Opamp's stabilty in positive feedback

[dushyanth]

i simulated the circuits in an mobile app. below is another circuit with negative feedback

i understand how the second circuit works(-ve feedback) but the first circuit... here is how the -ve feedback works(i think) we apply 1 volt to inverting input the output keeps falling(since the input is given to inverting input) from 0V and as it approaches -1V the inverting input becomes 0V(approx) and becomes virtual ground.now the system is stable but in positive feedback circuit (the first one) the input is positive and so is the output when we feed it back the input only becomes more high and must take the output to +ve saturation but as you can see the simulator is not in agreement with it... is it the simulator fault or is my explanation ?

 

[berkeman]

The positive feedback simulation looks wrong. Did you give it initial conditions? Or did the simulator somehow start with the 1V input to its + input and come up with those current values? The voltage at the + input will not be near 0V. It will start at 1V and then get higher as the output pegs up near the + supply input for the opamp. What opamp model are you using?

 

[jim hardy]

is it the simulator fault or is my explanation ?

at first glance the simulation is at fault. Nobody would build that circuit with any hope of its working as an amplifier.

There does exist a phenomenon called "Latchup"
wherein an op-amp that's overdriven, as yours would soon be, reverses the sense of an input because some internal transistor saturates.
That causes the circuit to drive to one limit and stay, ie latch-up.

I daresay nowadays most opamps are "latchup-proof " but in the 709 days we had to watch out for it.

Also, the 709 could latch up whenever the input CM voltage rose high enough to saturate the input stage. And, excessive differential input voltages could blow out the input transistors. Although savvy users could work around these 709 application quirks, it sometimes took extra parts to do it. So, in one sense the above use-related issues served as a general lesson towards the necessity of bullet-proofing an IC op amp. against various application stresses.

http://www.elektropage.com/default.asp?tid=177

That's doubtless why Berkeman asked what opamp you used. Maybe your simulation program is realistic enough to catch that quirk.

 

[meBigGuy]

There is something wrong with the precision. In order to get -1V out of the inverting opamp you need 10uV on the input. It shows that on the non-inverting opamp, but not on the inverting example.

It seems to be a 100KV/V voltage controlled voltage source, not an opamp model.

So how does 100K times 0 give -1V?

Also, the currents at the positive node of the positive feedback example do not balance out. There is 1V + 10uV across one resistor and 1V -10uV across the other.

You can mess with pspice so it doesn't converge properly. Or, you can misinterpret what you are seeing and draw it incorrectly.

 

[zoki85]

i simulated the circuits in an mobile app. below is another circuit with negative feedback

What has happened to Kirchoff's Law for currents in/out OP amp here?