Electronic – Passive Attenuator + Level shift simulation slightly off from hand calculations not sure where mistake is

analogcircuit analysislevel-shiftingltspicevoltage divider

I am trying to take a +/- 3.5V signal and shift it into the 0.65V to 2.65V range for input into an ADC with common mode of 1.65V. I have calculated the attached circuit by hand but when I simulate it on LT Spice I am just slightly off. I also have the constraint that R1 = 50 ohm in the circuit. enter image description here

At -3.5V I want an output of .65V but I am getting about .5V. At +3.5V I want an output of 2.65V but I am getting only 2.05V. This is my first time simulating on LT Spice so I am not sure if my calculations are wrong or maybe it is my simulation.
Here is simulation at 3.5V
enter image description here

And this is simulation at -3.5V
enter image description here

Best Answer

You're just wrong with voltage dividers, twice .

In the beginning, if you have to map 7V peak to peak onto 2V you need a 1/3.5 voltage ratio that turns into a 2.5 resistance ratio, i.e. R2=20ohm.

Then on the Thevenin voltage V1, you have the same voltage divider but upside down and so

$$V_1\frac{R1}{R1+R2}=1.65\,\mathrm{V}$$

$$V1=1.65\,\mathrm{V}\times\frac{7}{5}=2.31\,\mathrm{V}$$

From now on you should only design R3/R4 divider to have 20ohm and 2.31V Thevenin equivalent.