# 555 and 4017B

Discussion in 'Electronic Basics' started by Adrian, Jan 27, 2004.

1. ### AdrianGuest

Hey, for my school work i am using a 555 timer in astable operation to
provide teh pulse for a 4017B decade counter. I have it all working
properly however when i move my hand whithin about 15cm of the circuit
it pulses much faster - its not really a huge problem for my project
but im just curious to know what casues it. so if anyone knows i would
be interestd....thanks

2. ### John PopelishGuest

Is the circuit powered by battery or by an AC powered supply? This
can have an effect on how much AC hum is riding on your circuit.

I am suspicious that you may have left some input on the 4017
unconnected to anything, which allows that input to pick up any
electric field as a signal, altering the operation. Go over your
circuit and account for each pin on the chip that is an input. 555
also sometimes produce a glitchy output ( a very fast pulse preceding
the normal logic output. This glitch can be right on the threshold of
what it takes to clock the next chip, so that subtle influences can
make a difference. Put a 10 k resistor between the 555 output and the
4017 clock input to filter any glitch out, using the capacitance of
the input to complete the low pass filter.

3. ### Dominic-Luc WebbGuest

AC input for the 555??? Hmmm! John, you don't randomly shout nonsense
from my reading your many posts, so I know you have something in mind.
I am very curious what kind of circuits you might think would couple AC
to a 555.

Regarding the poster's question...

When I first started learning about the 555, my first IC actually, I
learned that some resistor/cap combinations on the 555 would prove
unreliable and I could get strange behaviors including changing
frequency, etc. It has been a while since I studied these in detail,
but I recall this happening when I chose very low resistance for
setting the timing. I believe both resistors had to be into the
thousands of Ohms and then I calculated the required cap based on
the equations in the spec sheets from the manufacturer. I am therefore
curious what frequency and what resistance/cap values you used to
get that frequency.

Dominic-Luc Webb

4. ### Dominic-Luc WebbGuest

One other thing, I have also had such problems when trying to drive
induction type devices with 555/4017 circuits without diodes between.
I don't suppose you are trying to run a motor, etc?

Dominic

5. ### John PopelishGuest

(snip)

I said, "AC powered supply". That is a power supply plugged into an
AC outlet. This could be a DC wall wart or a bench supply. But if it
is one that has no ground pin, it is possible for there to be quite a
bit if capacitively coupled AC riding on both DC rails. If you don't
believe me, plug in any DC output wall wart that has a line frequency
transformer and measure the AC voltage between either of its output
rails and the ground socket of the receptacle. Even if the supply has
the DC grounded on one side, the circuit has a current path for
capacitive current from other external line powered fields, especially
from fluorescent lights. Circuits powered from isolated batteries
have less problem with such capacitive currents because they have no
place to go.

6. ### AdrianGuest

hey, yea im using a DC power supply (9V) and the resistors im using are

R1 = 56 ohm
R2 = 56 ohm

and the capasitor is 2.2 uf

7. ### John PopelishGuest

Those low resistor values are working the grounding switch pretty
hard, but they are way to low in impedance to be directly affected by
the electric field changes caused by waving at the circuit. Have you
looked at the output of the 555 to see if it is changing frequency or
only at the 4017. At this point I still think a floating input on the
4017 is the most likely culprit and the 555 output glitch is the
second most likely.

8. ### Si BallengerGuest

You probably need to tie the clock line to ground on the chip
with a 10k or 22k resistor to keep it from floating. I made
something similar (below) and it would start switching when the
cat got within a couple of inches until I added the resistor.

http://www.geocities.com/zoomkat/RSswitcher.htm

9. ### JeffMGuest

using a 555...to provide [the] pulse for a 4017B
I was thinking the Control/Modulation pin of the 555.
Got it bypassed with a capacitor?

10. ### Dominic-Luc WebbGuest

I for sure had problems with some of my 555 circuits with
somewhat higher resistances and these problems were resolved
once I switched to values in the thousands of ohms. Also, as I
recall, is it not so that these two resistors should never have
same value? I think I saw this in an application sheet. There
is also the famous "555 Timer Cookbook" by Walter Jung one
could consult. I recall trying to keep them about 2 fold different
values or more. I am now curious what frequency the poster is trying
to get. Perhaps there are some preferable values that could be used?
Perhaps the equation can be recalculated for resistances of at least
a couple thousand to yield a new cap value giving same frequency?

