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

Strange resistor voltage curve in an RC

  1. Dec 16, 2004 #1

    My data gravely perplexes me. I've set up a simple series RC circuit with an AC emf of 14V @ 20Hz. Resistor and capacitor values are 3.9E4 ohms and 4.7E-5 farads respectively. Ideally, the voltage across the resistor should be a simple sine curve, since this is basically an RLC series circuit with L=0. However, what I'm getting off of a digital voltmeter (outputting data to the computer) is a strange but pretty series of superimposed sinusoidal curves. A general pattern can be discerned over a period of seconds, which is what confuses me. I'm thinking the large pattern is caused by premature discharging of the capacitor midway through an emf cycle, thereby causing perturbations in the resistor voltage. The time constant may also be slyly manifesting itself, although i've recieved considerable opposition to this idea. Please try to destroy the interpretations put forth here.

    I can't seem to upload my graph bcoz it exceeds the upload size limit.
  2. jcsd
  3. Dec 16, 2004 #2


    User Avatar
    Science Advisor
    Gold Member

    Sounds like you might have 60 hertz noise superimposed on your readings. I realize this may be out of your budget, but a digital meter dumping data to a PC is no substitute for a scope. Many real-time scopes (non-storage) are coming down in price. You can also reasonably buy handheld scopes with LDC displays. Like I said, it may be out of your budget. As far as the cap discharging midway though the cycle, unless you are exceeding the voltage rating of the cap there is no reason for this to happen. I also wonder why you need to dump the data to the computer unless you are slowly sweeping the frequency. You will get ONE voltage across the components as long as the input voltage and frequency do not change.
  4. Dec 17, 2004 #3

    A digital voltmeter samples the waveform at whatever it's own rate is , which maybe anything , it is almost certainly not synchronous to your waveform , and so the samples will gradually cycle through your waveform and repeat(or nearly so ) at some
    other rate which is close to common.
    As the last contributor mentioned a scope is the only way to see this -- unless you create your own sampler using the computer ports and a software program to simulate a scope .. I have done this using Turbo basic for signals of 10's of KHz but it is rather tricky. There maybe available software to do this.
    If I have your units correct you are using a 47 micro farad capacitor
    which maybe an electrolytic -- if so the value of capacitance is dependant upon the DC bias and it should not be subjected to voltage reversal as at the very least will change the value observed.
    Last edited: Dec 17, 2004
  5. Dec 20, 2004 #4
    Wow, i forgot to account for that possibility. I hooked up the circuit to a dual channel oscilloscope right after Averagesupernova suggested the digitial voltmeter itself might be the case, and sure enough, it was. The oscilloscope gave perfect EMF vs. Resistor Voltage sine curves, altho the phase shift was barely noticeable (as it should have been). I guess the digital/computer equipment lost this battle! :smile:

    Thanks a lot for the replies.
  6. Dec 20, 2004 #5


    User Avatar
    Science Advisor
    Gold Member

    So basically you were attempting to sample a 20 hertz sine wave with a Fluke DVM. Hmmmm.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook