Help grounding analog input

[evanmacq19]

Hello,

I am trying to hook up a voltage-biased photodiode to the analog input of my DAQ. The photodiode is home-built and grounded to my optical table (which is itself grounded). When I hook up my photodiode to an oscilloscope (grounded to the electrical outlet), I get a steady signal that fluctuates between some maximum value and zero as I open and close my laser shutter. When I hook it up to the DAQ, however, the signal does not go to zero when the laser is shuttered. In fact, it only drops to a minor (~10%) fraction of its unshuttered value. However, whenever I hook up to both the DAQ and the oscilloscope simultaneously (using a BNC T to split the signal), I get the proper reading on the DAQ. 

I have tried the various combinations of single-ended/differential input and differential/RSE/NRSE configuration with no success. Any ideas what's going on? I think it has to do with the various components not sharing the same ground, but I am not sure how to fix that.

Thanks,

- Evan

[johnsold]

Evan,

You did not say which DAQ device you are using, so the suggestions are general.

Do you have the DAQ AI Ground (possibly called something slightly different) connected to your ground? Some devices may have an AI Sense line as well.  Is your DAQ AC coupled at the input? Most are not, but some have capacitors which could be charging to your bias voltage. Can you post a diagram of all the connections?

Lynn

[evanmacq19]

Sorry. My DAQ is a BNC 2090A.

An image of all the connections is attached. I don't see an AI Ground or an AI Sense port on my device. There is one listed on the pinout, but I am not sure how to access the pin out terminals. 

Thanks again,

- Evan

[johnsold]

The BNC-2090A is not a DAQ device.  It is just a connector.  The DAQ device is at the other end of the shielded cable going to the computer.  According to the manual the BNC-2092A has several possible grounding options.  Look at several pages in chapters 1 and 2 of the User Manual.

Regarding your image of the connections: The PC is probably grounded through its power supply if it is a desktop computer and may or may not be grounded if it is a laptop. There is also some kind of ground connection between the PC and the DAQ device/connector block. Your problem may be somewhere in those ground connections.

The grounds connected to optical benches are often not very good grounds when it comes to measuring small signals. I have seen some which had very long and very noisy grounds.

Lynn

[evanmacq19]

Thanks,

I sorted out the problem by terminating the signal at the BNC-2090a with a 1 megaohm resistor. Not sure if this was the best way to resolve the issue, but it is working 🙂