PCI 6363 connector box 2110 indicates higher than actual voltage

[Shalvan]

Hello,

I am building an optical autocorrelator setup and I am using PCIe-6363 DAQ card (http://sine.ni.com/nips/cds/view/p/lang/pl/nid/207412) with two BNC-2110 connector boxes for both AO (adjusting the time delay via piezoelectric stage) and AI (acquisition of the signal). My photodetector is a photodiode, which can work in reverse polarity (12 V).

My trouble is that when the photodiode works in reverse polarity, when I use an external oscilloscope to acquire the autocorrelation signal, my measured voltage is no higher than 13,6 volts - which is the maximum voltage the diode can return. Of course this value is above the limit of DAQmx, so I reduce the amount of light in the setup. But even when the voltage is in range (let's say around 7 volts maximum), my program indicates voltage above the limit (showing constant 10 volts). Only when I reduce the amount of light so much that oscilloscope shows voltages in milivolt range I can acquire the signal using the connector box (and it indicates it is in range of 5 volts, which is riculous).

Thinking that something is wrong with my VI, I discovered it is unoptimally written and I tried coding it a different way (which hastens the measurement), but then I still get 10 volts unless I use NRSE measurement mode, then I get the signal, but monotonicall falling, and after around ~5 measurements the signal saturates again.

What is disturbing for me is that all programs and NI MAX test panel indicate 10 volts ambient voltage (when nothing is connected), which only falls to zero if I short-circuit it. This 10 volt indication appears in both connector boxes, in every connector I tested.

As attachments my regular VI and the test VI.

[stockson]

Dear Shalvan,

First of all, your testing VI is opening and closing references in each iteration which is not a good idea. It would be the best to use one of the examples installed with DAQmx to test your device. 

Second, measuring +/-10V when nothing is connected is perfectly normal. You have infinite impedance on the input (air).

Anyway, to help you find solution to your problem, I will need an answer to the following questions:

1) Did you try to connect a battery to the 6363?

2) What are the FS/GS switch positions on your BNC-2110 terminal blocks?

3) Are all the devices properly grounded?

4) How are the signals connected? You wrote about changing the mode to RSE, but nothing about changing the physical connectors.

Best regards,

[Shalvan]

Thank you for the answer. The problem was solved by connecting a 7,5M resistor in parallel. I believe the connector was charging, so what was needed was a sufficiently high input impedance so that it wouldn't kill the signal while allowing the "condensator" to discharge.

To answer your questions: I did use test VI's, getting the same results. Connecting a battery yielded correct voltage. The switches are set to FS. The signal source was a photodiode, not grounded.

Looking at the second link you provided, that was probably the problem, as without grounding the voltage was accumulating. The current solution isn't perfect, but it works.

Honestly change the mode to RSE was an act of desperation, when I was using every option to check if it'd work.

[stockson]

Thanks for sharing.

You can compare the input impedance of the scope with the impedance of PCIe-6363 (its >10GOhm). It should confirm your suspicions.