Capacitance Measurement

Discussion in 'General Electronics Chat' started by viju, Sep 29, 2008.

  1. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    I must thank all the members whoever supported me in this forum.I am surprised of getting solutions immediately and effectively.

    I have been given an interesting project of measuring a capacitance of 0.22 microfarad and its leakage current.How do I go about it. I need to measure it and display the value.I have tried to find the way for measuring the value of capacitor using astable multivirator.I generate a 1 Khz pulse from my PLC and read the value;It works fine.But I am helpless in measuring leakage current.If measured I need to amplify it and feed it to a analog input module of a PLC ( 0~10 Volts).Guide me.
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,646
    2,344
  3. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    There are many methods to measure capacitance. Amongst the simplest, you basically charge it to a predetermined reference voltage, then discharge it through a resistance / constant current, then measure how long this process takes.

    Other methods involve solving for impedance with a modulation, charge transfer, etc. These are usually employed when the capacitance is small or high bandwidth is required.

    As far as leakage, I would read how Bob Pease did it here

    http://electronicdesign.com/Articles/ArticleID/15116/15116.html

    also, you can read about 'soakage', which shows that leakage is actually a function of time and voltage.

    http://www.national.com/rap/Application/0,1570,28,00.html

    Steve
     
  4. theamber

    Active Member

    Jun 13, 2008
    318
    0
    Diferent types of capacitors materials need diferent methods for testing leakage. Capacitors have a maximun allowable leakage. I see that good analyzers now days charge them to high levels of current to see their effects. Also leakage measurements on nonpolarized electrolytics must be made in both directions.
    To have a simple test you can build a simple circuit like this:
    http://www.angelfire.com/electronic/funwithtubes/Testing_caps.html
     
  5. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    Thankyou all for the reply.But I was asked to measure the leakage currrent when the capacitor reaches the supply voltage.
     
  6. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    Can I connect a resistor( 10 meg-ohm) in series with the capacitor and charge the capaciotr for few seconds( Until the capacitor reaches the supplied voltage) then measure the voltage drop across the resistance.If suppose current drawn by the capacitor is 2 micro- amps then my drop across my resistor would be 20 milli volts. I can amplify the 20 millivolts to the desired signal level.Am I going in right direction.
     
  7. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    So this is a manual operation? I was under the impression that you wanted to build a piece of test equipment.

    A capacitor is governed by the fundamental equation i(t) = C dV/dt

    I was suggesting using a constant current, then measuring the voltage after a particular length of time. This way your i(t) = a constant amount of current, the amount of time or the current selected could be the range that the meter is on, you can easily see that C is the only unknown in this situation.

    You can use a basic RC circuit, then measure the amount of time it takes to get to a paricular voltage. One problem with this is that as Vsupply is reaching Vcapacitor, your current drops off. This is why there is an RC time constant, because of this non-linear type function. With a current source, you do not get this.

    Steve
     
  8. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    No. I am building an automated fixture where an assembled component is placed in that and with the help of this circuit I need to measure and process them through a PLC.

    By going through the equation and method suggested by you,

    If my current source is 100 milliamps and I charge the capacitor for a period of 2 seconds and measure it, C = 100 milliamps / Voltage read at that particular time will be my capacitor value.Is that OK.
     
  9. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    Steve Suggest me a method to measure leakage current for 0.22 microfarad capacitor.
     
  10. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    Hi,

    The charging current should be kept low, so that you have a longer measuring time. The longer the allowable time, the better resolution you will get. Also, the more accurate the current source, the better.

    Leakage is another issue separately. Bob Pease used a very low input current opamp and basically hooked up the capacitor to a unity follower. He precharged the capacitor and measured it over long periods of time. I don't really know another way to do this. Maybe use a really high gain transimpedance amplifier to measure the current?

    Steve
     
  11. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    Steve:

    I am confused at the basic level.What does leakage mean? Is it the current drawn by the capacitor after charging to the supplied volatge?Practically when voltage becomes equal there shouldn't be any current flowing.
    Rightnow I have connected a microampmeter with input impedence of 1K in series with the capacitor.After switching on the power the capacitor current reduces and stays constant at 10 microamps( I tested with 10 mf).When I measured the voltage drop across the meter it read 10 millivolts.Is that the leakage current? Or something else.Please help.
     
  12. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    Hey,

    Capacitor leakage is actually inherent due to the dielectrics used. In ceramics, it is defined by the insulation resistance. Basically, there is a shunt resistance of very high impedance across all types of capacitors, some lower than others.

    Your test sounds about right. As the capacitor leaks, you're supplying the missing current to keep it charged. 10uA is actually a pretty common leakage figure,

    http://measure.igpp.ucla.edu/hemisphere/Discussion2/Cap Leakage.htm

    Steve
     
  13. viju

    Thread Starter Senior Member

    Sep 28, 2008
    120
    0
    Thanks Steve.My ammeter fullscale reading is 200 microamps(digital).Is any meter available with range of 50 microamps.Can it be made by ourselves?
     
Loading...