before we begin, it is important to know which pin is which and where
the top of the chip is. The picture above illustrates this. Just about
all inline chips, that is chips with a line of pins on both sides, are
numbered in this manner. To orient yourself, find a notch in the
plastic on the chip or a painted dot. These indicators mark the "top"
of the chip. On the left side, running top to bottom, the pins are
numbered 1,2,3... depending on the size of the chip. The chip pictured
above has 16 pins or 8 on each side. When you number to the bottom,
finish by numbering the rest of the pins on the right side from bottom
to top. That's all there is to it! Sometimes you may find documentation
for a chip you want to use that only mentions the function of each pin,
When physically building a circuit with a real chip,
rather than in theory on paper, you will have to know the pin numbers.
Be sure to check the datasheet of any chip you are unfamiliar with
first to prevent possible damage to your circuit or the destruction of
Some chips are static sensitive and can actually be "blow-out" or
destroyed by the static electricity that builds up on you. This can be
avoided either by wearing a ground strap, working in a special
anti-static workstation, or by avoiding touching the metal pins of the
is the schematic representation of the circuit you will soon be
building. It consists only of a single 4017B Decade Counter, 10 LEDs,
and 10 Resistors. (The LEDs are the green circles in the picture above
and the zig-zag lines are the symbolic form of a resistor, the circle
with a "+" in it is the positive end of the power supply, and the 3
lines forming an arrow pointing down is the symbol for ground or
negative end of the power supply/0volts) This circuit takes a clock
input signal through its CLK input pin and "sequences" the output
controlling the 10 LEDs. The effect is something like this:
*although the ones and zeros
might seem familiar, this output is not normally considered as the
binary number system. Click here to see why... BINARY
Don't worry if this schematic looks intimidating. As
you can see, many wires (connections/pathways, represented by the black
lines) over lap and look very messy. Such a drawing increases the
chances of the builder making errors. It is often acceptable for the
designer to draw a schematic in a block diagram showing the pin
numbers, but not necessarily in the same order as on the actual chip.
The following diagram is a block diagram schematic for the same circuit
as previously illustrated.
people, this form is much easier to understand and looks cleaner. Pin
numbers are still used to show how to build this circuit on a real chip
as well as a written description of what the specific pin does. For
example, pin 12 is not used (IN THIS CIRCUIT), but it's function is
still defined for clarity.
Now for an explanation of the circuit. Pins 8 and 16
power the chip. The 4017 IC typically can function within a wide
variance of voltages (3-16 volts). Be sure to check the datasheet for
the exact chip you are using before using 16 volts. Some versions of
the 4017 require less power. In this diagram 5 volts was used for the
supply. Pin 16 connects to the positive end of the power supply and pin
8 is ground.
The chip also has some configuration pins, which
give the user more design options, which will be discussed later in
detail. In this circuit the chip has been configured to count from 0-9
in order and then reset. Pin 13 enables the chip (ground=enable and
high=disable). Additional circuitry could be used to provide a pause
function by temporairly disabling the chip! Pin 15 is also grounded
which gives the chip its full range of counting. As soon as pin 15 goes
positive (high) the chip resets (output 0 goes high / LED 0 lights up
and all other outputs turn off). If this pin were not grounded the chip
would not count because it would be always reset. The designer can also
use this to create special effects. By connecting the reset pin (pin
15) directly to output 3 (pin 7) the circuit will count like this:
Notice that outputs 3-9 don't even light up! As soon
as output 3 goes high, the chip immediately resets. One of the chips
own outputs controlled its reset pin to produce a different effect. Can
you see how this could be used to build a digital clock??
Pin 12 --- When the counting sequence of the chip is reset this pin temporarily goes high ( for one clock cycle ) This pin can be connected to another 4017 chip to provide a "slower" counter. Each chip can divide the frequency of an incoming clock signal by a factor of 1,2,3,4,5,6,7,8,9, or 10.
Pin 14 is the clock input. When the clock signal
goes high the values of the output change. So if the outputs were in
this state: (100000000) and a clock signal came along the outputs would
change to this state: (010000000). The clock input is very important,
it tells the chip when to count and without it the chip would do
nothing. Almost all digital devices have some type of clock which
provides timing and synchronation.
So, how do you build a clock signal generator? Its
easy, really. All it is is a alternating signal (high,low,high,low
(1,0,1,0....) ) In principle this could be built using nothing more
than a push button switch, where in one position the switch is high and
in another the switch is low. However, as I said this can only work in
principle. Real switches are plagued by an inheriant property known as
contact bounce. At the microscopic level, think of someone fliping a
switch or pushing a button. The two metal contacts inside come closer
and closer (imagine this in slow-motion). When the contacts impact, the
force (although small) causes them to recoil and bounce. Its like
dropping a ball, when it hits the gorund it will bouce and then it will
be overcome by gravity and hit the ground again, however it will also
bouce again. In digital electronics this problem cannot be tolerated
because today's electronic circuits are actually fast enough to detect
these bounces. So if you were trying to make a counter that counts how
make times you pushed a button it would be inaccurate. Every press
could generate a different amount of bounces so the system would be
unreliable. The problem is solved by a process known as conditioning.
Capacitors and latch circuits can be used to provide a "debounced"
Most people, however, would want an automatic clock
signal generator, especially for this type of circuit when it would be
used in a digital clock or christmas tree chaser lights.
This is the
simplest way of building a clock generator. By attaching a wire between
the output of this "NOT GATE" to the CLK input of the 4017B decade
counter chip you would get a self incrementing circuit. But still,
there is a problem! There is no guaranteed way of knowing how fast it
will count (oscillate). There must be a way, considering people can
build clocks which keep time down to the second for years.
At this point you should realize that a clock
generator can be built using virtually any method. It could be a wheel
turned by water with a contact on it where it makes conatct with a
stationary brush once per revolution. What I'm trying to say is that
almost anything can be made to work. CLICK
HERE to learn more about making clock generators and even how to
make a 1.000000 second signal generator (great for digital time pieces)
Now its time to build the circuit, in reality...
(You are Here)
Further Projects for the 4017 are under constructions, check back later!