Light to Frequency Converter Sensitivity


So I've been working on small arduino uno based spectrometer in a shoe box.

A quick look on the process is you shine UV light on a specific sample to excite it, and this excitation, in turn, produce light.

At current, I'm using the TSL235 L-to-F converter.

My problem is that when the UV light is on and there is no sample, the frequency read is 0, which is perfect. When the sample holder is placed, the frequency becomes around 3 Hz, also good, however, when the sample is placed it will give the same value of 4 Hz regardless of what the concentration of the sample is (higher concentration gives off more light).

I read the spec sheet, and my guess is the intensity is too low to measure with the sensor I'm using?

There's also the BH1750 sensor that i would like to try. It can detect at lows of 1 lux, however I can't seem to tell which of the two is more sensitive.

Even a basic method of comparing would be more than great.

What wavelength(s) of UV are you using for excitation?

What wavelength fluorescence do you expect the sample to emit?

Keep in mind that the TSL235 is most sensitive to light in the near infrared (800 nm) and just looking at the size of the detector, not very sensitive to low light levels. People normally use photomultipliers or large blue-sensitive photodiodes to detect fluorescence.

The UV source is 320nm and the expected florescence is around 375nm to 450nm.

From what you're saying I understand it is very much likely a hardware issue and I will be looking into the photodiode option.

However, is there any way to tell whether it is indeed a hardware limitation? Also, I'm assuming the BH1750 does not work also because it has relatively low responsitivity to lower visible spectrum end wavelengths (around 0.2-0.3). Would the high resolution option compensate for it though?

However, is there any way to tell whether it is indeed a hardware limitation?

Of course, but with the sample you have, you need some other, correctly functioning hardware to compare to your detector.

If your detector responds to room light in a reasonable way, it is probably functioning correctly.

If you look at the design of a commercial spectrometer or fluorometer, you will see that the manufacturers use very sensitive detectors and take great pains to eliminate stray light. There just isn't much signal!

Here is a DIY fluorometer project using a photodiode: DIY Fluorometer for DNA Concentration | PhysicsOpenLab