OpAmp Output Not Matching Expected Gain 2 With ±9V Supply and Dual Pot Inputs

Hi all.I made a simple circuit (see pic) to try and clear up my confusion with opamps. Eg, does the + input get amplfied as well as the - one, and how? That sort of thing. I just wanted to see it all happening in practice.Unfortunately.... guess what, I'm as confused as ever.
It runs off +-9V. The 1K pots for the inputs have a 1.5K res on each end so I can get about -2V to 2V from each. The gain should, I believe, be 2.
The second table contains what I thought the output should be if the gain was 1, ie the - input inverted and added to the + input value.The third table shows the same thing, but with a gain of two, ie everything doubled.
In the first table is what I actually got, ignoring some decimal places in the measurements. Not very similar to table two...
Can anyone please explain??Thank-you very much again!Graham.

Gain from NI to Out = 1 + R2/R1 = +2.Gain from I to Out = -R2/R1 = -1.This agrees with your measurements.

Graham...Peter is right, the measured values DO agree with your calculated values...up to a point.You are trying to make a difference amplifier and you have not got the circuit completely right.This is what you should have (from the Wikipedia Article):
Ideally R1=R2 and Rf-Rg, all 1 Meg in your case.The fact that you don't have the voltage divider on the + input will give your weird results.I did the following in Excel (sorry, it doesn't post really well here) which agrees with your results.The formula I used was =2*(A$n-0.5*$Bm) which equates to 2 x (NI -0.5xI) and lo and behold I get your results exactly:

I/NI	-2	-1	0	1	2
-2	-2	0	2	4	6
-1	-3	-1	1	3	5
0	-4	-2	0	2	4
1	-5	-3	-1	1	3
2	-6	-4	-2	0	2

Why do I multiply the NI input by 0.5 when you are not dividing it? Because we are looking at the DIFFERENCE between the two inputs.  Dividing the NI input will result in a bigger DIFFERENCE between the two inputs.   Why are your results correct for NI=0V?  Because 0 x 0.5 still = 0, so the lack of the voltage divider on the NI input does not affect the calculationI'd refer you again to the TI Op Amp Circuit Collection which gives standard circuits for many applications.  Get these standard circuits in memory, or refer to them, and you should not go wrong.  Departing from them, as you have seen, can give weird results if you don't know exactly what you're doing. I've fallen into very similar traps myself before 

Hi Graham, I would agree with what David has said.
Except my go to book for OP_Amp theory and practice has been George Clayton's Operational Amplifiers now in its 5th edition. Mine is a 1st edition but still seems relevant and easy to read and understand.Happy analogue computing.

Ok, thanks for all the advice. Maybe I'm too er un-young for a proper intuitive grasp, but I'll use those books and chalk it up to not re-inventing the wheel!Cheers,

Hi Graham-You might want to take a look at Max's article: Free super cool analog design tool. There's lots of good information hiding in the free design tool.Also- Tomorrow (Tuesday, May 21st.) ST Electronics is presenting a Webinar: Understanding operational amplifier parameters and accuracy measurements. 
This webinar will cover:
  Operational amplifier basics
  The key op amp parameters affecting precision
  Vio - input offset voltage
  CMRR – Common Mode Rejection Ratio
  Iib – Input bias current
  Noise
  Other key op amp parameters
  Zero-Drift vs. standard precision amplifiers - pros and cons-Rick

Graham if you are playing around with OpAmps on a breadboard (or even a simulator) then stick to the standard configurations as per any of the references given.  These are "building blocks" that you will use over and over again.  Once you have the standard inverting and non-inverting amps (and maybe differential amp) done you could try a Summing Amplifier.With opamps, bear in mind that because of the super-high gain, the opamp will act to keep the voltage between the inputs at near zero in a well-designed circuit..  This is nicely illustrated by the summing amp - because of the "Virtual Earth" caused by grounding the + input, the inputs don't interfere with each other.It occurs to me that the Electronics Tutorials site's series on OpAmps is great - and it's also an Aspencore site    Start here, and maybe work through their circuits on your breadboard.  And Electronics Tutorials is a great site for info on anything to do with electronics generally.Conrad's recommendation for George Clayton's Operational Amplifiers is also good.Following a similar recommendation from him some time ago I got a nice copy off EBay for a few dollars. It does get pretty deep compared to say the Electronics Tutorials site, but as a go-to reference it will have pretty much anything you want to know.As Crusty (aka Conrad) says... Happy analogue computing  - enjoy.  Crusty and I are both un-young as well