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

Linear Power Amplifier Design

  1. Feb 5, 2010 #1
    Hello All,

    I am faced with a very interesting challenge and I'd like to see if you all could give me any insights. I'm currently planning to design a linear power amplifier loosely based on the design on page 22 of this PDF file here: http://www.wadsworthmedia.com/marketing/sample_chapters/1401848567_ch05.pdf

    So far, I have just been doing simulations of the circuit to get a feel for it's operation. However, I want to do a few things different with the design that are, at the moment, throwing me for a loop.

    In my application, I am building an amplifier specifically for driving coils of varying inductances at frequencies from 0Hz to about 1MHz (up to 20MHz ideally, but I I'm settling for lower frequency for simplicity). I found that sticking an inductive load works fine with this amplifier design except for the fact that the amperage through the coil varies with frequency due to the coil's inductance.

    What I am thinking about is modifying the schematic so that it doesn't utilize voltage regulation but current regulation to keep the coil current constant throughout the frequency range. So, in other words, I want to basically make this into a voltage to current amplifier, with, say 100mV on the input translating to 100mA through the output load.

    To be more clear, I'm mainly going to be driving very low impedance coils, (0-5Ohms, ~10uH - 1mH max) at frequencies from DC to about 1Mhz or so. The maximum current I want to be able to source is about 1-5A max. To implement the current regulation scheme, I found this circuit from a Wikipedia page which shows how one can make a simple op-amp voltage to current converter:


    I modified the amplifier schematic in the simulator to use the a 0.1 Ohm sense resistor for a current shunt in the feedback circuit. This allowed me to get an output of 1mA for an input of 1mV into the circuit. This *seemed* to work fine until I attempted to drive a given coil (100uH for example) at higher and higher frequencies. What I found is that the output current started to look more like a triangle wave than a sine wave, I saw that this happens because the feedback voltage was turning into a square wave in attempts to regulate the current, but due to the impedance of the coil, the current couldn't respond fast enough.

    So I scrapped that idea and went back to the original schematic, which did produce a nice sinusoidal output, though the current varied with frequency. My next idea was to somehow come up with a feedback circuit that could increase or decrease adjust the output voltage across the coil to keep the current at a set amplitude but without the same issues that I was having with the other scheme. This seems like it shouldn't be that hard to do, but I have no idea how to approach this problem. Is their a better way to do this? Ultimately. what I am trying to make is a current-regulated AC power source.

    Jason O
  2. jcsd
  3. Feb 5, 2010 #2
    Going through some calculation using.
    1 Amp, 10 microhenry, 1 megahertz

    Xl=2*3.14*10^6*10^-5=62.8 ohm

    E=1*62.8 Volt

    So if you are talking about RMS volts, Peak volts will be 88.8 and peak to peak volts will be 177 volt.

    So roughly speaking you require an operational amplifier using +100 VDC and -100 VDC
    power supplies.

    Of course higher inductance coils or higher currents will require even higher voltages.
  4. Feb 5, 2010 #3
    Hi Carl,

    Thanks for the info there, that's rather depressing to hear seeing as I want to use +-12V for the supply rails! Perhaps I'll shoot for about 100-200kHz or so where I can reasonably get that much current through the coil. But my main problem still remains. How would one make a constant current output into an inductor with the circuit?

    To simplify things, I found a really nice power op-amp IC that looks like it should great for the frequency range I want to work in:
    http://www.mouser.com/ProductDetail/Texas-Instruments/OPA541AP/?qs=sGAEpiMZZMtCHixnSjNA6Mrdwin7F8A2zuoa2IM%252bibI%3d [Broken]

    But the voltage to current converter circuit is still a challenge. Would work great using the circuit from the Wikipedia page it I weren't trying to drive an inductive load.

    Jason O
    Last edited by a moderator: May 4, 2017
  5. Feb 5, 2010 #4
    Circuit from Wikipedia page should work with following modifications.
    For power supplies use +12 volt with negative grounded and -12 volt with positive grounded. (Standard connection for power supplies)
    Delete Dz and Rz
    Apply sine wave signal to operational amplifier + input/ground

    Expect to get maximum output voltage of about +/-10 volt or about 3.5 VRMS.
  6. Feb 26, 2010 #5
    Hello Everyone,

    I'm still putzing around with this amplifier design. I finally decided to not concern myself with the current control aspect for now as I am running into another problem, that being bandwidth. I did several simulations in Multisim to see how high a frequency I could push the design from pg. 22 of the link I referenced in my first post (http://www.wadsworthmedia.com/marketing/sample_chapters/1401848567_ch05.pdf). But as the PDF itself also states, this design tops out around 200-300 kHz as the crossover distortion gets ridiculous if you push it any faster.

    I'm wondering if anyone here knows of an alternative design that could make it up to around 1MHz? I don't know enough about the subject and I'm not sure what I should be searching for. Most of what I find is either related to Audio amps or RF power radio amps, which all leave a big void in the range of 100kHz to 1MHz, which is what I would like to also operate in. Is there even a such thing as a DC-1MHz power amplifier??

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

Similar Discussions: Linear Power Amplifier Design