amplifier gain not stable

[Chengjun Li]

Hi,

I found a differential amplifier board which was made several years ago.


The gain of this differential amplifier is 213.765, but I made a simple measurement and found that the gain decrease as the input voltage increases.

I wonder is this something normal? Does this ampilifer still work?

Thanks.

 

[Harald Kapp]

A few questions:

• Where and how did you connect the input voltage? Am I right supposing between V- and V+?
• Did the voltage souce have any connection to ground? (This is an instrumenation amplifier, it amplifies V+-V-, but both sources need a reference to ground for the opamp input current to return to ground. if the input voltage source is floating, add a 1MΩ resistor from each input to ground)
• Were the opamps powered by a dual supply (the LM324 is not suitable for single supply operation in this circuit)?
• How did you calculate gain?

Supply the complete test setup, please, there is more than one possible source of error.

 

[Chengjun Li]

A few questions:

• Where and how did you connect the input voltage? Am I right supposing between V- and V+?
• Did the voltage souce have any connection to ground? (This is an instrumenation amplifier, it amplifies V+-V-, but both sources need a reference to ground for the opamp input current to return to ground. if the input voltage source is floating, add a 1MΩ resistor from each input to ground)
• Were the opamps powered by a dual supply (the LM324 is not suitable for single supply operation in this circuit)?
• How did you calculate gain?

Supply the complete test setup, please, there is more than one possible source of error.

Thanks for your reply.
Please allow me to give a little more description.
The voltage I want to amplify is the output voltage of a Linear displacement sensor.

This sensor has four electrical connections, two for excitation voltage(here I use 5V DC from a power supply), two for signal output which is the voltage I want to amplify.

Here are the answers of your questions.
1. I connect two signal output of the sensor to the V- and V+ of the amplifier.
2. The sensor is excited by the following power supply. I am not sure whether or not the sensor output is floating, I will try add a 1Mohm resistor and test again.

3. The opamp was power by +12V and -12V.
4. I calculate the gain using the following equation.
Vout/(R1+R2+R3+R4) = (V+ - V-)/(R2+R3)

 

[AnalogKid]

The schematic in post #1 is a photoshopped version of a Linear Tech datasheet image. Where did you get it and where did the gain equation come from? I've never seen anything like that gain equation anywhere, ever. For the circuit in post #1, the gain is 1.212765. Note the similarity of part of the number sequence. Looks like a math error to me.

ak

 

[Harald Kapp]

I connect two signal output of the sensor to the V- and V+ of the amplifier.

This is correct, but you also need to connect the ground of the 5 V supply and the +-12 V supply together to allow input bias currents of the opamps to return to (common) ground.

For the circuit in post #1, the gain is 1.212765

Right.
To the op: The equation is Vout = (V+ - V-) *(R1+R2)/R2 with R4=R1 and R3=R2

 

[Chengjun Li]

The schematic in post #1 is a photoshopped version of a Linear Tech datasheet image. Where did you get it and where did the gain equation come from? I've never seen anything like that gain equation anywhere, ever. For the circuit in post #1, the gain is 1.212765. Note the similarity of part of the number sequence. Looks like a math error to me.

ak

Thanks very much for your reply. You are correct, the gain is wrong due to my mistake. Actually the R1=R4=1000kohm. I deduce the equation by myself by considering the current is same between V+&V- and Vout & GND.

 

[Chengjun Li]

This is correct, but you also need to connect the ground of the 5 V supply and the +-12 V supply together to allow input bias currents of the opamps to return to (common) ground.

Yes, I did exactly like that, I am sorry I forgot to mention that. I do have a return path for the input current. Is there any other reason that could lead to the strange measurement result? I actually did the same thing for another amplifier in this board(LM324N has four op amp and therefore two instrumentation amp), I got the same result, as input voltage increase,the gain decreases. I don't think this is the problem of the opamp chip, there must be something I did wrong.