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

Charge amplification

  1. Feb 22, 2009 #1
    I may need to amplify a very small, momentary charge. - a pulse. I have little information about charge amplification and have limited access to more.

    What are some related search terms for amplifying charge?

    Do you have any suggestions for doing so?
  2. jcsd
  3. Feb 22, 2009 #2
    Depending on the size of the charge, and length of the pulse, a simple non-inverting op-amp might be sufficient. Also, how long do you think the pulse will be? How much charge? What are the accuracy requirements? Wikipedia has an article on Charge Amplifier and the fourth external link, http://www.designinfo.com/kistler/ref/tech_theory_text.htm" [Broken] might be useful.
    Last edited by a moderator: May 4, 2017
  4. Feb 22, 2009 #3
    If you know the capacitance and voltage then you know the charge. Therefore if you know the input and output capacitance you can set the voltage amplification to give the desired charge amplification.
  5. Feb 22, 2009 #4
    my issue is the sensor I am using puts out a variable amount of charge dependent on force / impact etc. I have yet to test the sensor in my design, so this is somewhat preliminary.

    It is a piezoelectric sensor with 1 v/g (max) - I doubt I will approach this with my design.

    The ratings for the sensor say it 'works like a capacitor' and gives this value: 244 pF
    Do they mean that charge can be applied to a series of these sensors and when force acts upon it, a greater charge will be released?
  6. Feb 22, 2009 #5


    User Avatar
    Science Advisor
    Gold Member

    Welcome to PF zepto-. Sounds like this sensor produces a voltage depending on how much force you apply (hence the 1 v/g) and its internal capacitance is 244 pF. I concur with ravioli that an op-amp is probably the way to go. See figure 4 in the link he provided. Notice that the pieozoelectric transducer (in dotted box) shows its internal capacitance (C_c).
  7. Feb 22, 2009 #6
    1. Do you have a data sheet?

    2. What you are asking for depends upon your application.


    3. Any amplifier will present a current path and discharge (or adversely bias) the transducer. For this reason a class of amplifiers called charge amplifiers are often used that take a longer period to drain the charge.

    http://en.wikipedia.org/wiki/Charge_amplifier" [Broken]
    Last edited by a moderator: May 4, 2017
  8. Feb 23, 2009 #7
    I am not sure what you mean by 'datasheet' - I am not an engineer, yet. I have data specs for the sensor, but not a full set for the application and intended circuitry.

    I have been reading about charge amplifiers and my current question is: How do I determine the ratings for a charge amplifier? (perhaps this was Phrak's Ques.)

    The accuracy is negligible, this is a backyard operation. If it is any reference, I was able to charge a 0.1 uF capacitor with 3.72v (=372 nC) to accomplish my goal. My concern is time, I do not know how that plays into the piezo element and op-amp requirements.

    you guys are brilliant.
    Last edited: Feb 23, 2009
  9. Feb 23, 2009 #8
    For a "backyard" operation you might get away with just a voltmeter with a high resistance in series. The charge will start to go down pretty quick but with reed switch in series with the meter and a quick eye (or a peak hold reading voltmeter, readily available for cheap) Depending on your voltage, themillivolt scale of most DMM's are very high (>1000M) with pretty low bias current (current leaking backward into the cap)

    Just a possibility from one garage experimenter to another.. Oh sorry, back yard :)

    btw. What kind of instrumentation do you have access too ? A very simple way would be to forward bias the cap with a DC voltage and view the current pulse on a scope or meter. The Fluke 187's have a pretty fast MIN/MAX hold function.
    Last edited: Feb 23, 2009
  10. Feb 24, 2009 #9
    Excellent, Zapitgood! I hadn't thought of that.

    What's the input current on a fluke 187 on the 2 volt setting?

    The device has a capacitance of 244 pF. 50 pAmp of drain will decay a 1 volt signal to ... 0.8 volts in one second.

