Hello, so I have two input voltages coming from op amp buffers and I need to sum them. The circuit would work pretty well if it wasn't for the negative feedback on each buffer(I think it's that, I'm just a beginner and I might be terribly wrong). To sum them I'm using another op amp configuration, the summer. Any suggestions on how to proceed? Thank you!

 

hi UG,
The two OPA's as input buffers should not cause any problems.
What is the output from the final OPA, when the voltage inputs are +1v and +0.083 as shown in your circuit.?
E

Try this layout, NOTE the two 100k's on the Non Invert Input

 

The buffers are important to your circuit and without the negative feedback they plain would not work. The important characteristic of the buffers is that as a result of the feedback they have a very low output impedance so their output impedance will have a negligible effect on the accuracy of the circuit. Otherwise you would be at the mercy of the ouput impedance of the voltage sources driving the inputs. Of course if you use the 100k resistors at the noninverting input of your output amplifier as Eric suggests, you probably would not need the buffers at all.

 

Hello, so I have two input voltages coming from op amp buffers and I need to sum them. The circuit would work pretty well if it wasn't for the negative feedback on each buffer(I think it's that, I'm just a beginner and I might be terribly wrong). To sum them I'm using another op amp configuration, the summer. Any suggestions on how to proceed? Thank you!

When you say that the "circuit would work pretty well", that implies that is isn't working pretty well.

So how is it not working pretty well? What is it doing that is shouldn't be doing (or not doing that is shouldn't be)?

Are your opamps spec'ed to operate for input voltages close to the negative rail, because 83 mV is getting pretty close?

What opamps are you using?

What is their current output capability? 1 kΩ resistors are getting pretty low for some opamps. Bump that up to at least 10 kΩ, if not 100 kΩ (in which case you might not need to buffer the inputs at all, depending on the output impedance of what ever is suppling the input signals).

 

1kΩ resistors are a little low for most standard real op amps.
Use at least 10k.

 

Does that sim use ideal op-amps?

Did you build that circuit in real life?

 

Here is a proper summing circuit:

 

But your "proper" summing circuit requires bipolar supplies. The given circuit works with unipolar supplies (subject to min/max voltage specs of the particular opamp).

 

Here is a proper summing circuit:

The sum is equivalent to the mean multiplied by the number of inputs.
The only added complexity is that the op-amp needs a gain equivalent to the number of inputs.

 

As we have been shown there is more than one way to do this. Which one is best depends on your needs.

 

Hello! And thank you all for your replies, I've read them carefully and I understand my question had a little lack of context and information.
So my objective is to add the median values coming from some switches that can be randomly pressed or not, with a single other value that's activated with the same switches.
I've put the 100k resistors as you suggested, but what happens is that on one side of those resistors I have the desired voltages, but after the resistors, on the non-inverting I have an 8V value.
All of the switches are buffered. So that it's possible to use a lot of them without altering the value for each one every time another is pressed, because their outputted value is being regulated by a trimmer.

This is my current circuit, and after I just want to add any number of switches and always have the same result. At this moment I'm trying to have a median value of the values of the switches pressed between the op-amps and the 100k resistor. And it's not happening, I'm sending 400mv on the first and 1.400v on the second but the result is 1V. And the other value witch I'm trying to sum to it is regulated by another trimmer and it's variable between 1 and 5V.
After the 100k resistors the value just jumps to 8V. I've tried several things, including some witch I've seen from your answers. But I think I might be doing some other thing wrong. And on the output of the op amp I have 2.2V no matter the values inputted.

I'm using the lm324 op amp. I have this circuit built and it gives me more or less the same results on the meter as this simulator, the op amps on the simulation are not ideal, they are the same lm324 I'm using.

Thank you all for your help again!

 

Hi UG,
I would recommend you read this link ref Non Invert Summers.
Scroll down about 50% on the linked pages
When ready, set up and post the last OPA stage of your circuit, with just two known input voltages.
Say 1V on each input and measure the Vout of the LM324.
E

https://www.electronics-tutorials.ws/opamp/opamp_4.html

 

Hi UG,
This LTS sim may help you follow the operation of the Non Inv Summer.
E

 

Are you sure that it's the MEDIAN that you wish to add? Electronically, selecting the median is not easy and will involve a lot more circuitry. Are you sure it's not the MEAN that you want?

 

I looked in Google for Summing amplifier and found a few tutorials made by Engineers and found hundreds of videos on You Tube from students (most from one country) that get things wrong.
I even found the schematics here in posts #12 and post #13 made by students in 2 countries.