Difference amplifier. Simulation, reality and math... mismatches.

[uliano]

I expected the circuit shown in picture to gain 10 (and it does) to have an output centered at VCC/2 = 2.5V and it does only in reality while the simulation gives a 0 centered output, why???

Also I expected to have an fc = ((R1+R3) * 2pi * C1) = 1/(6.28*2000*47^-9) = 1693 Hz while both simulation and reality give results between 5500 and 5850 Hz. fc is -3dB or half the voltage, right?

[RoGeorge]

- First, you'll need to correct the schematic for the Vref opamp.  I think you meant R5 has to be connected to VCC, not to Vref as it is in the drawing.
- Second, R8 (20k) together with R4+R3 (2k together) will make a DC voltage divider that will make the Out to be centered around Vref/11.
- Third, please also attach the .asc file, so others can download and simulate the same schematic as yours.

[uliano]

Yes, sure, stupid mistake, and that accounts for the difference between simulation and reality. Now I just want to understand how to compute fc.

I considered that the source injects a current that can only go through R1, C1 and then R3 as opamp inputs don't draw current.

here is the.asc

[wasedadoc]

I considered that the source injects a current that can only go through R1, C1 and then R3 as opamp inputs don't draw current.

That -ve input of the opamp is a virtual ground.  Signal current does flow through R2.

[uliano]

So how should I calculate fc with paper and pencil?

[MathWizard]

I'm not saying you're mixing this up, but sometimes I've seen the -3dB point at closer to 0.708,  so like 10^(-3/20), I guess that's from the power relation when talking about deciBel's. Other times it is just 0.501=10^(-3/10), when just taking the log of some ratio. There's some more fine print I have to read on all that, especially sometimes when comparing my calc's to LTSPICE or a book.

[Terry Bites]

fp=1/(2pi*(R1+R2)*C1) but it will be attenuated by more than -3dB a pointed out
use 10xR1 for R2
You may find the attached helpful