Opamps not at the write outputing right

(Sorry, my knowledge is a little rusty, just take this a preliminary explanation until some real expert shows up )

You are just observing the difference between an ideal OpAmp and a "real world" OpAmp. Your expectation seems to be - based on the ideal OpAmp - that if there is no difference between the voltages of both inputs, then a voltage difference of 0V is amplified and should result in an output voltage of 0V.

However, due to imperfections of a "real world" OpAmp, the OpAmp internally "sees" a tiny voltage difference between both inputs ("Offset Voltage"). The amount of this voltage varies between different kinds of OpAmps, but it is never 0. I would have to look up the values for your OpAmp, but maybe 1mV is typical for a "standard" OpAmp.

Anyway, the OpAmp tries to amplify this tiny voltage by its open loop gain (since you have not provided any feedback path from the output to the input), let us assume 100000 as a typical value. So 1mV * 100000 = 100V ! Clearly, this is not feasible, so the OpAmp does as much as he can and drives the output into saturation (in your case, -VCC).

 

Thats exactly right....and dont ber surprised if you find one that goes to the positive supply....

i came across the profiles for opamp input offsets the other day and it showed that a batch will typically have a range which looks almost gaussian in profile.
the max voltage offset is what is quoted in the datasheet.....however, you may find one that sits bang on the middle but it would be rare....