I think there was even a post here some years back in which this
problem was even discussed by someone who seemed to know the
internals of these timers pretty well.

Dominic-Luc Webb

11. ### John PopelishGuest

Dominic-Luc Webb wrote:
(snip)
(snip)

There is no such requirement. The resistors have to be low enough to
provide bias current for the differential amplifiers that measure the
capacitor voltage and low enough to not overload the grounding
(discharge) switch. That is all.

12. ### Dominic-Luc WebbGuest

Adrian,

Are you trying to get a frequency of f = 1.44 / [(Ra + 2Rb)*C] = 3896 Hz?

Calculating a little, I think for such frequency I have used
(successfully):

R1 = 2 KOhm
R2 = 18 KOhm
C = 10 nF (note this is nanoF)

One of my R's was a 25 turtrim potentiometer, maybe a 47 KOhm.

Dominic-Luc Webb

13. ### BalajiGuest

I think there is something wrong with that notion. You see, the values
of these resistances determine:

a. the time taken to charge the capacitor up to the threshold and back
to the lower limit. The time period of a 555 timer ckt connected as an
astable is T = ln2 * (R1 + 2*R2)*C. In this case, we get a frequency
of 3.9KHz. When you design a ckt with a 555 make sure you choose R's
and C's such that your frequency does not exceed 100KHz. (Try to keep
within 80KHz) If you need higher frequency, use the pin compatible
7555 IC. (ln2 = 0.693 approx)

b. the amount of current flowing into the pin 7 when the output goes
low (or pin 7 starts sinking huge currents) This should be kept
generally to a minimum. A value of about 100 - 150mA should be about
the maximum allowed (check datasheets here ... i don't remember that)
since the transistor inside then goes into saturation. How to find if
it is limited in a design - (Vcc - 0.2)/R1, where R1 is the resistor
connecting the Vcc terminal to pin 7. Check up the datasheets first,
and then see if you have exceeded that.

c. differential input to the comparators consume some current. Check
that from the datasheet. It should be around 50nA.
Now, that factor of 2 is because, if you have a CRO and if you observe
the output of the timer, it remains high for more time than it remains
low. In fact, the time it remains high is given as TH = ln2 * (R1 +
R2) * C, and goes low for TL = ln2 * R2 * C. Now if we have R2 ~ 2*R1,
we get TH ~ 2.1 * R1 * C and TL ~ 1.4 * R1 * C.

See that if this ratio between R1 and R2 is increase further, the duty
cycle (fraction of time period for which) the output is high
increases. We generally prefer 50% duty cycle, but here it does not
matter, since no time bound operations are done in the rest of the
ckt. during the stable states of the timer. Besides, since 4017B is a
CMOS IC, it does not consume power during the time the input is in any
stable state - it consumes power only during transitions in logic
levels.
That's partially correct and others have posted pretty good values of
resistances and capacitances. Make sure the capacitor is a ceramic
one. At any cost do not use a electrolytic one for this purpose.
Tantalum may do fine - check that up - your own risk. (Big deal!?)
The problem being faced here is mostly that of what is called EMI
emission. Although the effect of EMI is not generally so strong, for
circuits powered with such a high voltage, it may be powerful. It will
surely be powerful if

a. you have used an electrolytic capacitor somewhere in the circuit,
where it should not be. An electrolytic capacitor is allowed at the
power supply to regulate and get a clean source.

b. you or some part of your body is touching or is very close to a
mains socket. Even a power source of 120V 60Hz is enough to cause
fluctuations due to EMI emissions.

c. Finally, the layout and design of the PCB is poor. To learn
designing good EMI emission protection in PCBs, look up google or get
this free book from Cypress Microsystems - "Perfect Timing". This
deals with many aspects of timing and board layout and EMI.

In fact this principle of EMI emissions is what is used to detect the
presence of a working socket without touching it or detecting a
concealed wire inside a wall. Many such circuits are quite popularly
known.

-Balaji

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Continue to site