I need an amp circuit with different gain for + and + inputs

[Mito57]

Hi. I'm new here and need some help. I need a simple circuit (if exists) which amlifying a diferential input gives me an amplified output with 2 different gains depending of the input polarity. I need to have a slightly higher gain when the input is positive. The output should also have the same polarity than the input if possible. Can this be done with some simple circuit using Op Amps?

 

[Sal]

Re: I need an amp circuit with different gain for + and + in

HI

this is a simple solution employing and inverting configuration where the gain for positive signals is Rb/Ra1 and for negative signals it is Rb/Ra2, "remember the polarity will be inverted" so you need to consider this at the time of taken the signal from the amplifier. This circuit has the disadvantage that the signal will drop by about 0.5V at the entrance because of the diodes.

cheers

Sal

 

[Mito57]

Re: I need an amp circuit with different gain for + and + in

Sal, thanks for your answer. That .5 volt drop is an issue for my project. Any other solution? I've been thinking in using a step gain circuit and use a different reference for the input, so all inputs will be positive, and just have a different gain when the voltage reaches the point I want. I know this can be done but I don't have the schematic. Can anyone post it or give me a link?

 

[newelltech]

Re: I need an amp circuit with different gain for + and + in

Here's an idea for you.
Starting with Sal's circuit, remove the diodes.
Connect Ra1 to Signal input. So at this point you have a simple inverting amp with gain of -Rb/Ra1.

Add a voltage follower (IC1) and diode (D1):
IC1 Output to Anode of D1
Cathode of D1 connected to Ra2.
- input of IC1 connected to junction of D1 and Ra2.
+ input of IC1 connected to Signal input.
Hope this is descriptive enough - don't have time for a schematic right now.

How it works:
When the Signal input is < 0 the voltage follower turns off and you have a gain of -Rb/Ra1.
When the Signal input is >0 the voltage follower becomes active and the gain is -Rb/Rx, where Rx = Ra1 || Ra2 so the gain will always be higher for positive signals. You can figure out how to reverse it if need be.

The voltage drop across D1 has no effect on the circuit.
One thing you may want to do is limit the output swing of the follower so it doesn't have to slew back and forth as the input crosses 0.

Again, this is just a concept - you need to go through all the details and test it.