# Problem in understanding DC offset voltage in OPA847

#### Sanket Chaudhary

Joined Sep 15, 2020
6
I am working with the OPA847 for learning op-amp theories. I am using the PSpice model available from the Texas Instrument resource. I am trying to evaluate the offset voltage. As per the theory, if I ground the inverting and non-inverting terminals (with +/-Vbias), I suppose to get amplified offset DC voltage (0 in ideal case!). Instead, I am getting a constant 4.452V DC. I have tried to nullify it but nothing works. I don't know if the problem is from a model or something. I have also attached the model I am using.

Any help would be appreciated.

Thank you.

#### Attachments

• 4.5 KB Views: 0

#### BobTPH

Joined Jun 5, 2013
4,899
Since you have no feedback, the op amp has a gain of about 100K to 1M at DC. An offset of several microvolts would send the output to one rail or the other, which is exactly what you are seeing.

Bob

#### DickCappels

Joined Aug 21, 2008
8,707
Better way to connect this is to Ground the + input pin and connect the - input pin to the output pin. Then get out your DVM and measure the offset on the inverting input.

#### Sanket Chaudhary

Joined Sep 15, 2020
6
Since you have no feedback, the op-amp has a gain of about 100K to 1M at DC. An offset of several microvolts would send the output to one rail or the other, which is exactly what you are seeing.

Bob
I completely agree with you. Now it makes sense to me. However, I have tried to implement the unity gain differential amplifier to get more inside. Please see the following schematic. I was expecting the null or very very low o/p voltage as my all 'R' are same and input voltages are also same. I have saperately tried to get the offset voltage from the trick that DickCapples have suggested. I found offset around -23nV. So things found bit complex to understand.

#### Sanket Chaudhary

Joined Sep 15, 2020
6
Better way to connect this is to Ground the + input pin and connect the - input pin to the output pin. Then get out your DVM and measure the offset on the inverting input.
I have tried to make this trick to find the offset voltage. Since OrCAD was showing error as follows:
"Voltage source and/or inductor loop involving V_V7. You may break the loop by adding a series resistance".

So I have implemented it in ADS simulator. I think simulator would not make difference. I have found -24nV as an offset voltage. but again when I have implemented the unity gain differential amplifier as replied to "Bob", things were not making sense when looking at the output.

#### DickCappels

Joined Aug 21, 2008
8,707
Remember, any resistance added to the inputs can result in affecting the observed offset because of input bias currents from the input pins.

#### Audioguru again

Joined Oct 21, 2019
4,680
Your opamp with +500mV input has a circuit and voltages that make no sense.
Your opamp as a follower with a voltage gain of 1 has its output shorted to ground with a 50 ohms load.
The offset adjust circuit shown on the datasheet assumes that the opamp has negative feedback so that near offset null the opamp does not suddenly switch its output to the other rail voltage.

#### pitalokabell

Joined Apr 22, 2022
3
I have tried to make this trick to find the offset voltage. Since OrCAD was showing error as follows:
"Voltage source and/or inductor loop involving V_V7. You may break the loop by adding a series resistance".

So I have implemented it in ADS simulator. I think simulator would not make difference. I have found -24nV as an offset voltage. but again when I have implemented the unity gain differential amplifier as replied to "Bob", things were not making sense when looking at the output.

View attachment 262390
Hai I have tryin like yours, but without R 50Ohm, and the result I got 500mV in the output. Do u get the solution of this problem offset OPA847?

#### pitalokabell

Joined Apr 22, 2022
3
Your opamp with +500mV input has a circuit and voltages that make no sense.
Your opamp as a follower with a voltage gain of 1 has its output shorted to ground with a 50 ohms load.
The offset adjust circuit shown on the datasheet assumes that the opamp has negative feedback so that near offset null the opamp does not suddenly switch its output to the other rail voltage.
Did u has the answer why the output of the op-amp is 500mV? with input + gnd, and - connect to output. Supply +-5V

#### crutschow

Joined Mar 14, 2008
29,785
Did u has the answer why the output of the op-amp is 500mV? with input + gnd, and - connect to output. Supply +-5V
Yes.
You didn't read the data sheet (why do newbies not read data sheets?)
That is not a single supply op amp.
Here's a snip from the data sheet showing the input common-mode range with ±5V supplies:

Note that it can go no closer then about 2V from the voltage rails.
Since your negative rail in this case is ground, then the input can go no lower than 2V above ground for the op amp to operate properly.

So if you add a -5V supply to the negative terminal you should get the correct results.

#### pitalokabell

Joined Apr 22, 2022
3
Yes.
You didn't read the data sheet (why do newbies not read data sheets?)
That is not a single supply op amp.
Here's a snip from the data sheet showing the input common-mode range with ±5V supplies:

View attachment 265637

Note that it can go no closer then about 2V from the voltage rails.
Since your negative rail in this case is ground, then the input can go no lower than 2V above ground for the op amp to operate properly.

So if you add a -5V supply to the negative terminal you should get the correct results.
I have read the datasheet, and I connect the supply to +/- 5V (not just +5V). And the result I got is output 500mV. I just use OPA847 as buffer. But the result has offset about 500mV. I tried in about 4 op-amp, and get the same result

#### crutschow

Joined Mar 14, 2008
29,785
I have read the datasheet, and I connect the supply to +/- 5V (not just +5V)
I apologize. Must have stayed up too late.
the result I got is output 500mV. I just use OPA847 as buffer. But the result has offset about 500mV. I tried in about 4 op-amp, and get the same result
If it's wired as shown, then that voltage is incorrect.
Measure the power voltages at the op amp pins.
Are both the supplies connected to ground?