Real Time monitoring

I think it may help to break this down into three questions:

1. What advantages or capabilities does DLP technology bring to real-time monitoring spectroscopy applications?
2. What is necessary to achieve a real-time monitoring spectroscopy system?
3. Can spectroscopy using DLP technology detect specific things? (conductivity and glucose)

Please let me know if you have any questions on the below answers.

#1:

The main advantage of DLP technology in spectroscopy for real-time monitoring is that it can be adaptive. While a particular pattern sequence may be perfectly adequate for a general purpose scan, the adaptability allows the user to use already existing information about the material that is likely to be scanned to optimize the patterns to extract particular information. For instance, in the DLP NIRscan Nano and the DLP NIRscan, scan configurations can be set up which alter the number of patterns used, the scan type (Hadamard or sequential column scan), and the wavelength extents to measure. You can imagine a system which monitors an expected substance on an assembly line, and when a discrepancy is found, it can take a higher resolution scan without an operator having to remove the sample to a different instrument to see what the problem is.

Other patterns and scan types are also possible, which can allow very fast clustered differentiation between expected materials. In this way, it is similar to compressive sensing in that you use knowledge of the likely objects which you may encounter to reduce the number of patterns necessary to differentiate between them. An experiment and analysis of this technique can be found in an SPIE paper given at Photonics West 2015: "Techniques and applications of programmable spectral pattern coding in Texas Instruments DLP spectroscopy"

#2:

Two pieces will be required: Physical sampling mechanics and optics, and a computing system to analyze the results and take any necessary actions (executing additional scans, halting a production line, alerting an operator, etc.)

The DLP NIRscan Nano includes a reflective sampling head, while the DLP NIRscan includes a transmissive sampling head. For prototyping purposes with the EVM, both incorporate mechanical mounting features which can be used to design your own sampling head for other uses (fiber illumination, etc.). Full optics and mechanics for these systems that show these mounting features so you can investigate a custom sampling head can be found in the TI designs: NIRscan TI Design, NIRscan Nano TI Design.

A computing platform would communicate with the DLP system in a variety of ways for a final product. Again, for prototyping purposes, the DLP NIRscan can be communicated with via Ethernet or ethernet over USB via RNDIS drivers. The NIRscan Nano can be communicated by USB (HID protocol), UART, or BLE.

#3:

In general, TI has provided an instrument design and evaluation modules for spectroscopy, but we are not chemometric experts. With that said, the important factors for those experts to understand what DLP spectroscopy systems could sense are listed below:

• The DLP4500NIR DMD is compatible with wavelengths from 700-2500nm. The DLP NIRscan EVM only covers 1350-2450nm, but a different optics design (possibly requiring a different type of single point detector) could cover a greater or different portion of the DMD's compatible wavelength range.
• The DLP2010NIR DMD also is compatible with wavelengths from 700-2500nm. The DLP NIRscan Nano EVM only covers 900-1700nm, and likewise, a different system design could cover a different portion of the DMD's compatible wavelength range.
• System spectral resolution of the EVMs is roughly 12nm for the DLP NIRscan EVM and 10nm for the DLP NIRscan Nano EVM. This resolution is primarily limited by the ability of the optics to clearly image the slit to the spectrum-dispersed image of the slit at the image plane incident on the DMD. Other dispersive designs could make different system tradeoffs for size/cost/resolution/signal level.

Also, the TI design links above provide some test data that we have gathered for the EVMs, which show some example spectrums.

Jim,

The nano reflectance head is indeed removable.  There should be a small Allen head bolt holding it on.

Fizix