Open-loop Gain Measurement

[xdrfvg]

open loop gain

I am trying to measure open loop gain of LM324 in real circuit with this configuration below.
I am using:
R=1Meg (feedback)
C=1Farad (feedback)
The power supply: 15V ~ -15V
Then I input the ac sinewave with 100mvp-p @ 10Hz in the non-inverting input.
I'm trying to get the open loop gain with this calculation: {V(+) - V(-)}Ao=Vout.
So I take the measurement of the output voltage, but I realize that the output voltage is already saturated (it reaches rail power supply).
I think I have something wrong.

Can somebody show me the correct ways to measure open loop gain with this configuration?
And why we need large resistor value in the feedback path?
What is the different if I short it?

Thanks for the answer~

 

[sutapanaki]

how to measure open loop gain

Look for this paper by Dr. R.D.Middlebrook:

Measurement of Loop Gain in Feedback Systems
Intl. J. of Electronics, 38, 485—512, Apr. 1975.

 

[FvM]

measure open loop gain

The 1F capacitor seems to indicate, that you are simulating rather than "measuring". It's O.K. of course, but should be mentioned to my opinion. If so, there are simple methods to determine open loop gain from a closed loop circuit simulation. without needing particular bias tricks.

A loop gain simulation is usually done in AC sweep mode, which implies small signal models and also that absolute signal amplitudes are meaningless. In a real circuit and in transient mode simulation (=large signal), you have to keep linear voltage range of course, so 100 mV input is a much too high magnitude.

 

[xdrfvg]

open loop gain measurement

Look for this paper by Dr. R.D.Middlebrook:

Measurement of Loop Gain in Feedback Systems
Intl. J. of Electronics, 38, 485—512, Apr. 1975.

Thanks. I will look for it.

The 1F capacitor seems to indicate, that you are simulating rather than "measuring".

Sorry for my mistake, it should be 1uFarad, not 1Farad as I mentioned in the first post.

If so, there are simple methods to determine open loop gain from a closed loop circuit simulation. without needing particular bias tricks.

If so, what the methods is it?

In a real circuit and in transient mode simulation (=large signal), you have to keep linear voltage range of course, so 100 mV input is a much too high magnitude.

Yes, you get the point. The problem is how to keep the output voltage in the linear voltage range. If 100 mV is too high, how about 10 mV. It is the smallest voltage that my signal generator can generate.

 

[FvM]

loop gain simulation

So the discussion is about real measurements. To measure a loop gain around 100k, you must be able to generate low voltages below 100 uV and/or measure voltages in this order of magnitude. If you don't have the equipment and don't know how to build suitable measurement circuits, it can't work. You should also consider, that the OP noise most likely requires a frequency selective measurement of output voltages.

Loop gain is Vo/Vi, it can be detremined also my measuring the OP input voltage in closed loop, but it's a uV quantity at maximum gain. Thus I mentioned the method primarly for simulation. With suitable equipment, it can be a real measurement as well.