How it all started

I did first attempts of sensor monochrome modification in late 2020. My first „patient“ was old DSLR Pentax istD with 6 mpix APS-C CCD sensor. Modification went „quite easily“, Bayer CFA went off using Stanley blade (ha-ha-ha!) without scratching the surface, but I messed sensor because had zero experience and very tiny gold wires were broken during process.
Yeah, now I think it was crazy idea to use a steal blade for Bayer CFA removal, but it worked in that case. The most difficult thing was to remove sensor protective glass, which is glued to plastic/ceramic cage, but local heat application and patience did the job.
Inspired by this first attempt of quite easy debayering, I picked up Pentax K100D 6 mpix CCD camera (donated by another Pentaxian) and modified it in the same way as Pentax istD. This time sensor worked, but... had 4 pixel wide vertical line defect near the center of image area. I possibly damaged mechanically some circuitry of the sensor near imaging area. Not good, even manageable to remove the defect using Photoshop clone / content aware tools. But not convenient to use for sure.
To say the truth I was impressed by the images and how subjects pop-up even at f4 aperture (shot with Pentax FA 31/1,8 Ltd).


Picture 1.  Pentax K100D monochrome with Pentax FA 31/1.8 Ltd at f4.

6 mpix is OK, but I wanted more, because some internet sources claimed that monochrome modification gives up to 4 times better „resolution“ (details). Here came Pentax K200D with 10 mpix CCD. Protective glass was harder to remove, it crashed in pieces, but gold wires were still intact. What‘s next? Ah, sure, Stanley blade...:)  I thought all CCDs are physically same or very similar, 6 or 10 mpix, no practical difference. But result was absolutely different. K200D has very scratch sensitive surface under the CFA layer. All sensor surface was heavily scratched and I got still working sensor, just with „analogue film effect“ and couple of defected lines. This time Bayer CFA on the edge of imaging area was not even touched but defects occurred.


Picture 2. Pentax K200D 10 mpix CCD monochrome (debayered) sensor with defects.

More damaged sensors, more experience! Next K200D conversion in the Spring of 2021 was almost „ideal“ and I shot first portrait session (Picture 3). Sensor still had some very small artifacts of the surface, but pictures had very good details and I liked my new monochrome camera a lot. Now the „problem“ became original UV-IR cut filter I left in place, because it was a stack with anti-aliasing (AA) properties. As we know, AA filter blurs the image a little in vertical and horizontal directions and this effect is undesirable if maximum details are wished. Even with this AA filter effect monochrome camera gave very detailed images, but K200D was too slow for me (2008 year model).

Picture 3.  Monochrome picture shot with Pentax K200D debayered sensor and Pentax DFA* 50/1.4 lens. Copyright Mindaugas Grybauskas.

Then other Pentax DSLR models were tried to convert but unsuccessfully: K-x (maybe 3 cameras), K-30 (also maybe 3 cameras), K-01, before I came to Pentax K-5 in late 2021 – very good 16 mpix CMOS sensor model. Actually it was my infrared converted camera I did my self in 2017. I had two of them and decided to convert one on my risk.
Starting from K-x model the biggest issue in conversion process became removal of sensor‘s protective glass. I broke it intentionally many times as careful as I can, in most attempts sensors survived but this was very drastic and unsecure method. Sony sensors have protective glass which is glued with very strong adhesive material. Even 450^oC heat does not help – the glass simply cracks and removal of shards is even more difficult. Moreover extreme heat possibly can be harmful for the sensor circuitry.