# Multisim 8 question - negative voltages?

• posted

Hi!

I am trying to simulate some kind of regulated power supply using Multisim 8.

Look at the circuit (It amplifies voltage at 200 miliohm serial resistor)

It is minimal circuit which shows my problem.

As you can see at picture - LM324 produces negative voltage relating to ground! But it is impossible - opamp is powered with +12 and 0V voltages.

When i change LM324 with 'virtual opamp' it works as expected - there is no negative voltage at output.

I know that LM324 is far from ideal but i think it doeesn't have ability to produce negative voltages, especially when there is no capacitor or inductor in this circuit.

My question is - what am i doing wrong? What should i set in Multisim to have correct simulation ?

greetings Marcin

• posted

in

Who knows- are you doing DC analysis with an unrealistic limit on the maximum number or iterations? IIRC, the LM324 is problematic for SPICE in general, and the real part has a tough time getting Vo below 0.7V while sinking currents in the 3mA range anyway, and your circuit requires it to go to 0.24V- not going to happen.

• posted

resistor)

How much negative voltage? Spice models like to include input offset voltage, and often they use the part's worst-case spec'd value. As for the output range, most opamp models fail to accurately limit near the part's voltage extremes, which is ground in the case of the LM324. If you study the model, you'll see why it goes negative. But if the model has any integrity at all, it should be setup so it can't go far.

• posted

resistor)

Almost -4V at a gain of 1....

Spice models like to include input offset

Input right at Vcc/2...

Not a viable suggestion....

• posted

You have a bad simulator *and* a bad opamp model. A real-life circuit can't do this.

John

• posted

Dnia 10.02.2007 John Larkin napisa³/a:

So which simulator should i use to have proper results? Could you recomend some software to me ?

• posted

napisa³/a:

• posted

I don't agree with John's remark that your results are due in part to a poor spice simulator. It's the opamp model you're using that's at fault. The best simulator package (most use the same Berkeley spice engine, with minor enhancements) can have a few poor models in its library, and even the worst simulator surely has a few excellent models in its library.

Most simulators let you add models to their library, and most serious spice users spend plenty of time seeking better models or making our own. Here, you need a better LM324 macromodel.

One suggestion: make a schematic of your '324 model by translating its netlist to a drawing. This will let you see for example how its output stage works.

One other suggestion: test your models first with some simple test circuits. Then you'll know in advance if you can place much faith in the results.

• posted

What's bad about Multisim (aside from the price) is bait-and-switch business practices, rotten support, and buggy device models.

John

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.