Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

555 Oscillator Frequency Fluctuating Rapidly

  1. Aug 9, 2009 #1
    hey all,

    I've wired up a standard 555 timer, using two resistors, a capacitor and a diode to make a clock in order to feed a stepper motor controller.


    Its wired like the schematic above.

    My resistor values are the following: R1=4.66K, R2=13.560K,C=2.2uF. This should give me a calculated clock freq of 36Hz with a duty cycle of 25.6%.

    But, when I measure the frequency using my DMM, its rapidly fluctuating. For instance, it'll be 38 one moment, and then 36, then 24 or whatever value it wants to be, really! Its quite random.

    Looking at the stepper motor, it does not appear to slow down or speed up, its rotating at a constant speed.

    What could be the cause of this rapidly fluctuating frequency? It only occurs when the controller is attached to the driver and the motor is running.

    Any advice would be massively appreciated.
  2. jcsd
  3. Aug 9, 2009 #2
    I'd watch the power source voltage. i think you need more current and good regulation.

    EDIT: or you might also be overloading the 555 output.
    Last edited: Aug 9, 2009
  4. Aug 9, 2009 #3
    Across what points are you connecting your DMM?
  5. Aug 9, 2009 #4
    I built a version of your circuit, and it runs OK. I didn't have the pots you specified, so I had to use fixed resistors. Sorry. See thumbnail.
    Is your DMM programmed to read frequency? If so, is your threshold properly set? Are you AC coupled?

    Attached Files:

  6. Aug 9, 2009 #5
    I'll get a fresh power source tomorrow and check again.

    Skeptic, I use point 3 to check my frequency.

    Bob S, I'm sure the circuit is good and yes, my meter does have a freq. feature but I'm not quite sure what you mean by threshold and AC coupling?

    The meter can detect from 1 Hz to 100 Khz, according to the specs.
  7. Aug 9, 2009 #6


    User Avatar
    Science Advisor

    And...? You need to connect both leads to get a proper reading. The red lead should be on pin 3 or wherever you wish to take a reading from and the black lead should be connected to ground, say, pin 1.

    Most typical DMMs with a built-in frequency counter don't have a threshold setting and are DC coupled. So, don't worry about those.
  8. Aug 9, 2009 #7
    Because pin 3 swings from a few millivolts (positive) to over 8 volts, your DMM frequency counter needs to be ac coupled, because there is no polarity change in the pin 3 output voltage. If you set your ac threshold (discriminator) voltage to about +/- 1 volt, you will not be sensitive to millivolt noise.
    [Edit] If you do not have an ac-coupled input on your frequency counter, use two 5 kohm resistors to make a reference point at 4.5 volts, so you can use DC coupling to pin 3. Alternatively, use a small capacitor (say 0.01 uF) to couple the frequency meter to pin 3.
    Last edited: Aug 9, 2009
  9. Aug 9, 2009 #8


    User Avatar
    Science Advisor

    There doesn't need to be. DMMs are almost all DC-coupled but they have no trouble counting frequency because they don't rely on zero crossings. Typically, they have counter threshold fixed at several tens to a few hundred mV.

    EDIT: of course, I just checked my Fluke 117C and it's AC-coupled for AC volts Hz but DC-coupled for AC current Hz. :rolleyes: So, might want to check your particular specs. In any case, whatever it is, it's generally fixed for a particular meter.
    Last edited: Aug 9, 2009
  10. Aug 9, 2009 #9


    User Avatar
    Science Advisor

    It seems like the problem is in your method of measurement rather than in the 555 circuit itself.

    Can you borrow an Oscilloscope? That would make things easier.

    Two things seem possible.
    You might be getting 60 Hz hum into the multimeter and confusing it.
    You have too much signal going to the multimeter. If this has even a small 60 Hz component, it may be enough to get counted in the frequency counter.

    Try to reduce the signal with resistors. Maybe a 10 K in series with a 1 K and take the output across the 1 K.
    Use short leads to the multimeter.

    Measure the frequency across the LEDs. This should be a smaller voltage than on pin 3.
  11. Aug 9, 2009 #10
    The 555 is powered by a 7805 regulator and in addition, the power source is also a 12V DC battery, so I don't think it would have a 60Hz hum anywhere at all... unless I'm missing something.

    But I'll try out your suggestion. Should have a recharged battery in a few hours too. Will report back here.

    Oh and, I forgot to mention, I actually do measure the freq. using two pins, Pin 3 and Ground.
  12. Aug 10, 2009 #11


    User Avatar
    Science Advisor

    Anywhere in a city a few inches of wire will pick up hum 50 Hz or 60 Hz, especially if the circuit is at high impedance.

    If you make up a cable for this, the resistors should be at the frequency counter end of the cable so that any hum pickup is attenuated. In any case, the cable should be as short as possible and shielded if possible.

    Also the 7805 is quite capable of oscillating if it is not bypassed properly. There should be 0.1 uF capacitors across from its input and output to ground.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook