Amplifying a sense resistor...

[smanches]

This is taxing my opamp skills, as I cannot figure out how to set the gain when you have a differential input.

I've attached each input to either side of the resistor, with normal VDD and GND connections. I can see it working, but the gain is way too high.

Anyone learn me how to reduce the gain in such a setup?

 

[EN0]

Yes, you have two resistors called Rf that bring both positive and negative feedback to the input. Take a look at this link:

https://www.electro-tech-online.com/custompdfs/2009/08/sloa064.pdf

The first schematic you come across shows an example. It is marked "2.2 Gain."

 

[smanches]

I'm using a single output opamp though, since I want the output to be from VDD to GND based on the voltage drop of the resistor. How do you setup the feedback with a single output?

Looked through the document, but there doesn't seem to be this setup.

 

[EN0]

It would be much easier for me to have a visual aid. Could you please make a schematic of the differential circuit you have? A differential circuit needs two things to difference.

 

[Grossel]

Read about differential amplifiers here.

A better circuit hence to input impedance is a instrumentation amplifier. It's described here.

 

[smanches]

Looks like an Instrumentation amplifier is what I need. Thought a single opamp would have done it.

Thanks.

 

[audioguru]

Does your opamp have a single-polarity or a dual-polarity supply?
Some opamps have inputs that can sense a voltage that is at the positive supply voltage.
If the opamp does not have a negative-polarity supply then some opamps have inputs that can sense a voltage that is at 0V.
A new kind of opamp is "rail-to-rail" and has inputs that can sense a voltage at the positive supply voltage or at 0V.

What is the voltage that you want to sense?

 

[Roff]

Yes, you have two resistors called Rf that bring both positive and negative feedback to the input. Take a look at this link:

https://www.electro-tech-online.com/custompdfs/2009/08/sloa064-1.pdf

The first schematic you come across shows an example. It is marked "2.2 Gain."

There is no positive feedback in that circuit. Positive feedback is not generally used in a linear amplifier.

 

[smanches]

I've worked it out using a couple of AD8227 instrumentation amplifiers. They are rail-to-rail with a programmable gain. Also handle wide supply voltages and have very rugged inputs. Since this is going into an auto, I want it to be very robust.

I'm just trying to amplify the voltage drop from a high side sense resistor in a buck converter. At max load it only develops 100mV, which I figure would be too error prone to put straight into a PIC. I'm configuring the amps with a gain of 40, which will make the max load drop 4V, giving me an extra volt to measure the over current itself.

I haven't built it yet, but this seems to be the way it's done, after many hours of research.

 

[Roff]

I've worked it out using a couple of AD8227 instrumentation amplifiers. They are rail-to-rail with a programmable gain. Also handle wide supply voltages and have very rugged inputs. Since this is going into an auto, I want it to be very robust.

I'm just trying to amplify the voltage drop from a high side sense resistor in a buck converter. At max load it only develops 100mV, which I figure would be too error prone to put straight into a PIC. I'm configuring the amps with a gain of 40, which will make the max load drop 4V, giving me an extra volt to measure the over current itself.

I haven't built it yet, but this seems to be the way it's done, after many hours of research.

It should be a lot simpler and cheaper than that. Try Googling "high side current sense amplifier". There are lots of them available.

 

[k7elp60]

Here is a circuit I use on occasion. It doesn't require an extensive instrumentation amplifier.
If you make all the resistors the same value the voltage difference between point a and point b is 1/2 the voltage across the sense resistor.

 

[EN0]

There is no positive feedback in that circuit. Positive feedback is not generally used in a linear amplifier.

Thanks for correcting me Ron!