I'm using an OPA381 to provide trans-impedance gain from a photodiode. The voltage output from this stage drives a single pole low-pass filter realized via a THS4531 which is also performing single-ended to differential conversion to drive an ADC. The existing system already employs a +/-2.5V supply and from looking at the OPA381 datasheet I thought it would be easy to run it off of this split-rail supply as well; unfortunately this has not proven to be the case.
Connections were originally as follows:
PIN_2: photodiode cathode
PIN_3: photodiode anode & 0V (GND)
PIN_4: -2.5V
PIN_6: output to diff. stage through series 3K3 resistor
PIN_7: +2.5V
A 1M5 resistor and 15pF capacitor in parallel form the feedback path between PIN_6 and PIN_2.
The result provides an excellent AC signal for the system's purposes, unfortunately there is an almost +2.4V DC offset at the output as well, which makes the output unusable. I was very surprised at this since the datasheet states that the device is supposed to be self-nulling.
I tried several configurations, including referencing pin 3 to both the positive and negative supplies, as well as other diode reference configurations. Finally I de-referenced anything from the system's GND (faking a 5V single supply), with the same result.
I am unsure if this is normal behaviour or if the device is simply damaged.
Any insights on conditions which could cause a significant DC offset from this device and any information regarding running it from a split-supply would be greatly appreciated.
(AC coupling is not desirable between the stages, but may be considered if no other solution is available)
