Question About A Fractional Gain Op Amp

Thread Starter

Glenn Holland

Joined Dec 26, 2014
698
I've searched the AAC site for info about a fractional gain op amp, but I haven't found anything specific on the subject.

With 100% feedback, the gain will be approximately 1. However, to obtain a gain of less than 1, somehow the feedback would have to be greater than 100%. This might be done by connecting the op amp to a second one which has a gain greater than 1 and feedback its output to the inverting input of the first one. Then the output of the first amp would be fractional.

One idea is to apply a small reference voltage to the non-inverting input (instead of connecting it to ground) and amplify the difference between the inverting and non-inverting inputs.

Another idea is to feedback the output to the non-inverting input through a voltage divider connected to ground, so the difference would always be proportional to the input. I haven't done any calculations yet and at this time, I'm just looking for peer review.

An example of a fractional gain amplifier are "voltage attenuators" which sample a high voltage (such as the 600 VDC input to a traction motor) and provide an output in the range of 0 to 12 VDC which is then inputted to an analog to digital converter.
 
Last edited:

#12

Joined Nov 30, 2010
18,186
One way to achieve fractional gain is to use 2 resistors and no amplifier. Another way is to use the inverting configuration and make the feedback resistor smaller than the input resistor. If you don't like the polarity inversion, add another inverting gain stage with a gain of one.
 

Papabravo

Joined Feb 24, 2006
12,774
In the inverting configuration an opamp can have any gain in the range 0 <= | G | <= M.
In the non-inverting configuration an opamp cannot have a gain of less than 1, so 1 <= G <= M.

I use the symbol M, to represent any large gain limit that exists due to practical limitations like the power supply voltage or the maximum slew rate, and the absolute value function makes the gain a positive number. To be pedantic about it I could have written that -M <= G <= 0, where M is some large positive number.
 
Last edited:

GopherT

Joined Nov 23, 2012
8,012
Use a voltage divider and feed that into the non-inverting input and and connect output to inverting input. A unity gain, potentiometer with low impedance output.

image.jpg
 

Papabravo

Joined Feb 24, 2006
12,774
Use a voltage divider and feed that into the non-inverting input and and connect output to inverting input. A unity gain, potentiometer with low impedance output.

View attachment 104876
Works best on a very low impedance source, otherwise -- not so much. To alleviate this problem, as #12 suggested, use a unity gain inverter followed by your attenuation factor with inversion.
 
Top