# op-amp outputs supply voltage on GND Signal

I am trying to use an op-amp (first time). I know the basic theory of its amplification circuits. To test I am using an Arduino Nano Microcontroller to create a Signal (0-5V). I want to amplify the signal by a Factor of 3. I measured the input and the output of the opamp relative to GND with an oscilloscope. The amplification works (5V to 15V --> factor 3), however when the signal goes to ground the opamp always reaches the supply voltage (tried with different supply voltages). I know that an op amp amplifies to the supply voltage if there is no feedback, but i cannot explain how it can happen in my case. How do I prevent the opamp from switching to supply voltage once the signal raches GND?

EDIT: The Op-Amp is a MC33078 (datasheet)

• All opamps are not created equal. Yours might not have a common-mode input range that includes its negative rail. If so, the opamp might be going into an error condition when the input is driven negative too far. And as above, it should NOT be inverted. What is the part number? Mar 21, 2019 at 18:09
• @AnalogKid I added the part into the description. Mar 21, 2019 at 18:18
• What do you measure if you connect In to the Arduino's +5V directly? What if you connect it to the Arduino's GND directly? Mar 8, 2020 at 14:20

The circuit you draw is the right one for a gain of 3, but the waveforms you show indicate a negative gain and an offset. The equation for Vout would be something like Vout=15V-A*Vin where A is your gain.

I think your circuit you've built is not what you've drawn. Double check your connections. 5Vin should make 15Vout. If the opamp is rail to rail, you'll also get 0Vin makes 0Vout.

Looks like you driving the output tri-state (open circuit) rather than pulling it to ground. You can add a resistor from the non-inverting input to ground or change your firmware. When it's open the op-amp input will float up or down, predictably if it's a bipolar op-amp to one supply or the other, and unpredictably if it's a MOS input op-amp. For example, an LM324 input will float upward because the bias current flows out of the op-amp input.

You should also be using a "single supply" (or rail-to-rail I/O) op-amp - the output should swing to near ground and the input common mode range should include the negative rail.

Edit: The MC33708 is not 'single supply', as you know. Some op-amps (not sure about the MC33708, I've used them, but have not knowingly violated the input common mode range that way) may behave the way you describe. LM358 is the the same pinout and should work okay.

• Unfortunately adding a pull down resistor didn't impact the output signal. Mar 21, 2019 at 18:16
• But i think its the single supply that i have bought wrong. Mar 21, 2019 at 18:22
• What is your op-amp type? Something like an LM324/LM358 should work fine. Mar 21, 2019 at 18:37
• Its a MC33078. I added a Datasheet into the EDIT of the question. The Type is Dual Supply Mar 21, 2019 at 18:40

Unfortunately the MC33078 does not have an input voltage range which includes the negative supply rail. From page 3 of this datasheet, the parameter $$\V_{ICR}\$$ is quoted to be ±14V, or perhaps even as bad as ±13V, when the supply is ±15V.

What this means for you, with your supply potentials of 0V and +30V, is that you can expect your opamp to work with input voltages from +1V to +29V. An input of 0V is outside this range, and the opamp's behaviour in this condition is not defined.