Microscopy of Nikon sensor

Ever wonder what is it like to see the bayer pattern on an image sensor. Last year I got a sensor from one camera repair guy and it's a LBCAST from a broken D2H. I assume it's useless but might be the sensor can still work if you solder it to a new camera imaging PCB in D2H.

Here's the top left corner view in 1x

And at 11.5x on CCD imager, bayer pattern can be resolved:

Note the brighter 17 columns and 10 rows are fabricated with metal layer between color filter and photodiode, this is known as the "Optical Black" regions on the peripheral of imaging sensor that allows you to calculate the black signal. And because of the metal layer that reflects light, it appears to be much brighter. This light shielded area is included in the "effective pixel" count. The darker region is the actual "Active pixel". For total number of pixel, the dummy pixels outside the light shielded area is included.

Let's take a look at the bottom right corner at higher magnification. (23x objective)

In B/W image, we can see the reflection from the microlens array overlay on top of bayer color filter. Because light is not shining from above but from aside, the reflection is off center on the dome shaped microlens

In color image, we can see the dummy pixels did not have microlens and you can see the organic color filter directly.

The vibration from the CCD cooling fan is really a issue at higher magnification. To clearly resolve the microlens may need 100x oil immersion objective with depackaging of sensor cover glass or even send to Scanning Electron Microscope.

Thanks for viewing!