Where did you connect
"Across the volume control. I wanted to
be able to turn the volume on the radio down and still feed my hi-fi."
"Well, let's sketch out a typical receiver circuit.
Notice that the detector is designed to work into a very high resistance
"I see !" exclaimed Johnny. "The transistor booster
would short out the volume control, and put a heavy load on the
detector circuit. Well ... how can I get the input impedance high
enough so that it won't affect the circuit?"
"That's a story
in itself, Johnny. The best way is usually to put a resistance in
series with your input. But even though your radio isn't an a.c./
d.c. job, you'll probably have hum problems if you try using a common
ground. For the moment I'll say this: the best way will probably
be to use an input transformer that matches from 500K down to about
"In the meantime, dust off some of those old transistors
and see if you can put them to work. A lot of design 'failures'
result from the experimenter forgetting to take into account the
very low impedance of transistors.
"Look me up when you've
digested this session and I'll show you how to check out some of
the transistor design parameters, and incidentally how to check
out some of the transistor design parameters, and incidentally how
to test the little gadgets."
Coupling and Testing
It was a week later
that Johnny once again sauntered down the stairs into Pete's basement
workshop. Pete looked up from a small amplifier he was wiring. "How's
"I'm just a little confused, Pete.
All this dope you've been giving me on transistors is hard to digest
so fast. For example, you said that the input is a short circuit.
Well, the Radio Amateur's Handbook has a table of specs for quite
a few transistors, and for the 2N107 they say ... "
"They told you the input impedance was around 600 ohms, I bet. Well,
in a sense we're both correct. Here, Johnny, look at it this way
- you've done a fair amount of wiring. How much resistance would
you say there is in hookup wire?"
"I never thought
about it. Oh ... maybe a half an ohm a foot ... but it doesn't make
much difference to the circuit."
"Why not?" demanded Pete.
"Because even if the resistance were as high as an ohm or two,
it's usually so much smaller than any other resistance in the circuit
that you don't count it. Who cares about half an ohm or so when
you're dealing with resistors of several thousand ohms?"
"Right you are, Johnny. And the same thing goes for your question
about a transistor input impedance. Even at six hundred ohms, it's
so much smaller than any other resistance in the circuit that you
can count it as a short."
Short and Simple
very simple to figure out a circuit this way, too. Using the short-circuit
input approach, you eliminate most of the calculating."
"But wouldn't that make your calculations inaccurate? You couldn't
get the exact values of gain. You'd have to be a few percent off."
"Look at it this way, Johnny. To begin with, transistors
are never rated exactly. A transistor with a rated gain of 20 might
have an actual gain as low as 10 or as high as 40. So, fancy calculations
usually aren't worth the trouble."
Johnny nodded. "True.
But let me draw you a circuit. Now, if the voltage supply were very
low - such as a single dry cell - then the collector resistor R
would have to be small enough so that it didn't drop all the battery
voltage. Wouldn't that mean ... ?"
Pete chuckled. "Johnny,
you're too sharp for me today. Yes, that would mean your small collector
resistor would prevent some of the signal from getting to the next
stage. Not only would amplification be poor, but the low resistance
would call for some mighty hefty coupling capacitors. And to put
another fly in the ointment, as soon as your signal began to get
lost in the resistor, distortion would increase sharply. But don't
worry - unusually low voltages are almost the only cause of such
"Pete, you lost me. I guess I still don't
dig this whole low impedance input idea."
"You've been brainwashed
by vacuum-tube theory, Johnny - you've got to learn to rethink the
problems. Let's look at some volume controls."
vacuum-tube circuit volume control transferred to a transistor circuit
would look like this. Whereas a tube draws almost no grid current,
the base input of the transistor does - and right through the control.
You can see how moving the "pot" arm up and down is not only going
to tap off a part of the signal current, which is what we want it
to do, but unfortunately it's also going to change the value of
the base bias resistor. A volume control with no problems looks
"If you run into volume control problems such
as distortion at low volume, abrupt changes in volume as you move
the control, or just excessive noise from the control, check to
make sure that you're not goofing up the bias in some way and that
there isn't too much current flowing through the control."
Capacitors - Filtering and Coupling
find that transistor filter circuits, too, have to be approached
differently," Pete continued. "Here's a typical tube low-pass filter.
A transistor input following the filter would short out the capacitor,
making it useless.
"We'll redraw the circuit for transistor
use like this. It could function as a scratch filter or a simple
equalizer for a magnetic phonograph cartridge. The values, of course,
depend on the use."
"Could I put two filter sections together,
like this?" asked Johnny.
"You sure could. That will give
you a sharper frequency cutoff."
"Looks like all you have
to do to adapt a vacuum-tube circuit for transistors is turn it
end for end."
partly right, Johnny. But it applies only to coupling circuits between
transistors, and it won't always work. Remember to make the resistances
smaller and the capacitances larger than with tubes. And check out
your circuit practically after you work up the design."
"Speaking of capacitors, Pete, I've been meaning to ask you: how
do you figure the size of your blocking capacitor?"
to your collector resistor, usually. You know the formula - it goes,
'C equals over ... " "Ouch! Spare me those formulas!" "Tarnation!"
grumbled Pete. "If you ever want to get anywhere in electronics,
you'll have to start using formulas. Things are getting too complicated
for anyone to get along with a wet finger and a screwdriver.
"This time I'll give you a break. Here's a table for you. I've
rounded off the values for a frequency response down to about 20
looked at his watch. "Pete, I'll have to run soon. But first, you
told me last time that you'd give me the low-down on how to test
"Well, it's fairly easy. A nice thing about
transistors is that they usually don't gradually deteriorate-they
burn out. And you can spot almost all faults real quick by two simple
tests: leakage and gain.
makes the transistor easy 'to check - no fancy testers needed. Just
select n-p-n or p-n-p, check for leakage and current gain ... and
that's it !"
