I'm working on a project where I need to amplify a signal of ~0.9Vdc over 0-5Vdc. The signal source/voltage is coming from one axis of a joystick which is equipped with a 5K linear potentiometer possessing ~355deg of electrical range in place of a 60deg ranging pot. I initially believed that the AD623 would be a proper fit, but I'm having troubles implementing the device. At best I seem to be able to get a 0-3.6V range with a gain somewhere around 4. If I try to up the gain to 5 as I had intended/hoped the device saturates to its maximum output voltage (for the given gain) and stays at this value until the input signal is decreased. The higher the gain (further into saturation), the more the input signal must be decreased to cause a decrease in output.
I've done some reading/research and I fear that my common-mode voltage is beyond that which the AD623 can manage for gains approaching/passing 5. Does this sound like a possible culprit? Furthermore, I can't seem to get a clear answer on how to determine what exactly the common-mode voltage is for my system. From what I understand, I should connect the inverting and non-inverting inputs to ground. Is the resultant output then the common-mode voltage? Forgive my lack of understanding as this is my first adventure into the land of amplifiers.
I've attached a screen shot of the circuit to aid in diagnosing the cause of the device's behaviour. As seen, I'm using a single sided supply of 5Vdc, have the wiper of the 5k pot attached to the non-inverting input and a 20k pot attached to the inverting input. The measured signal range of the 5k pot is ~ 2.089Vdc to ~2.910vdc, as such the 20k pot (R3) is utilized to set the inverting input to ~2.05vdc allowing amplification of the difference.
Am I asking too much of the AD623? Do I have a fundamental design flaw, or is there a better alternative in a PDIP package? Any thoughts or insight anyone may have is greatly appreciated.