Op-Amp voltage follower output different from input

[jgauthier]

I am using this op-amp. I have a positive rail (about 10V) and a negative rail (about -13V) explained it in this other post.
The op-amp is basically being used a voltage buffer.

In my application I have 10 ADC inputs reading current transducers. I am offsetting the current transducer by 1.65V (half of my ADC voltage reference). I felt like an op-amp would be a good use-case for this. the 1.65V comes from a voltage divider after a 3.3V regulator.
Since there isn't any load on the 1.65V, I felt like a voltage divider should be okay. I used the op-amp to remove the impedance from the voltage divider.


Now, that said, the op-amp input is 1.65V, but the output is 1.8V. I am currently unsure why this is. Could it be the imbalance on the +/- rail? I am also using the Op-amp as a buffer for 1.65VAC signal, and the signal is equal. Maybe there is something else after the op-amp that is somehow adding 200mV. I need to explore the circuit more. I wanted to put this out here for any other tips I might have missed. Thanks!

 

[crutschow]

the op-amp input is 1.65V, but the output is 1.8V.

What is the output with 0V in?

 

[jgauthier]

What is the output with 0V in?

Hm. Well, I removed R22 and R23, for the voltage divider input. I verified on the schematic that was not used anywhere else.
But, the input showed 1.7V and the output now showed 3.9V.

Pictured here, pins 1,2,3 are the ones I brought up in the initial post.
Pin 5 is 1.65VAC that is obtained through the voltage divider shown in the original post with R2 and R12.
Pin 10,9 are used as a summing input for 1.65VAC offset with with buffered 1.65VDC.
Pin 8 is the output of that in ADC range (0-3.3V)

I removed the op amp, and as I suspected the place where pin 3 was in 0V.
So, something else is wrong causing the op-amp to have a voltage on pin 3, but since there are no other connections it must be happening inside the op amp?

 

[Alec_t]

Leaving input pins 12 and 13 floating is not a good idea. It could lead to unpredictable effects, such as oscillation.

 

[jgauthier]

I removed all the other inputs, and put the 1.65V back into pin 3. 1.8V comes out. Still.
I checked the datasheet and it agrees with you:
"Any unused channel of the ADA4522-1/ADA4522-2/ADA4522-4 must be configured in unity gain with the input common-mode voltage tied to the midpoint of the power supplies"

I need to see how I can potentially run a jumper(s) for that.

 

[jgauthier]

Unfortunately, utilizing the last channel did not change the result. I connected my 1.65V input to the unused channel, and the output is also 1.8V. I can however, easily input 0V to that channel to test it. But I have to put it down for now. Thanks for the help.

 

[bertus]

Hello,

Have a look at this page:

The Unused Op Amp—what to do?

I’m not referring to op amps in your parts bin. Those should be in anti-static bags or conductive foam. What about the one on your circuit board—the unused...

e2e.ti.com

Or this page:

Avoiding Noise and Power Problems with Un | Maxim Integrated

If connected improperly, an uncommitted op amp leads to problems such as increased power consumption and added noise. This application note shows how to properly terminate an uncommitted op amp.

www.stg-maximintegrated.com

Bertus