8
$\begingroup$

I am trying to establish an appropriate distance between a spectrograph output window and a CMOS sensor.

The spectrograph creates a focused image on the CMOS sensor but I am trying to restrict the light as the initial pixels are capturing light of higher intensity from 0th order wavelengths. Post processing is possible, but I would rather physically restrict the wavelengths.

I need to block out some of the pixels in the border. Therefore, I need to determine at what distance the blocking window needs to be placed.

I hope the image below will illustrate my problem.

A 16.64 by 14.04 mm sCMOS sensor with 6.5 micron pixels is positioned at a distance of 43.35 mm from a spectrograph exit window. A square blocking window of unknown dimensions is placed between the sensor and exit window at an unknown distance from the sensor. All components are parallel.

The setup is:

  1. Motor operated scanning mirror
  2. Headwall VNIR Spectrograph
  3. PCO Edge VNIR 2560 x 2160 pixels sCMOS sensor

I am trying to "see" 2560 spatial pixels and 500th-1600th spectral pixels. So, I am thinking the horizontal window dimensions would remain the same (at 16.64 mm), but the vertical dimension is unknown. For example, vertical dimension would be 7.15 mm if the window is placed EXACTLY on the sensor. I am trying to establish how far to place the window. I can place the window anywhere between 5 and 10 mm from the sCMOS sensor.

Improve this question
$\endgroup$
2
  • 2
    $\begingroup$ Why block when you can reject the pixels in post-processing? $\endgroup$
    – Carl Witthoft
    Dec 29 '15 at 11:31
  • 1
    $\begingroup$ The answer to this question depends very much on what is happening behind the spectrograph window. The spectrograph does not 'create a focused image on the CMOS sensor'; it separates white (ish) light into its constituent wavelengths. The different wavelengths exit the window at different angles, but the angular spread is dependent on what happens inside of the spectrograph. Could you post a model number of the spectrograph? $\endgroup$
    – Chris Mueller
    Dec 29 '15 at 13:16
0
$\begingroup$

I would place the window directly on the sensor, if at all possible.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.