• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 1079 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Gain/Sensitivity of the DLPNIRSCAN

Benjamin Viquerat
Prodigy 120 points


Hi everybody,

I have tried to use the DLPNIRSCAN with another light source.

However, I seem to get less light into the spectrometer as for the initial configuration (halogen lamp+transmission module+lens).

Therefore, it is difficult to measure a proper spectrum.

Is there any way to control the gain/sensitivity of the detector or the integration time?

Thanks for any advice

Best regards,

Benjamin

  • 0
    TI__Guru 54705 points

    Hello Benjamin,

    The EVM gain is set to be compatible with a wide range of inputs (reflective or transmissive, and columns scans or Hadamard scans), which means that it is not optimized for any particular application. 

    What source are you using and how are you coupling it into the system?  The optics are intentionaly very selective (geometrically) regarding acceptance of light into the spectrometer, so this may be a mismatch in the optical interface.

    Fizix

  • 0 in reply to Fizix
    Prodigy 120 points
    Hello Fizix,

    Thanks for the interesting answer.

    If I understand correctly, you're saying that the DLPNIRSCAN should work with lower levels of illumination?

    Anyway, I am using a quite powerful halogen lamp focused on the sample so I guess I do have enough illumination.

    I am collecting diffused reflected light with a fiber that is then put in front of the spectrometer slit.
    The problem may indeed be the coupling of the fiber into the spectrometer...
    What do you mean by 'very selective' ?
    Should I collimate the output light of the fiber to reduce the aperture?

    I find it difficult to find the right configuration since the interface doesn't allow to monitor the signal in real-time...

    Thanks for any advice.

    Benjamin
  • 0 in reply to Benjamin Viquerat
    TI__Guru 54705 points

    Hello Benjamin,

    I see you are doing reflectance which diffuses the light.  You may indeed need to make some adjustments.

    First, the slit is only 50 µm wide, so you may need to use a GRIN lens on the end of the fiber to conentrate the light into the slit.  The kit is re-imaging the slit at the DMD with each color image at a slightly different position due to the diffraction grating.  Coupling the light efficiently can  be a challenge.  You could leave the lens that the lamp was using in place and try to position the end of the fiber where the lamp element was at.

    Version 2 of the software gives you some control of the integration time and both v1 & v2 allowed sample averaging.

    If you download the technical documents for http://www.ti.com/tool/TIDA-00155 it is possible to alter the gain of by replacing some resistors and caps on the NIRscan detector board (dranwing number 2513458). 

    • R2, C14
    • R18, C28

    The product of each pair of resistor and capacitor (R2 * C14, R18 * C28) should be between 1E-5 and 2.4E-5 so that the frequency cutoff is between 6.5KHz and 15KHz, and R2 should be the same value as R18 while C14 is the same value as C28.

    We have tested this circuit with resistor values up to 240K and capacitor values correspondingly reduced to 50-100pF, with good performance.

    If you make this change, please confirm by inspecting the raw data to ensure the maximum values do not exceed the full scale digital value (8388607).

    Also at the bottom of the TI-Design page (00155) under software there is a Design File -->  http://www.ti.com/lit/zip/tidc293 it has all kinds of goodies  in it including a zeemax model that may help with alignment..

    Fizix

  • 0 in reply to Fizix
    Prodigy 120 points
    Hello Fizix,



    Indeed I'm trying to make diffuse reflectance measurements and I find it quite difficult.

    The illumination configuration and the fiber coupling into the slit are important parameters to ensure there is enough light entering the spectrometer.
    I am using a powerful halogen source that is focused onto the sample. This is quite the maximum illumination that my sample can actually bear and I know by experience this should be enough to measure a spectrum even with a low integration time.
    The fiber coupling is indeed a delicate thing to deal with. After some tests, placing the fiber end at the previous spot of the lamp an using the focusing lens of the transmission module seems to be a good way to do it. Nevertheless, I do have the feeling that the coupling is not yet optimized.
    In the best case scenario, the detector is getting a very small amount of signal and the spectra are therefore very noisy...

    Changing the gain of the detector is therefore a solution that I am interested in.
    But I find the "replacing resistor and capas" part quite extreme!
    I am actually afraid of damaging the device...
    If I were to do it, is there a known relationship between the resistor value and the corresponding gain?
    Thanks,



    Benjamin