opa 827 amplification

Hello Alasgarzade,

While you did not provide a schematic of your circuit, it is very likely you are experiencing the effects of offset voltage, input bias current, and noise in your circuit. Since 2mV signals are relatively small, every bit of your design effort must ensure that the signal is not covered by these sources of error.

If you are not familiar with offset voltage and input bias current, I suggest you take a look at the video series on those topics provided by TI Precision labs which can be found here.

Using the SPICE simulator TINA-TI, I will show you the basic effects of offset voltage:

Above is the OPA827 in a voltage follower configuration with a high pass filter on the input. Notice how the DC simulation still shows 72.67µV on the output. Looking at the OPA827’s datasheet you can see the typical spec for offset voltage is around 75µV which agrees with our model.

Like said in the video, the mismatch of the op amp’s internal transistors causes the offset voltage which is why you see a voltage with no input voltage applied.

Assuming your circuit has amplification or gain, the offset voltage will be amplified as well. Notice how a gain of 11V/V here causes the DC voltage to show approximately 800µV with no signal applied.

Like mentioned in the videos, there are digital correction circuits and other amplifiers with laser trimmed input transistors to minimize offset voltage. I would recommend these options based on your application.

However, I mentioned input bias current and noise because the videos also talk about tradeoffs. The OPA827 is marketed as a low noise amplifier which also happens to have a low bias current. If you decide to look for a new amplifier, one of these factors may be a problem. In addition, if an amplifier has a good performance in all 3 areas, the cost will most likely increase.

TI Precision Labs has a great video series on noise as well. Assuming you have a frequency of interest, filtering will greatly reduce the noise, and consquentially, the voltage you see at the output. 

Hope this helps.

Best,

-Cole

Hello again Alasgarzade,

You are experiencing extrinsic power line noise picked up from the environment in your lab. Based on what you’ve given me, there should be nothing wrong with your circuit design.

There are many possible sources for this noise. Make sure your oscilloscope is using earth ground as well as your circuit’s power supply. Your measurement devices—such as a scope probe—can act as an antenna that picks up this source of error and couples it into your circuit. Try using different measurement tools or ensure that they are configured to ignore this noise. Oscilloscopes usually have an AC Coupling feature that is configured to filter 50/60Hz noise—see if this fixes your problem. Another source can be fluorescent lighting hitting the package of the IC and coupling into the circuit so try turning off the lights.

This summary of the article titled Top 8 Ways to Deal with Noise in Data Acquistion and Test Systems, can help give you some exposure to possible sources of noise and how to deal with them.

Best,

-Coles