0
\$\begingroup\$

I made two simulations of real opamp and ideal opamp, why are results so different if other params are the same?

Ideal Real

Model I took from maxim site http://www.maximintegrated.com/en/design/tools/modeling-simulation/spice/operational-amplifiers/orcad/MAX4239.LIB

\$\endgroup\$
1
\$\begingroup\$

There are three obvious problems with your circuit.

1) Assuming you want a differential amplifier, you need to feed back the output to the - input, not the + input. To see why this makes a difference, add a little AC to V1, and watch the output swing from rail to rail. As part of the simulation process, the simulator has produced a solution for inputs/outputs at turn-on. This does not imply that the solution will remain stable for changes in inputs, and if you change them you will see that your circuit does not have a gain of 10.

2) You are grossly violating the MAX4239 input common mode limits. Your inputs should be no more than Vcc - 1.3 volts. See the data sheet.

3) You are violating the maximum supply voltage rating for the MAX4239. Maximum total supply voltage is 5.5 volts.

\$\endgroup\$
0
\$\begingroup\$

Op amps have offset voltages and leakage currents. These will be different between different amplifier designs, and the models should reflect this. If you use different models, expect slightly different results.

\$\endgroup\$
  • \$\begingroup\$ Thanks for reply. Well, I suppose there are difference between real and ideal op amp, but it cannot explain my results. Op amp configured on gain 10, and ideal op amp is working perfectly here, it have a 10 gain, but real amplifies signal by 50. May be something wrong with the maxim model, in pspice a didn't get any errors. \$\endgroup\$ – cy8g3n Feb 22 '15 at 19:28
  • \$\begingroup\$ CybGeen, you have positive feedback in your circuits (no negative feedback), which never can work in practice. However - please note that such a problem cannot be revealed using an ideal opamp for ac simulations. To compare both circuits - and to proof stability properties - you must do a simulation in the time domain (TRAN simulation) \$\endgroup\$ – LvW Feb 23 '15 at 9:17
  • \$\begingroup\$ CybGen, from what follows that the gain is 10 or 50? It seems to me you haven't done any analysis except for DC Bias point calculation, or am I wrong? As LvW already noted your main problem is you have mistaken OpAmp inverting and non-inverting inputs, so it can't work as you expect (that's why I think you haven't simulated it yet; AC Sweep or Transient analysis) \$\endgroup\$ – Eric Best Feb 23 '15 at 17:21
  • \$\begingroup\$ Gain should be 10, and ideal model works fine here, I went over limit op amp input voltage. WhatRoughBeast answered about it. \$\endgroup\$ – cy8g3n Feb 23 '15 at 18:52
  • \$\begingroup\$ CybGen, OK, but you have still mistaken inverting and non-inverting inputs in your circuit, so in reality it will not work properly due to instability. \$\endgroup\$ – Eric Best Feb 24 '15 at 19:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.