No Script, No Fear, All Opinion
RSS icon Home icon
  • EEVblog #600 – Opamps Explained

    Posted on April 6th, 2014 EEVblog 36 comments


    The most often requested video! Dave explains what Operational Amplifiers (OpAmps) are and how they work. The concepts of negative feedback, open loop gain, virtual grounds and opamp action. The comparator, the buffer, the inverting and non-inverting amplifiers, the differential amplifier, and the integrator circuit configurations are also explained.
    Then a practical breadboard circuit to demonstrate a virtual ground and the effect of voltage rail limitations.
    All EEVblog Opamp related videos are HERE

    Forum HERE

    Be Sociable, Share!
    • Chris

      Woo, 600! Congratulations. Love the Fundamental Friday videos.

    • GMCurrie

      nice thx – heh – jus rapped into google images: ‘art of electronics horowitz man transistor t shirt’

    • Johm

      I’m definitely expecting followup #60X Non-idealities of OPAMPS :)

    • huh

      Wow.. thanks Dave, this one was great! You have the gift of making things easier to understand; opamp basics aren’t exactly rocket science but not many people would explain the concept in a such understandable way.
      I would like to see some videos from you about their applications such as active filters, instrumentation (linear or logarithmic) amplifiers, etc.

    • Chris

      OK, this is probably a stupid question, but let’s say I were to make an inverting amplifier, and I hooked up ground to the inverting input and the resistors to the non-inverting input and ouput. Would that still work just fine? Or if not, why not?

      • Johm

        Chris: This configuration wouldn’t work:

        1. The opamp is working under POSITIVE feedback, that means it will become unstable and saturate quickly. The voltage at the output is going to be added to the one your feeding in. Fortunately this can be useful to generate fast transients from a less sharp input signal. This is called the Schmitt trigger and is used, mainly, to regenerate dicky digital signals.

        2. As there is no negative feedback, the opamp isn’t working in lineal zone therefore, rule 2 doesn’t apply so there is no virtual ground. Probing the inputs would give a non zero value, depending on the saturation at a given moment.

        In other words, the gain would be that of the opamp (e.g. 100000) and the output would always saturate.

        Hmm, hope it helps though I’d never reach Dave when explaining.

        • Chris

          Thank you. That was quite helpful. It’s been a while but I think the Schmitt triggers I made always used a 311 comparator and I didn’t make the jump that it would also occur for op amps.

    • Paul

      Awesome video Dave, hopefully there’s more to come, it left me with more burning questions…

      Virtual Ground: Is this useful for something? For example, with a few op-amps in the same virtual ground configuration could you hook them up in series to make a larger voltage (acting as an indpendent/isolated ground) up to the limits of the rail voltages?

      I also assume that the op-amp would be unable to maintain the virtual ground level if the input frequency exceeded that of the op-amp’s capability, so you’d probably start to see some of your input signal at the input in that case.

      Voltage Divider Resistors:

      Why would you not use absolutely massive values for these? Like 100MΩ or something… wouldn’t that effectively lower the power consumption of the circuit? Isn’t using lower values just burning off extra power for no reason?

      Using A Single Rail:

      To get the “AC” back why can’t you simply use a capacitor to decouple the op-amp’s output and remove the DC bias component? That seems like a much cheaper alternative than a dual supply. At least for audio applications.

      • Raphael

        @Paul
        You can’t use extremely high resistor values because then your opamp input bias current becomes an issue which can affect the result.
        Having very high resistor values can also create problems in noisy environments, because the high-impedance nodes are susceptible to noise pickup. Selecting resistor values is always a trade-off…

        • tchicago

          Chris,
          instead of an inverting amplifier, the configuration you described would make a well known circuit called Schmitt Trigger. Scroll to the middle of this page and your exact schematic is there: http://en.wikipedia.org/wiki/Schmitt_trigger

          Schmitt Trigger is a bistable circuit, a kind of memory element that can have two different outputs for certain range of input values, can also be called a comparator with hysteresis. It is is widely ised to filter out the analog noise from a digital signal.

          • tchicago

            looks like I’ve replied to a wrong thread. Guessing Chriss will find it anyway.

    • Gustavo

      Really great!!! Thanks!!

    • http://www.danielbjohnson.net septer012

      5 Years * 52 Fridays = 260 open places to cover OP AMPs, you finally got around to it eh?

    • http://www.saturnpcb.com Kenneth Wood

      Dave,
      The blip on the output using the lower R values should be due to the input offset voltage of the LM358. The higher R values help the blip on that particular op-amp. However it would be interesting to see the same test over the entire BW of the LM358. The op-amp has a bias current flowing out of its inputs. This plus the capacitance around the input plus the impedance help to reduce the effects of input offset voltage. Also the output stages tend to run cleaner through the zero cross over at lower currents. It also helps to run more current out of the op amp. You can also design the circuit using a constant current draw with a resistor to v-, this forces the output stages to stay biased on.

      Regards
      Ken
      Saturn PCB Design, Inc.
      http://www.saturnpcb.com

      • tchicago

        The lower R does not affect the input. I’d say it is rather overloading the max output current of the opamp. The datasheet says the output can only sink ~20 microamps when the output voltage is around 200mv.

