Photo sensor current too low

[Eugeneo]

This is a little off topic since it is not a compiler question but here goes.

I'm trying to read a sensor that has 40 nano amps max output. Here's my setup.

OPA129 opamp which has an untral-low bias current of 100fA max.

I'm using a 100 meg I feedback resistor, so in theory the output should be 4volts at 40 nano amps (40na * 100 megohm). Instead all my test equipment reads 1-20mv.

Board traces are goood and flux has all been removed.

Standard FR4 PCB board

When using a voltage follower with no gain the output voltage is over 1000mv, and I've tried different sensors.



Any thoughts would be greatly appreciated.

[newguy]

Can you post a schematic? It's a little hard to follow your description of the circuit.

[Eugeneo]

This is the exact setup I have with the exception of Rf being 10^8
The -V and +V have less than 500 nv of ripple at 10v, but I don't think this should matter.
All capacitors are very low ESR.

One thing concerns me, there is a linear voltage output measured at the ouput of the opamp in the mv as the sensor is excited. Is there anything else I could look at?


http://members.shaw.ca/eugeneo/schem.bmp

This is the sensor

http://members.shaw.ca/eugeneo/PDU-S101.pdf

And this is the opam

http://members.shaw.ca/eugeneo/opa129.pdf

[Ttelmah]

Do you really need to use the diode in photovoltaic mode?. This is much less sensitive than using photoconductive mode. Photovoltaic mode, is the preferred mode for really small levels (because of the lower noise), so this may be where you are coming from. The active area is small, so unless your source is very well focussed, a lot of the light may not be influencing the current. What is the source power?.
What is the wavelength of the source?. The 0.14A/W, is a peak figure, and if your light source is down nearer to 200nm, the responsivity falls to only 0.02A/W.
Remember that the cell itself, has a internal resistance (typicfied at 500MR), which is also loading the signal, as well as the current drawn by the inputs.

Best Wishes

[SherpaDoug]

Try substituting a small battery and a large resistor for the sensor. Start with a combination that gives 1000X the current you are expecting from the sensor, a current you can actually measure and verify, and make sure your circuit behaves. Then work your simulated sensor current down to real levels, and finally use the real sensor.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[Eugeneo]

Thank you guys for the help