Spikes in the analog input measurements with NI DAQ 6363 BNC

[SatbirSS]

I am measuring the output voltage of a photodiode Newport 2051 with NI DAQ-6363 BNC using a USB cable on my computer in lab view. During the measurement of optical power stability of laser, i observe unusual spikes and noise in the output voltage, which i am not sure is because of bad laser or some electrical disturbance. I observed the similar behavior with other similar photodiodes and NI DAQ USB 6251. To arrive at some conclusion, i measured the response of photodiode in dark environment and I have observed a few cases as follows:

1) In the dark environment, when i ground the photodiode by connecting it to the power supply and measure using GS configuration on DAQ board and differential terminal configuration in DAQ task, i see spikes in the data as seen in the following image

  the level of these spikes is  reduced when i remove the grounding of photodiode and measure with the battery of the photodiode, as seen in the image

and it is completely disappeared when i measure using FS configuration as seen in the following image 

what could be the possible reasons for this behavior as these might be related to my original problem of noise in optical power of laser. Any help is highly appreciated. Thanks  

[LV_Warrior]

I see bit noise.

You might want to use a transimpedance preamplifier

A photo diode usually makes a very small signal

[SatbirSS]

Hello!!

Thanks for the reply. The photodiode i am using is having gain of the order of 10^6. I thought it be fine for measurement, Do I need to use preamp? I am worried about some sort of electrical disturbance in grounding, which is giving some noise. But i am not able to understand it well. Although, I have grounded my optical table and power supply sockets to the same ground, yet i see this noise.

[LV_Warrior]

Your setup is unclear to me. I'm curious aubout the type of photodiode you use.

I see quantisation noise. That means you're not even measuring dark current.

Are you measuring DC ?

Do you see the voltage change if you apply light ?

[SatbirSS]

Yes, with applied optical signal the voltage increases. The quantisation noise is different in the three cases i shown in the post. Is it usual to observe such changes in noise with different terminal configurations?

[LV_Warrior]

Hi Satbir_SS,

I read your original post a bit better this time. You wrote:

I am not sure if the unusual spikes and noise are caused by a bad laser or some electrical disturbance.

You did excellent tests.

- No, it is not the laser because with shorted diode it is still there.

- So yes, it is electrical disturbance.

When you run the amplifier on internal batteries (pic3) with "floating source" the noise is gone.

When you run the amplifier on an external powersupply (Newport 0901 ?) you introduce a groundloop via PE where common mode current from the mains supply can sneak in.

That dirty current highly is contaminated with HF noise because we all pump in the noise we are legally allowed to dump there.

That current will flow through the connection between the Photoamplier and your AD converter. And generate a differential mode current there.

If your amplified optical signal is large enough it is no reason for panic.

I think you're OK.

Nevertheless, you could try to reduce the current with a common mode inductor handling ALL THREE wires of your symmetrical 15V power supply. This is usually not very effective.

You could at the AD converter side filter out the HF noise. Analog or digitally.

Putting the switch in the "FS" position intends to eliminate the ground loop.

( It is not perfect because it still leaves a resistor to GND needed tor define the DC common mode voltage of your AD converter input. Maybe there is even a small capacitor there totally spoiling the common mode isolation for HF )

[SatbirSS]

Thanks for such an elaborate response. I have also checked the photodiode with another power-supply (Agilent E3631A), and got the similar response. One thing is still unclear to me, if there is high frequency noise in mains supply, it may also appear in the case where i use batteries for photodiode, as the DAQ is still connected to the mains. 

[LV_Warrior]

When you use both floating source and battery supply, the common mode voltage noise is still there, but it doesn't turn into a common mode current flowing through delicate parts of your analog frontend.