Debayering a DSLR's Bayer matrix.

[Gina]

"Reverse the polarity!"

Peltier TECs can be used to heat as well as cool and are resersible devices.  With "normal" current polarity one side gets cold while the other side gets hot.  If you reverse the current flow the side which normally gets cold now gets hot and vice versa.

[octachoron]

So, I finally took the plunge and attempted debayering my 600d. I was able to remove the cover glass in one piece. I used a heat gun I have access to set at over 800 degrees F! It came off easily, with no damage to the sensor or gold wires whatsoever. I reinstalled it into the camera and it worked flawlessly.

I've had considerably less luck with the microlenses and bayer mask. I've barely scratched the surface, and I think I've damaged a couple rows of pixels. Also, about 3/4 of the screen shows darker than the remaining portion. Any advice?

I'll upload a picture when I get more time.

[JohnC64]

looks like JTW are now offering debayered cameras on a commercial basis

http://www.jtwastronomy.com/products/monochrome.html

and also a primer on how to debayer

http://www.jtwastronomy.com/tutorials/debayer.html

John

Edited October 15, 2013 by JohnC64

[Sternfreund]

Hey,

Thanks for this thread!

It inspired me to mod my 450D.

First i removed the filter for greater red-sensitivity and then i started to debayer.

It worked fine and all gold wires are intact.

BUT:

See these shots and tell me what happend here?!

Flat:

Dark:

Thank you in advance!

Arno

[Gina]

I think you've gone a bit too deep with the debayering in one or two places.

[Sternfreund]

Hi Gina,

Do u think its still usable?

[Gina]

I rather doubt it but the only way to be sure is to try it - I think you'll lose the right hand side and a band across just below the middle as the dark shows.

[Sternfreund]

hmm...

maybe ill try it tonight. if the upper spots works i'm still satisfied. a little bit at least...

when i started disassembling the cam i had huge doubts that it would work. 

i just wonder why there is so much noise in the dark.. is it also due to the fact that i went too deep on the right side?

well.. i got the cam for about 70 euro.. 

[Luis Campos]

hmm...

maybe ill try it tonight. if the upper spots works i'm still satisfied. a little bit at least...

when i started disassembling the cam i had huge doubts that it would work. 

i just wonder why there is so much noise in the dark.. is it also due to the fact that i went too deep on the right side?

well.. i got the cam for about 70 euro.. 

Hi,

Same thing happened to me on a 350D sensor, just a small scratch was enough to cause two dead pixels collumns and ended up with very high noise on long exposures, the sensor is useless for astrophoto but still ok for regular BW day photography, except for same ligh horizontal banding...

Better luck next time

Edited October 31, 2013 by Luis Campos

[Gina]

This is not an easy job - I have quite a large graveyard of sensors I have managed to destroy

[Tf2k]

Yes, debayering is an expensive hobby. 8 dead sensors on my desk...  

But my debayered D40 is working since my last post here.

Doing research has many drawbacks until a breaktrough can be achieved.

[Tf2k]

Just to add another method to convert the  raw files without debayering into tiff files into this thread, the libraw library has a sample program which is called unprocessed_raw.exe.

it is contained in the binary distributions from:

http://www.libraw.org/download#stable

http://www.libraw.org/docs/Samples-LibRaw-eng.html

e.g . unprocessed_raw -A -g -T Filename

[nytecam]

Yes, debayering is an expensive hobby. 8 dead sensors on my desk...  

But my debayered D40 is working since my last post here.

 Doing research has many drawbacks until a breaktrough can be achieved.

A very sobbering thought that has answered for me :-)

[Gina]

Sometime I think I ought to add up what I've spent on buying used and faulty DSLRs and other bits for the debayering project   I may be able to recoup some of the cost by selling spare parts on ebay and/or I might try on the Classifieds here too.

[Gina]

Later... Just added up DSLR spending on ebay and it comes to a touch under £800 for 11 cameras.  I'm sure there are a couple more but I've been through all this years purchases.

[Gina]

Have just spent a day reading every post on this epic thread.

