This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Questions about TI Precision lab Opamp Series-- Low Distortion Design

Other Parts Discussed in Thread: OPA192, OPA172

Dear Ti Precision Lab, 

I am studying the TI precision Lab Low distortion design Part 2. I have three questions here and hope someone could help. 

(1) On Page 9, It is clearly stated running over crossover region will contribute to DC offset change, this will distort the signal and introduce harmonics. But how about " shift in AC parameters" what does it mean?

(2) On Page 11, the instructor determines the crossover region from the upper band of the common mode voltage range, saying that the crossover region at V- 2V. How could we determine crossover region, is that the same as the upper common mode range? 

(3) On the summary page, Page 16, why inverting amplifier has low input crossover distortion and can prevent common mode impedance effects than non-inverting amplifier?

Thanks,

B.F

  • Hello B.F.,

    Thank you for your questions regarding the TIPL - Op Amps course on low distortion design. My responses to your questions are given below.

    1. Operating in the crossover region can affect a variety of parameters, so you must check the data sheet of a particular part to see what is specified. For example, the OPA192 is a rail-to-rail input CMOS amplifier that exhibits crossover distortion. On p. 8 of the data sheet you can see that noise and CMRR are specified over a variety of Vcm ranges, since the behavior shifts significantly after the crossover point.

    2. In most cases, the valid input common-mode voltage range extends beyond the crossover region, such as the OPA192 example above. The OPA172 example from the TIPL material is an exception. In general to determine where the crossover region occurs, check the test conditions for CMRR to see if it is specified across different ranges of Vcm (like in answer #1 above) or check the data sheet curve for offset voltage vs. common-mode voltage. For example, Figure 13 in the OPA192 data sheet shows that the crossover/transition region occurs from around (V+) - 3V to (V+) - 2V.

    3. An inverting amplifier configuration will not encounter input crossover distortion because its common-mode input voltage does not change. In this configuration, the non-inverting input to the op amp is held at a fixed voltage (often GND), and since the input pins of an op amp are a virtual short the inverting input is held at the same voltage. Since the average voltage at the op amp inputs is the same, the common-mode voltage is the same, so no common-mode effects are encountered.




    Best regards,

    Ian Williams
    Applications Engineer
    Precision Amplifiers

  • Hi Ian,

    Thank you very much for your detailed explaination. I can understand the my question1 and 2. But I still have questions on 3.

    I do not quite understand why the common-mode input impedance varies with Vcm in the lecture P13, espectially what the instructor said "With an input sine wave the variation and input impedance causes the current drawn through the source impedance to no longer be sinusoidal, because the current changes with the changing input impedance."

    Besides, I do beleive that, when we do non-inverting topology, we still need to bias the non-inverting terminal for Vcm and the feedback will set Vcm at inverting topology. In this sense, I did not see the difference between inverting and non-inverting topology in terms of input impedance varies with Vcm.

    Thank you again,

    B.F.
  • Hi B.F.,

    Let me provide some additional evidence on the behavior of input common-mode voltage (Vcm) on inverting amplifiers vs. non-inverting amplifiers.

    In a non-inverting amplifier configuration with a gain = 1V/V, the input signal is applied to the non-inverting op amp pin VIN+. Since the circuit is in unity gain, the output voltage is equal to the input signal. Since the output pin VOUT and inverting input pin VIN- are connected with a short circuit, the voltage at VIN- is equal to the output voltage and the input signal.

    If the input signal is a 1Vpk sine wave, that means the output voltage and the voltage at VIN- are the same 1Vpk sine wave. Since Vcm is defined as the average of the voltages at VIN+ and VIN-, and the voltages at VIN+ and VIN- are both equal to the input signal, that means that Vcm is equal to the input signal. Here's a simulation circuit showing this.

    You can see from the simulation results that the common-mode voltage is equal to the input voltage of 1Vpk.

    Contrast this with an inverting amplifier circuit with a gain of -1V/V. In this configuration VIN+ is shorted to GND or 0V, so by one of the fundamental properties of op amps the two input pins are a virtual short and therefore VIN- is connected to 0V as well. The input signal is applied to the input resistor RI. 

    Let's take a look at the simulation for this case, again with the input signal equal to a 1Vpk sine wave.

    You can see that in this case, Vcm is equal to about 50uVpk. For an ideal op amp this would be zero, but a real op amp does not have the infinite open-loop gain and bandwidth required to force its inputs to exactly the same voltage. Still, the Vcm level of 50uVpk is 1/20000th of the 1Vpk Vcm seen with the non-inverting circuit.

    Hopefully I've now proven that the input common-mode voltage Vcm is equal to the input signal for non-inverting amplifier circuits and is approximately equal to the constant bias voltage at VIN+ for inverting amplifier circuits.

    The reasons that changing Vcm affects the common-mode impedance is described in the script of the TIPL video, so I recommend you now go back and re-watch the video, but one of the points made in the video is that changing the reverse-bias voltage across a diode changes its input capacitance.

    In a CMOS amplifier like the OPA192, there are large ESD protection diodes at the inputs to the op amp. They are reverse biased while the op amp is in normal operation and create some common-mode capacitance at the op amp input. Changing Vcm changes the reverse-bias voltage across these diodes, so the common-mode capacitance changes. For JFET amplifiers a similar phenomenon occurs, as changing Vcm changes the reverse-bias voltage on the diode formed between the gate of the input FET and the substrate of the op amp. Finally, on bipolar amplifiers, changing Vcm changes the reverse-bias voltage on the diode formed between the base and the collector of the input BJT.

    In addition to common-mode capacitance, changing the common-mode voltage on all these types of transistors can also affect their current gain (beta) and bias current. This looks like a change in input resistance.

    Best regards,

    Ian Williams