        • http://www.saturnpcb.com Kenneth Wood

          Unless I’m missing something, it looks like Dave is running the 358 with no load so that should not be the issue.
          Ken

          • tchicago

            OPAMP’s output is loaded with a 10K resistor which connects output to the (virtual) ground.

    • speedyant

      Dave,
      Please don’t use “Dave CAD” because I’m laughing like an idiot…
      My parents thought I’m crazy!

    • Phil Tirino

      I have a degree in mathematics and no formal educational background in EE, and I watched this whole video and I can’t say there was anything on here that I didn’t already know… no fault of yours Dave. This is a topic I am very interested in and I have taught myself quite a bit from a book available online called Opamps for Everyone. I would have loved to have had a video like this when I first started learning. Especially with my dyslexia, videos are like my go-to sorce of information and I read to fill in the gaps. You did a very nice job explaining opamps as real objects and not some theoretical entity. I am definately looking forward to follow ups in the comming Fridays.

    • Guido

      Been watching the videos for quite a long time but never commented or posted on the forum but here goes.

      Not all that new stuff for me but I always enjoy watching the fundamentals Friday videos. I would have mentioned the all so common differential input audio power amplifiers among the power opamps :P They are after all still in essence the same thing and also used in the non inverting or differential amplifier configurations. For that future video I might like to see You explain on some limitations to feedback one can apply on an opamp and maybe take a look at some parasitic oscillations too. Even though these are more of a problem in discrete opamps they can still be considered a problem related to all amplifiers having limitations to gain and bandwidth.

      Keep that camera rolling and have fun so that we can have fun.

    • http://poonisteps.wordpress.com/ Raja Balu

      Congratulations on 600th video blog upload!

      Thanks Dave. I love your Fundamental Fridays.

    • Sue

      Good job of explaining the fundmentals of OP-Amps. Back in the days of cast iron and discreet transistors I learned Op-Amp theory, your explaination was concise and to the point.
      Anybody who knows what end of the soldering iron gets hot should be able to start building simple circuits and find uses for this little workhorse. Forty years and still an indespencable part.

      Take Care
      Sue

    • eevBadict

      DAVE!!!!

      YOU ARE SO MUCH MY HERO !!!

      well, you was allready long time ago but now you are totally my SUPERHERO !

      i never saw it all so clear with OPAMPs

      please recieve from me an infinite gain loop of THANKYOUs

    • V_King

      great revision of op-amps! thanks.

      waiting for a part two of practical aspects of op-amps use :)

    • Ronald1962

      What a great video.

      PLEASE go on for follow ups.

      Thank you very much!!!!!!!!!!!

    • Sherri

      Thank you Dave. Great video. I think I’m beginning to get a handle on these. Thumbs up to more opamps, please :)

    • Simon

      Dave,

      Just a quick correction which I’m sure you’ve spotted.

      In your example of the inverting amplifier configuration with a d.c. offset applied to the non-inverting input.

      You applied half rail volts (+5V from your 2x10k divider)but remember that from a non-inverting amplifier standpoint the gain is [1 + Rf/Ri] (11 in the case of Rf=10k and Ri=1k) so your output would attempt to sit at +55V.

      This of course cannot happen with Vcc at 10V.

      To make the output sit at half rail you need 5/11 Volts at the non-inverting input e.g. a divider using 10k to Vcc and 476R to 0V.

    • Edy

      Thanks Dave, finally understanding a bit more. So the non inverting configuration is basically using a voltage divider on the output and inverting input to set the output. If non inverting input is set to 1 V and the opamp must make the inverting input the same, by virtue of the voltage dividers 1k and 9k resistor setup it must make the output 10V to get the 1V on the non inverting input. It makes sense! Thanks!

    • Petter

      Can you do a current feedback opamp session? I see most sites struggle to explain what current feedback opamps are, how they work etc.

    • Joseph Buganski

      Brilliant Dave! Enjoyed this video bigtime!!! You Rock! Lexington, Ky-USA

    • Trev

      I’ve paused the video so I can type this comment and hopefully the ‘answer’ isn’t in the rest of the video…
      When you’re describing the gain (Av) of the inverting amp configuration, you wrote it up as Rf/Ri, but surely that should have a minus sign in front of it to highlight the fact that it’s an INVERTING configuration? i.e. -(Rf/Ri)

      • Trev

        Dammit!!! I shouldn’t have paused it as you ‘corrected’ it 25 seconds later… (Normally you insert that overlay caption in such cases…)

    • Warren

      A bit off subject, but do you realize your EEVblog outtro buzzes at 60 Hz, not 50?

    • Mike

      Thanks for the video Dave! Keep up the good work!

    • Worf

      I’ve heard of the virtual ground, but when analyzing op-amp circuits, I’ve been told the inputs are a “virtual short” rather than virtual ground. I.e., the inputs act like they’re shorted together, except of course, rule 1 means there’s no real short.

    • Pingback: Videotipp: EEVblog #600 – OpAmps Tutorial | Laudamm 22