Congrats on all your endeavors. Just one thing occured to me. I see in order to avoid breaking the gold wires some have been using the filter assembly as a shield. This seems to be partly effective as there is still a gap under it where the cover glass used to be.

Could you use the filter assembly as a guide to cut the glass using a diamond tipped cutter. Then remove the center section which should leave a cover for the gold wires and the blue area that is subject to damage. Another advantage would be a clean edge to the bayer removal and less loss of pixels.

If you could achieve that then there would be far less possibility of damage and no heat or major mechanical shock to the sensor.

Having never disassembled a Canon I am not sure if this is possible, but just an idea for consideration.

Terry

I tried cutting the cover glass with a glass cutter - it wasn't successful - the glass didn't cut cleanly and a small particle fell on the gold wires, breaking one of them.  That was the end of that sensor   The frame used as a stop did work though.

[woodbob123]

Been practicing on cheap 99p + post cameras off e-bay no sucess so far broken wires each  time so next week when some more arrive I plan to try and fix wires in position using some casting resin, Tilt sensor and fill one side at a time hope it works. Using such small sensors as I hope first to make some wildlife camera traps, then progress up to an DSLR

[Davey-T]

Been practicing on cheap 99p + post cameras off e-bay no sucess so far broken wires each  time so next week when some more arrive I plan to try and fix wires in position using some casting resin, Tilt sensor and fill one side at a time hope it works. Using such small sensors as I hope first to make some wildlife camera traps, then progress up to an DSLR

Think this has been tried but even the shrinkage as the resin sets is enough to break the wires.

We still need to find a solvent to dissolve the matrix without scraping.

Dave

[libraryman]

Been following this thread and trying my own methods on duff sensors, this is my last attempt, i have found a material that i believe will remove the matrix without scratching the layer below!

there is a small scratch in the lower level but this was caused by the glass removal.

I used a small piece of Melamine with a flat end.

If someone else tries it please post results.

Ray 

[Gina]

DSCF5131.jpg

Been following this thread and trying my own methods on duff sensors, this is my last attempt, i have found a material that i believe will remove the matrix without scratching the layer below!

there is a small scratch in the lower level but this was caused by the glass removal.

I used a small piece of Melamine with a flat end.

If someone else tries it please post results.

Ray 

Looks very good   Which camera model does this sensor come from?

[libraryman]

Not sure Gina, it was given to me by a friend who works with cameras, it did not matter too much to me since it was a test, the glass was a pain...need to work on that but the bayer must be very similar.

i will try it next without the glue because the tool i used really works very well.

and no matter how (reasonably) hard i press i cannot damage the gold coloured layer.

Are you going to try again?

Ray

[Gina]

I have a couple of 450D sensors I would like to debayer, one is half done but can't remember now if I started the other one.  The idea is to have 3 matching cameras for NB widefield imaging.  I was hoping to do 1100Ds as these are slightly better but the cover glass has proved virtuall impossible to remove and needs heat which can damage the sensor, so I gave up on those.  The 450D glass comes off quite easily and the CFA comes off alright with very fine abrasive.

[libraryman]

Did you still have to use heat on the 450D glass?  it is this aspect that worries me the most...i think that i can successfully remove the bayer without damaging either the sensor or the gold wires but that glass!!!!

Ray 

[alistairsam]

I have a couple of 450D sensors I would like to debayer, one is half done but can't remember now if I started the other one. The idea is to have 3 matching cameras for NB widefield imaging. I was hoping to do 1100Ds as these are slightly better but the cover glass has proved virtuall impossible to remove and needs heat which can damage the sensor, so I gave up on those. The 450D glass comes off quite easily and the CFA comes off alright with very fine abrasive.

hi Gina

instead of three cameras and changing cameras, why not use a filter wheel?

so much easier and you only need one camera, no issues with the frame getting rotated when you change.

I have the sx usb wheel and 36mm unmounted filters and they're great.

cheers.

Alistair

[JamesF]

hi Gina

instead of three cameras and changing cameras, why not use a filter wheel?

so much easier and you only need one camera, no issues with the frame getting rotated when you change.

I have the sx usb wheel and 36mm unmounted filters and they're great.

cheers.

Alistair

You can't use the filters in a filter wheel at the same time

James