Debayering a DSLR's Bayer matrix.

[Gina]

Now then... Connected up to camera and tried live view - Err 70 Swapped back to camera's own sensor and amera works fine.

So that would seem to be that - I've finally succeded in destroying the test sensor - guess one of the glass slivers must have caught a fine gold wire

And Puff The Magic Dragon sadly slides back in her cave...

Disheartened of Devon

Edited October 31, 2012 by Gina

[Photosbykev]

Now then... Connected up to camera and tried live view - Err 70 Swapped back to camera's own sensor and amera works fine.

So that would seem to be that - I've finally succeded in destroying the test sensor - guess one of the glass slivers must have caught a fine gold wire

And Puff The Magic Dragon sadly slides back in her cave...

Disheartened of Devon

If you don't try you'll never know

[Gina]

Does it still work?

It would seem NOT

I have no idea what that glue must be, but it clearly forms a very strong bond between the glass and the backing. it's almost like it's welded rather than glued.

James

I think you're right.

[Gina]

If you don't try you'll never know

Quite agree. I could try a couple of solvents and paint stripper on the sensor and see what happens just by eyeballing it.

[russellhq]

I can't see very well from the photos, but is there any way you could prise off the black plastic frame holding the glass from the rest of the assembly?

[studdedsole]

Just tuned into this thread and it gave me a real good chuckle at some of the antics, you guys have great big cojones to contemplate trying this.

I don't know about extreme mods but I think Gina needs some extreme counselling, jeez man you is one looney character

[Kbramley]

Been watching this thread with interest.

Gutted for Gina however I would reccomend maybe trying it once more in case it was a dodgy connection maybe? Err 70 is reported as being "shooting not possible due to an error, turn camera off/on" and "Malfunctions related to the image have been detected".

I have been caught out many times with poorly seated connectors, worth a shot I guess.

Keith

[Gina]

I can't see very well from the photos, but is there any way you could prise off the black plastic frame holding the glass from the rest of the assembly?

No. The black frame is what holds all connections and is NOT plastic I think it's some kind of glass - it's very hard.

[Gina]

Just tuned into this thread and it gave me a real good chuckle at some of the antics, you guys have great big cojones to contemplate trying this.

I don't know about extreme mods but I think Gina needs some extreme counselling, jeez man you is one looney character

:D Oh I is looney alright Mad as the proverbial hatter A true mad sciemtist But it's mad scientists who make breakthroughs

[Gina]

Been watching this thread with interest.

Gutted for Gina however I would reccomend maybe trying it once more in case it was a dodgy connection maybe? Err 70 is reported as being "shooting not possible due to an error, turn camera off/on" and "Malfunctions related to the image have been detected".

I have been caught out many times with poorly seated connectors, worth a shot I guess.

Keith

I'm going to try the same test sensor again on the camera - I've had Err 70 from bad connections before.

[Gina]

I've got a high power magnifier and been examining all the fine gold wires and their connections and they all seem fine. However, where I ground down the glass cover plate and then prised it off there's damage to the sensor itself - it looks like broken glass.

I've done a couple of tests with IPA and paint stripper - absolutely no effect. It actually looks like another glass layer

[Photosbykev]

I've been doing a bit of research on the application of the CFA layer to sensors and I think it is applied using a Photolithography technique http://en.wikipedia.org/wiki/Photolithography and that the Microlens over each pixel is above the CFA layer http://www.maxmax.co..._conversion.htm

[Gina]

I've tried using a small amount of buffing paste on the sensor using a cotton bud. Within a few tens of seconds I had removed the green layer leaving a smooth gold layer underneath. Here's a photo of the result.

[Gina]

I've been doing a bit of research on the application of the CFA layer to sensors and I think it is applied using a Photolithography technique http://en.wikipedia....hotolithography and that the Microlens over each pixel is above the CFA layer http://www.maxmax.co..._conversion.htm

Yes, I know what photolithography is and it is very likely that this technique was used. As for the microlenses, as long as the result is improved sensitivity and resolution I'm not that bothered if they go.

[allcart]

I'm sure we are all on your side Gina, and praying that the sensor still works. I haven't had such a laugh in ages. When I think back to the Canon forums and how much folks stress about being ever so careful with their sensors when cleaning them, I just LOL.

[Gina]

I've tried a 20mm buffing wheel and paste on the sensor covering most of the area. This was very quick. However, although all the green has been removed the next layer is softish and I've taken what appears to be a varying amount off - it's giving diffraction patterns. Here's a photo.

I'm thinking that if this stage went well I might have a go with the other spare sensor. However, I have to say that the problem with the cover glass makes this tend towards being a bit too extreme and hazardous unless I find the next trial run goes a lot better. It's absolutely essential to be able to remove the cover glass safely, let alone further risking damage when buffing.

[Gina]

I'm sure we are all on your side Gina, and praying that the sensor still works. I haven't had such a laugh in ages. When I think back to the Canon forums and how much folks stress about being ever so careful with their sensors when cleaning them, I just LOL.

Oh I know everyone is with me on this And I agree with you about the great care people take. The sensor is practically indestructible

[theodore]

Just tried collodion on an old non-working webcam sensor and it took some of the bayer layer off, so maybe repeated applications would work...

But I found the easiest way was just to scrape it off with an old fashioned flat razor blade. Seemed to work very well and didn't scratch the surface- its to small to get a pic of though.

Theo

[JamesF]

I've been trying to work out what is achieved by doing this to a webcam because there's no way (generally) to get at the raw data. Would you just end up with a monochrome camera with increased sensitivity that gave you what was effectively a 2x2 binned image?

I believe the SPC firmware can be altered to give the raw image data, but wouldn't having to transfer all that data over USB 1.1 cripple the frame rate?

James

[Cath]

The poor thing has been polished, ground, peeled, chipped, hammered, filed, and just about everything else by the looks of it. I'm speechless lol.

[Photosbykev]

The poor thing has been polished, ground, peeled, chipped, hammered, filed, and just about everything else by the looks of it. I'm speechless lol.

It gave it's life for the greater good. Well it didn't give it as such, more like murdered lol

[Gina]

R.I.P.

[Gina]

It was an accident - honest gov

[ArmyAirForce]

It was an accident - honest gov

An accident, I don't think so. It was "Sensorcide!"

[JamesF]

It was an accident - honest gov

"It just came off in my hand"

James