"Just how do you make these tests, Pete?"
"Easy. Let's take leakage. If I leave the base of a transistor
disconnected, current can't flow through the other two leads. The
transistor is just two diodes, back to back; so one of them will
block the current flow.
"Practically, this isn't quite true.
A small leakage current will flow - usually less than a tenth of
a milliampere for most low and medium power transistors. A reading
over 1 ma. means your transistor has 'had it'.
"So, to test
for leakage, just place about six volts across the emitter and collector,
using the normal polarity for the transistor type. A milliammeter
with a range of anywhere from 1 - 10 ma. will do the trick. And
for the protection of the meter, it's a good idea to add a 2000-ohm
series resistor. This won't affect the accuracy of the test.
"Testing for gain is just as simple. Think back to our first
talk. Whatever current is put into the base of a transistor appears
multiplied by the amplification factor (beta) in the collector circuit."
"Of course," exclaimed Johnny. "All you have to do
is put a milliammeter in the base circuit, another in the collector
circuit - and you can see your amplification !"
can do it even easier than that," replied Pete. "With an exact known
current in the base, you need only one meter in the collector, which
you can calibrate to read directly in gain, if you wish. The ratio
of the collector current to the base bias current gives you your
gain figure. An open element, of course, gives you a zero - gain
reading, as does a base-to-emitter short.
"If a transistor's
beta reads low, don't throw it out. Remember, you're not working
with vacuum tubes where low gain usually means that the tube is
"Power transistors are checked in the
same way - but you'll find much more leakage, sometimes as high
as 10 ma. Because of this, it's easiest to check this type at much
higher current levels.
"Transistor testers are quite
inexpensive. Kit models are available, and it's a lot easier to
build yourself a finished tester than it is to rig up a circuit
every time you want to check a transistor.
thing. You'd better learn how to use a little math! Those formulas
are there to help you. Dust off an old textbook, and ... "
"Yes, sir! Just as you say, sir! I'm on my way right now!" Johnny
seized the pile of circuit sketches and ran up the stairs. He waved
back. "Thanks, Pete. See you later !"
Driver Stage of the above amplifier
is mocked up on this months cover
This class B audio amplifier
will deliver 8 watts output with less than .5 watt of audio drive.
T1's impedance primary is chosen to load the previous stage properly.
Collector current in the first stage is set at 70 ma. at a with
R1. Pot R2 is adjusted for 100 ma. at b. Both output transistors
require the use of large heat sinks for proper operation.
Code Practice Oscillator
transistor version of the Colpitts oscillator is used here as a
code practice oscillator. The 1- megohm tone control may be adjusted
for the most pleasing tone. Ordinary magnetic headphones are used
and serve as an inductance in the oscillator circuit.
Pocket Regen Receiver
detectors are easily adapted to transistor radio use. In this circuit,
coil L1 is a standard transistor antenna coil and the feedback winding
(L2) consists of 10 - 15 turns of wire wound around the center of
L1. The 1-megohm potentiometer is the regeneration control and should
be set just below the point where whistles or "motorboating" are
heard in the headphones. If no regeneration is heard, reverse the
connections to tickler feedback coil L2.
of the better broadcast-band receivers include an r.f. stage for
increased sensitivity. Extra r.f. amplification can be added to
almost any receiver without any complicated rewiring by means of
a simple r.f. preamp such as the one shown here.
High-Frequency Xtal Oscillator
above is a general-purpose crystal oscillator for use at frequencies
up to 20 mc. designed around a 2N94A transistor. Approximately 10
milliwatts of r.f. output are developed at the crystal frequency.
The low-impedance pickup coil (L2) consists of two or three turns
wound close to the lower end of L1. Coil L1 and C1 are chosen to
resonate at the crystal frequency.
is a code keyer unit designed to eliminate the key clicks and thumps
that may be cluttering up your transmissions. With the key up, the
transistor is cut off, as is the tube being keyed. With the key
down, current drop across the collector-emitter circuit is about
0.1 volt, and you're on the air. Before adding the keyer unit to
your transmitter, make these checks. 1. Connect a 20,000-ohm
voltmeter across your transmitter key jack. If the voltage is greater
than 25 volts, it may be necessary to reduce the screen voltage
on the keyed tube to 150 volts or less. Use a voltage-regulated
supply or voltage-divider network rather than a simple series resistor.
2. Measure the current to be keyed by placing a milliammeter
across the key. This current should not be much greater than 50
ma. 3. If it was necessary to lower the screen voltage, check
the r.f. output. If the output is too low, return the screen voltage
to normal and use two complementary transistors in series back-to-back,
as shown in the alternate circuit diagram. Note that the keyer should
never be left in the key-down condition.
Dynamic Microphone Preamp
high-grade dynamic microphones have too low an output to be used
with tape recorders or p.a. systems. A simple self-contained preamp
can be designed using one transistor which will boost the output
of these low-level microphones sufficiently for almost any type
miniature hand-held wireless microphone can be constructed using
an r.f. transistor in a Hartley-type oscillator. Good short-range
results can be obtained on the broadcast band with a 2' whip antenna.
Follow the manufacturer's coil terminal numbering shown or oscillation
will not occur. Adjust the coil and trimmer for best reception on
a clear spot in the broadcast band.
Sensitive Photo Relay
two-stage direct-coupled circuit offers high sensitivity in photocell
applications. A 1-megohm pot is included for sensitivity adjustment
over a wide range. The relay is a sensitive Advance Type SO/1C/4000D