Hi,

Ocean spectrometers are highly sensitive devices. Black cloth may have fine openings in it. make sure you collect the dark condition noise in a pitch black dark room. Turn off the monitor light. Thumb rule while doing any optics experiments, turn off all the light sources in the lab and perform the experiments in the night in a dark room where you have a fine control over the light intensities!!!

Regards

Hmmm …. That’s interesting. Its hard to see with such a small difference, but it looks to me like the extra counts are uniform across the whole spectrum. The whole sensor is counting higher. You could subtract to be sure. If so, that suggests to me that the extra counts aren‘t from light. Oh, it could be light leaking into the box from somewhere other than the entrance, but, frankly, that almost never happens. You could double check by doing it again with the room lights off. If it doesn’t change, it’s not light.

The only thing I can think of that would cause this is temperature. Could it be that when you cover it you are preventing the room circulation from reaching the sensor, and perhaps also trapping heat? The sensor does produce a small but non negligible amount of heat, and, even though the visible spectrometers don’t have fans, they do rely on conduction through the case and room convection to shed that heat.

The dark rate is a combination of read noise (what you get at zero integration time) and dark noise (the part that is proportional to integration time). It is impossible for me to tell how much of the ~ 3700 counts is dark noise, but I assume most of it. Dark noise is a strong function of temperature and it would only take a very small temperature increase to raise the dark noise 5 or 10 counts out of 3700.

By the way, you can be confident that the black cloth isn’t necessary. Spectrometer housings don’t leak light. If it doesn’t enter through the fiber connector it doesn’t enter. (Unless someone has taken it apart and put it back together poorly). Again, you can verify by comparing the counts with the room lights on and with them off.

The quite constant difference between these curves is outstanding ( Mike Albert pointed already to the fact).

So any EM-interference or acoustic/vibration crosstalk (microphony) of the detector with the surrounding may be involved.

Has any change in environmental conditons taken place when doing the two measurement; such as for example any motor*) switched ON or OFF or any vibration?

Occasionally such phenomena are seen in photon counting x-ray detectors/applications.

Checking the background with ambient light ON and OFF is a good idea.

I had such issues when dealing with a low light camera combined with a pretended shielded housing...

Black cloth may not be enough, as mentioned by Amit Morarka.

*) in your lab or the neighbouring lab

I realized there is a glitch in the software of the Ocean FX spectrometer. The data shown previously is when the Ocean View software uses scan to average.

When I wrote a code and did the averaging by collecting individual spectrum of Background and the Dark Spectrum. The spectra looks as expected

Thanks Gerhard Martens, Mike Albert and Amit for all the suggestions and inputs.