Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DLPNIRNANOEVM: NIRScan nano - Minimal Optical Power it may detect

Part Number: DLPNIRNANOEVM

Hello all,

I am using  ultra-mobile near-infrared (NIR) spectrometer in reflectance measurements using optical fibers. My signal is very weak, I believe it is very close to the noise level. 

I need to define if I am measuring reliable signal or it is just noise. I have access to Optical Power meter thus my question is:

What is minimal optical power that I could detect? Preferably in Watts. If it is not possible to define it exactly, could you please provide the range, eg. nW, µW.

So far I have found SNR equal to 6000:1. However I cannot  use it (or I don't know how to adapt it) for my puprose, link below.

Thank you and best wishes, Lucas.

  • Hi Lucas,

    Welcome to the DLP forums and thank you for your question. Can you describe your optical fiber setup in a little more detail?

    - What type of light source are you using?
    - What size fiber are you using?
    - How are you coupling the fiber to the NIRscan Nano?

    It is difficult to say what the minimum optical power for detection is, it will depend on wavelength and you also don't want to be right at the minimum - you'll have very poor SNR. In general, I would guess that the limit is on the order of milliwatts.

    When measuring a signal, I would feel comfortable with raw intensity values of greater than 50,000, although brighter is better.

    You could attach some of your scan CSV files for me to look at. That will help me understand how low your signal level is.

    -Paul
  • Hello Paul,

    Thank you for quick answer.

    I am currently using 15 W IR halogen lamp, which should be stronger than built-in two small IR lamps. Fiber core in my setup has 125 µm of diameter. On one side of the fibers I have just 'raw' polished fiber tips, lets call this detector's side. On the other side those optical fibers are connected to the module and lamp using SMA 905 connectors. No lenses / sophisticated coupling.

    Yes, I understand your point with dependency on wavelength.  As far as I checked I should not receive more than 100 µW on this range.

    50, 000 counts in reflectance measurements? Is this possible to achieve with optical fibers? I am using Spectralon as a standard sample for calibration and even with this I cannot achieve amplitude higher than 1,200 units. Please find CSV file and a plot attached.

    To be precise, 50,000 counts on average or peak points? I am now not sure if I got that much even with transmittance measurements. 

    Could you please provide me some hints how I could improve signal strength? 

    Thank you again!

    Best, Lucas.

    Sample1.dat

  • Hi Lucas,

    From your attached graph, I too am concerned that your spectum is too noisy.

    The biggest reason that you're likely not getting the desired performance is the matching of the slit to the fiber. The slit is 1.8mm by 25um. Your 125um fiber is only filling a small portion of that slit, so not much optical power is getting into the optics of the Nano. If you have access to a fiber which is round on one end and linear on the other (stacked small fibers), that would be ideal.

    When you take a scan, are you getting the PGA gain as high as you can? Do you get better results with Hadamard vs Column?

    -Paul
  • Hi Paul,

    sorry for delayed answer but I needed some time more time to test. First of all I updated software after your idea to change PGA gain, which I couldn't in previous version.

    I checked Hadamard with biggest possible PGA (equal to 64). I believe it slightly improved SNR as the signal become less noisy. Meaning amplitude of 'spikes' decreased. However overall amplitude of the signal is not much better. I would say it improved by 200 units/counts. Meaning it is equal to ~2300 units/counts in highest peak.

    I think your solution with optical fibers may be the key. However, before I start I would ask you, if your team in TI conducted successful reflectance measurements with Nano, using optical fibers. If detector is able to detect signal on milliWatt range I'm not sure if I will provide much more than µW even with round-to-flat optical fibers. If you already did it, could you please provide more details on how measurement setup was made, e.g. fibers used(core size, type), light source(optical power), distance from the sample to fibers, sample type, reference sample type.

    Another theoretical solution would be changing the photo-detector, would you recommend this solution? I believe I would need to change ADC as well.

     

    Thank you for your time and help!

    Best, Lucas.

  • Hi Lucas,

    My range for optical power was a guess - it very well could be microwatts. Ordinarily, we use fiber input for transmissive type measurements and with a 20W light source can easily saturate the detector.

    Let me ask my colleagues about any testing done with reflective fiber setups. I know we have some of those fiber types in the lab, but I personally have not done any work with them.

    You could change the detector, but that's a fair amount of work - you'll need to adjust the compensation values and possibly the ADC as you mentioned.

    Best,
    Paul
  • Lucas,

    We have done some testing with reflective and fiber input. For our setup, we used a Thorlabs bifurcated fiber. The setup is drawn here (bottom of page):
    e2e.ti.com/.../3502.sampling-techniques

    The light source was a 20W source from Ocean Optics and the cable featured ~37 fiber cores, with 7 being used to direct light from the sample to the Nano (stacked linearly).

    One other thing to note is that for this type of measurement, you do not want contact the sample. There needs to be some distance to allow the light to diffuse and reflect into neighboring fibers.

    Let me know if you have further questions.
    -Paul
  • Paul,

    thanks a lot for your help. You clarified several issues. I hope that I will be able to push it from this point. 

    Best, Lucas