    Edit: So I was curious and looked at the "extended" specs for the Fluke 187/189,

    http://assets.fluke.com/datasheets/2153ExtSpecs.pdf" [Broken]

    The mV/V input has an input impedence of >10M ohm, so they say.

    10 Mohm * 244 pF = 2.44 milliseconds. A voltmeter won't do. He needs a low input bias current FET op amp.
    Last edited by a moderator: May 4, 2017
  11. Feb 24, 2009 #10
    I agree, that would be the best approach. I think the input Z is much higher, there are other models with a power on feature that removes any shunt R. That being said even a Gohm would be pretty quick, but using the 250uS peak hold function should capture the event nicely even with 10M/2.44mS decay.
    Last edited by a moderator: May 4, 2017
  12. Feb 24, 2009 #11
    Ok, a very crude experiment this morning yielded these results. Not knowing the actual parameters I used a 50uS 5V pulse (about 10nS rise/fall) in series with a 270pF cap and into the Fluke on peak capture, millivolt range. Readings were about 220uV avg, pretty poor resolution. So some amplification is definitely needed. My rise is probably a lot faster than the load cell response.
  13. Feb 24, 2009 #12
  14. Feb 24, 2009 #13
    Ooooh. I didn't catch that about the peak hold function, earlier.
  15. Feb 25, 2009 #14
    I am still absorbing all the posted information, forgive me for not having an intelligent reply.

    The event frequency from the sensor will be at least 6.25 Hz and preferably higher. .. if that helps understand the op-amp issue.

    thank you all very much for the insight.
  16. Feb 25, 2009 #15

    pulse freq: - - - - - - - - - - - 6.25 Hz

    sensor output
    voltage sensitivity:- - - - - - - - - 1.1 V/g (assume 0.25 g for app.)

    charge sensitivity: - - - - - - - - - 260 pC/g

    Sensor capacitance: - - - - - - 244 pF

    required charge input from amplification is at least 100 E-9 Coulombs.
    Last edited: Feb 25, 2009
  17. Feb 25, 2009 #16
    Last edited by a moderator: Apr 24, 2017
  18. Feb 26, 2009 #17

    I have used these picoamp input amps before, have to be very careful about layout, use guarding whenever possible. I had to lift the pins and connect the sensor "dead bug" style to reduce leakage. Keithley instruments has a really great handbook called "Low level measurements handbook" that is an excellent reference on these subjects. I'm not sure they are giving it out free anymore but I think there is a web version. P.S. They do still have it available in print...
    Last edited by a moderator: Apr 24, 2017
  19. Feb 27, 2009 #18
    I'm suprised as all get-out to find some listed. I hadn't been aware of their existance. I'd think that even blowing air on one would have a measurable effect due to the change in surface leakage due to water vapor absorbtion.

    So I was doing some thinking (a dangerous thing), wondering how to compensate for the remaining current--if one were so inclined. The input bias current into the op amp is a function of temperature. After the op amp is aged and stablized, you could calibrate the accelerometer plus amplifier pair as a function of temperature. A thermistor and an 8 pin ucontroller having an 8 bit ADC and counter could be calibrated to supply variable compenation current depending on the ambient temperature.

    Have you heard of this one? Instead of measuring the voltage out of the accelerometer as a function of acceleration, you can supply a bias to the accelerometer to keep the voltage at zero. The amount of bias required is also presented as an output signal. In this way the op amp input bias current is constant over acceleration.
    Last edited: Feb 27, 2009
  20. Mar 2, 2009 #19
    Never seen that approach specifically, guess it comes down to the accuracy/stability of the correction components.. Being fairly lazy, I would put everything in a crystal oven (mostly because I have a few lying around) and run above ambient. That might stabilize the bias current ??? But you will need to keep the device near ambient, outdoor temperatures might throw the oven regulation off.. The oven I have claims +/- 1/2 degree stability and you can mount the parts upside down right on the heater or box it up in foam.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Charge amplification
  1. RF amplification (Replies: 0)

  2. Current Amplification (Replies: 6)