I'm using an instrumentation amplifier and I'm trying to amplify as:
Last edited:
Instrumentation amplifier gain resistor
- Thread starter nothing909
- Start date
sorry, here's the simulated circuit
Last edited:
I think you need to have the power going to all the op amps for the simulation, even though it is actually a quad which only has one supply for all four in real life.
Also that circuit will not work so well when built due to the tolerance of real resistors, which will cause the common-mode refection to be significantly worse than with the ideal simulation resistors.
Also that circuit will not work so well when built due to the tolerance of real resistors, which will cause the common-mode refection to be significantly worse than with the ideal simulation resistors.
I don't. It's simulated like the real thing. The power is only going in one, which powers them all
Hello,
If not anything else you could troubleshoot by replacing all the AC sources with constant DC sources. You could then check to see if the input offset is affecting the output.
In a single stage, with 5mv input and 1mv input offset we get an error as high as 20 percent. LM324's have max input offset of something like 3mv so that might not be a good choice for the op amps.
I dont know how your simulator simulates the LM324 as to the input offset spec. It may use the same for all packages, or it may use a different value for different parts of the package. A test would show this one way or the other as mentioned above.
I think your requirement of 1v peak output is reasonable if your frequency is 20kHz but you may see limitations if you try to go too high in frequency.
If not anything else you could troubleshoot by replacing all the AC sources with constant DC sources. You could then check to see if the input offset is affecting the output.
In a single stage, with 5mv input and 1mv input offset we get an error as high as 20 percent. LM324's have max input offset of something like 3mv so that might not be a good choice for the op amps.
I dont know how your simulator simulates the LM324 as to the input offset spec. It may use the same for all packages, or it may use a different value for different parts of the package. A test would show this one way or the other as mentioned above.
I think your requirement of 1v peak output is reasonable if your frequency is 20kHz but you may see limitations if you try to go too high in frequency.
I know what the problem is: you need to use a faster op amp.
With the component values you've chosen, your instrumentation amplifier circuit will have a gain of 101. As best I can tell from your simulation plot, you are feeding it a 10 kHz sine wave. The LM324 op amp has a gain-bandwidth product of 1 MHz, therefore the open-loop gain of your op amps is only (1 MHz)/(10 kHz), or 100, so their performance in a high-gain circuit such as this will be rather miserable: the actual circuit gain will be significantly lower than desired.
As a general rule when designing amplifier circuits with op amps, the gain-bandwidth product of the op amp(s) you use must be much greater than the desired circuit voltage gain at the highest frequency of interest, for the circuit voltage gain to be accurate.
You can test this theory by doing another simulation with a lower input frequency, such as 100 Hz; if you do, I suspect you'll see the output amplitude increase to the level you expected.
If you really need to have a circuit gain of 101 at 10 kHz, you'll need to use a MUCH faster op amp than the LM324.
With the component values you've chosen, your instrumentation amplifier circuit will have a gain of 101. As best I can tell from your simulation plot, you are feeding it a 10 kHz sine wave. The LM324 op amp has a gain-bandwidth product of 1 MHz, therefore the open-loop gain of your op amps is only (1 MHz)/(10 kHz), or 100, so their performance in a high-gain circuit such as this will be rather miserable: the actual circuit gain will be significantly lower than desired.
As a general rule when designing amplifier circuits with op amps, the gain-bandwidth product of the op amp(s) you use must be much greater than the desired circuit voltage gain at the highest frequency of interest, for the circuit voltage gain to be accurate.
You can test this theory by doing another simulation with a lower input frequency, such as 100 Hz; if you do, I suspect you'll see the output amplitude increase to the level you expected.
If you really need to have a circuit gain of 101 at 10 kHz, you'll need to use a MUCH faster op amp than the LM324.
Last edited:
yea, you are right, i tried it at 1kHz and now it's working 
Bordodynov
- Joined May 20, 2015
- 3,179
LM324 ==> Ft=1MHz Gain=10kOhm/100Ohm=100 ==> F3dB=1Meg/100=10kHz=Finput ==> Gain10kHz=100*0.7 Voutput=10mV*70=0.7V
