TL082B: unexpected output voltage with 0[V] Vin during production testing?

Hello,

Could you please share what your supply voltages are?

Best,
Jerry

Hi Jerry,

See below from schematics.
I am asking customer to confirm on the E2E that U-5.0V really means -5V..


BR,

A.

Hi BR,

with supply voltages of -5V and +15V the circuit should work.

Are you sure that not the TL082C was soldered into the boards?

Some more ideas:

The DPO 4104B is a scope, right? Scopes don't show the highest precision when it comes to DC voltage measurements. I would repeat the measurement with a good digital voltage meter. Short the input (pin 3 of TL082) carefully to signal ground and connect the digital voltmeter via a 1k isolation resistor to the output of TL082. Have the isolation resistor sitting directly at the output of TL082.

Another cause for unexplainable DC offset voltages can be demodulation of HF-EMI hitting the OPAmp circuit. Because of this, turn-off all cell phones during the measurement and/or use shielding.

Another cause can be oscillation of OPAmp due to a too high load capacitance or an insane complex load. What else is connected to the output of OPAmp? I see two lines leaving the output of OPAmp? Mount an isolation resistor into every line leaving the output of OPAmp. From my experience the TL082 doesn't like unisolated capacitive loads at all, especially not, if the obligatory the phase lead capacitance in the feedback has been omitted. So, also mount a suited phase lead capacitance in parallel to R131. Something between 10...47pF should work in combination with a small capacitive load. But a higher capacitive load may need a bigger phase lead capacitance or even the implementation of the "dual feedback method".

Kai

Hi Kai,
Thanks for the feedback.
Some more info on the load of the TL082B.
From the info I got it really seem to be a TL082B (SOIC package):

Thanks in advance.

A.

Hi BR,

ok, R639 doesn't seem to be an issue. But I hope the two isolation resistors are not too far away from the output of OPAmp.

I don't see any issue with this circuit. Maybe I'm overlooking something, but to me everything looks acceptable.

If it was my board, I would to the following things:

I would cut everything from the output of TL082 and everyting from the input, so that only the two feedback resistors and the resistor and the cap at the input (shown in your first schematic) are present. Then I would short-circuit the input to signal ground. Then I would take a digital voltmeter and measure the output voltage of OPAmp by the help of said 1k isolation resistor. Then I would measure the voltage at the -input of OPAmp (pin 2) again by the help of a 1k isolation resistor. It should be about one tenth of the output voltage, indicating that the OPAmp is in regulation.

I also would measure the supply voltages of OPAmp with the digital voltmeter.

Then I would probe the output of OPAmp by the help 1k isolation resistor. Is there any oscillation? Can still the 100mV offset voltage be seen while there's no oscillation?

Then I would probe the supply voltages. Are the supply voltage clean?

If the 100mV offset is still present, then the OPAmp is bad. It could have been damaged somehow during the handling, mounting, soldering, measuring, etc. Can any ESD arrive the TL082? Is there anything else wrong with the board? Are the supply voltages clean, stable and noisefree?

Kai

Anthony,

Kai has provided a great suggestion. Another simpler test you can do is to see if the behavior follows the device. Since you have passing boards and suspect boards with the device, place one of the suspect parts on the passing board and see if the behavior follows it.

Best,
Jerry