I have an existing analog input circuit on a device that utilizes an op amp (TI LM2902) as a buffer. There is a reference voltage of 9V and the Op Amp is powered by 5V. The input signal to be read by the ADC originates at an off-board sensor.

ADC Schem

I am wanting to make a new device with more inputs and re-use the existing circuit. The new device I am making does not have any need for 5V, so I am wondering if it is possible to raise the Op Amp supply voltage to 9V without affecting the ADC values? I want the A to D readings to be consistent between the two device.

My understanding is that the supply voltage just dictates the ceiling of the output voltage. Is this correct and the ADC values will be the same whether I power the op amp with 5 or 9 volts?

  • \$\begingroup\$ That is correct. \$\endgroup\$
    – τεκ
    Aug 10 '18 at 19:08
  • \$\begingroup\$ How will the ADC survive a 9 V power-up transient, etc.? \$\endgroup\$
    – Transistor
    Aug 10 '18 at 21:16

My understanding is that the supply voltage just dictates the ceiling of the output voltage.

Yes, that is correct.

Yes, but it is also the problem.

The danger is that in a power-up, fault, or sensor disconnection or short that the op-amp output will exceed the maximum allowable input voltage for the ADC. This limit is likely to be 5 V or very slightly over. Exceeding this will destroy the ADC chip.


simulate this circuit – Schematic created using CircuitLab

Figure 1. Over-voltage protection.

To do what you require some form of over-voltage clamp is required. This could be a resistor and 5.1 V Zener (Fig. 1a) or resistor and diode to +5 V (Fig. 1b).

  • \$\begingroup\$ Thanks for the insights. There is also going to be a zener diode at the entry to board for ESD protection. For that I am using a SMAJ18A (re-used from another BOM) Would it make a difference one way or another if I put the 5V over-voltage preventing zener at the op-amp input versus the output? I would rather not put anything on the trace between the op amp output and the ADC input. Also, the zener diode for limiting to 5V should have a 5V BREAKDOWN voltage, not WORKING voltage correct? The 5V WORKING voltage zeners have breakdown voltages of over 6V so that would be too high \$\endgroup\$
    – zme
    Aug 14 '18 at 0:22
  • \$\begingroup\$ The Littlefuse datasheet agrees with your ratings on the SMAJ5.0A (5 V version) and I suspected as much so I added R1 on Figure 1a. I can't really advise you much more than that as I haven't faced this exact problem. \$\endgroup\$
    – Transistor
    Aug 14 '18 at 9:37

Yes, you should be fine.

The ADC values depend on the voltage range of the ADC itself. If the voltage going in to the ADC is the same, the ADC values should be the same. Changing the supply voltage on the op-amp should not change the ADC input voltage as in this configuration the op-amp is only acting as a buffer.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.