2
\$\begingroup\$

I am looking into a circuit simlar to what is attached here, when I place a probe between the opamp's output and the inverting input - I see oscillation when V1 = 0V. But when voltage is applied to V1 (i.e 10V), the oscillation is gone. Is it normal? Or something is going wrong with my circuit and I need to compensate for it?

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
  • 1
    \$\begingroup\$ Is the op-amp also powered by V1? Or does it have its own power supply? \$\endgroup\$ – Dan Laks Oct 17 '14 at 22:47
  • \$\begingroup\$ And is your "probe" the leads of a voltmeter or an oscilloscope? \$\endgroup\$ – Dan Laks Oct 17 '14 at 22:49
  • \$\begingroup\$ Try a small capacitor say 15pF across the feedback resistor. \$\endgroup\$ – user207421 Oct 17 '14 at 22:54
  • 2
    \$\begingroup\$ Huh? Are you connecting the 'scope probe's ground wire to inverting input? \$\endgroup\$ – George Herold Oct 17 '14 at 23:45
  • 1
    \$\begingroup\$ If the gnd probe goes to negative input you are likely causing R2 to become shorted at high frequencies giving very high gain and the noise on the scopes gnd line is amplified and this may indeed look like oscillation. Do you have a reason to connect the probe like this? \$\endgroup\$ – Andy aka Oct 18 '14 at 9:09
2
\$\begingroup\$

Here's what the spec of the AD712 says regarding operating either input close to the negative power rail. I say this because you haven't shown the power rails and, like @Rennex, I suspect you are not using a negative rail.

Typically exceeding −14.1 V negative common-mode voltage on either input results in an output phase reversal.

This was taken from the data sheet of the device when connected to a +/-15V supply and if your negative rail is in fact 0V, then going to within 0.9 volts of 0V will cause an inversion in the op-amp which will likely result in positive feedback and hence oscillation as you observe.

Lifting V1 up a few volts will likely stop that inversion and the op-amp stops oscillating. So if V1 is at 10V, then V+ is about 1 volt and it's going to be round about the area where input inversion stops taking place.

OK, it's not a long shot but, in my water, I feel this is the problem.

The other issue is that you appear to be saying that your scope probe is connected to V-. Try connecting it to 0V for sensible results.

\$\endgroup\$
2
\$\begingroup\$

Your question doesn't show how you're powering the op amp, but I'll take a wild guess that you're not giving it a negative supply, just ground. The chip's input voltages need to be at least a volt away from the supply rails to operate properly. E.g. with a 0V...12V supply, both the inputs need to be between 1V...11V.

\$\endgroup\$

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.