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

LRC Circuits

  1. May 16, 2004 #1
    I'm attempting to construct a demo using an LRC circuit.

    Here's the setup:
    I currently have three 1000 microF capacitors which I plan to wire in parallel for a combined capacitance of 3000 microF. (maximum voltage for this is 35V, but that shouldn't be a problem). I could get more capacitors if I really need them, but the local Radioshack seems to be charging exorbitant rates ($1.61 for a 1000 microF capacitor, 35V) and I would prefer to keep costs down if possible. I plan on winding 30 gauge wire around a galvanized steel landscaping spike (6" long, ~1/2" diameter IIRC) to create an inductor - assuming that's feasible. The setup would go like this:

    The three capacitors are wired in parallel to work as one capacitor, which will be wired in series with the inductor. There will be a switch that will, in one position, complete the circuit by including a nine volt battery, and in the other position will complete the circuit without the battery (just the inductor and capacitor). A basic LRC circuit. When the power is shut off, the capacitor will discharge, and then back emf from the inductor will charge the capacitor up again, etc, until everything is lost to joule heating. I would like this period to be approximately 5 seconds, but I would consider two or even one acceptable - I just don't want it to fly by real fast, because it's for a demo.

    I looked for some equations on the internet, and used them... it told me I needed some incredibly huge number of windings - infeasible. I have a lathe that I can use to wind wire around this spike, but there are limits of course. Something didn't seem very right about that, so I tried again with some other equations I found, and got an even more ridiculous 1.2 windings - obviously that's not going to do it.

    Does anyone know how to do the calculuations I want? I would really appreciate some numbers that I could use for this: C, R (if that matters), L (preferably in the form of number of windings needed), and anything else that I'm missing. Thanks.
    Last edited: May 17, 2004
  2. jcsd
  3. May 17, 2004 #2


    User Avatar
    Science Advisor

    You might want to look closely at the label to see if you have mF or uF or pF since a .5F 20V capacitor like you can find for car stereos is the size of a pop bottle.

    A light bulb might make for a good resistor as then you could see what's happening. How else do you see electricity?

    Is this for a science fair or some other project in school?

  4. May 17, 2004 #3
    I'm sorry. I copied and pasted that from somewhere else that had a mu symbol (nobody on the other site knew how to do it), and didn't realize that this one did not. I'll go back and edit the post. Anyway, I was planning on hooking up a voltmeter to demonstrate that the current is reversing directions.

    Yeah, it's for a school project - I'm trying to explain this stuff to people who know essentially nothing on the topic :-/
  5. May 19, 2004 #4


    User Avatar
    Science Advisor

    Ok, here's an idea for demonstrating the capacitor that can last long enough to show people what's going on. You can find the time constant for charging/discharging a capacitor by multiplying them together. So if you had a resistor that was 333 ohms you'd get:

    .003 * 333 = 1 second

    After 1 time constant the capacitor will charge or discharge about 2/3 of the total voltage, and after 5 time constants its bascially fully charged/discharged. You could show the charging of the capacitor with the resistor and how it starts fast and slows down as it gets closer to the source, that will take a little while and shows the charging curve.

    Now, for an underdamped circuit that will oscillate for 5 seconds, this seems like a tall order but that is a guess. Since most examples seem to be in the uS or mS range, no idea of the formulas still hold when you're trying for a time constant more than a 1000x bigger.

    Here's a link that may help:

    In the example on the bottom of page six with the spring-mass example, in order to get the very slow oscillation you would need a large mass with a weak spring and a very low amount of friction. As would be the issue with your setup, getting the resistance low enough with such small components that are affordable.

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?