Debayering a DSLR's Bayer matrix.

[Gina]

Gina, that is what I thought when I first saw them. If you take one of your busted sensors and rip the driver board off, you will see they are SMT components and not pins.

Ooops - sorry - you're right I've just blown my photo up and, yes, they are SMDs.

[alistairsam]

Alistair, if you only consider pixel size yes, but you also have factors like the microlenses, throughput of the bayer, readout noise etc. Also the 350D is 12bit as you say. The heat removal of the cover glass, you heat one spot, cool and then heat the edge of the spot, this makes the white spot longer and you repeat until you take the white spot all the way around the sensor. Obviously heating a sensor is not great and you really have to be quick with the heat and patient with the cooling. As for CCD vs DSLR, a 5um CCD pixel is almost pure active surface area, I think the DSLR sensor pixel size is not the actual size of the photosensitive part. CCDs are superior and here to stay for the foreseeable future, modding a DSLR is a compromise of performance and price. Modding CCDs can also be done, curious why it is so rare. Budget level cameras all really benefit from cooling upgrades.

Hi,

I meant the sensitivity of a debayered 350D vs the 1100D as the microlenses would be removed. but yes, the readout noise would be a major factor.

Does the generation of the processor make a difference at all? Digic 2, 3, 4 and so on. I'd expect speed to be a major improvement and even a lower noise floor.

I've got a spreadsheet here with some info I gathered on various parameters.

https://www.dropbox.com/s/ea2jurl3lglh2hv/dslr-canon-sensors.xlsx

the readout noise of the 350D looks quite low at 3.7e- and a full well of 43k.

would be great to see a like for like comparison with Luis or my 350D and Gina's debayered 1100D on the same target.

I'll give that heat method a go on the glass that's remaining on my 600D sensor. i'll get a pencil torch as well. i'm guessing heating for just 2 or 3 seconds would do?

I believe Guylian, a member here has debayered a Nikon which is a CCD. Might give that a go. pity that Nikon's don't have the same astro software support like the Canons.

Cheers

Alistair

[pixueto]

Thanks for all your help JTW. I've been thinking about all you've said and there is something I cannot get out of my head. Maybe you can shed some light on it. Do you think a DSLR camera with a CCD sensor such as the Nikon D80 could give better results with this mod? From your earlier posts I gather that CCD sensors will always be more efficient and the dichloroethanewill work better on those, right?

Thanks

[pixueto]

Hi,

I meant the sensitivity of a debayered 350D vs the 1100D as the microlenses would be removed. but yes, the readout noise would be a major factor.

Does the generation of the processor make a difference at all? Digic 2, 3, 4 and so on. I'd expect speed to be a major improvement and even a lower noise floor.

I've got a spreadsheet here with some info I gathered on various parameters.

https://www.dropbox....on-sensors.xlsx

the readout noise of the 350D looks quite low at 3.7e- and a full well of 43k.

would be great to see a like for like comparison with Luis or my 350D and Gina's debayered 1100D on the same target.

I'll give that heat method a go on the glass that's remaining on my 600D sensor. i'll get a pencil torch as well. i'm guessing heating for just 2 or 3 seconds would do?

I believe Guylian, a member here has debayered a Nikon which is a CCD. Might give that a go. pity that Nikon's don't have the same astro software support like the Canons.

Cheers

Alistair

Alistair, I've used for astrophotography all these cameras: canon 350D, 1000D, 450D, 1100D and 500D. I can tell you that the 1100D performance for astrophotography is far superior to the rest. I have been really impressed by that camera; the next best one would be the 450D.

[flolic]

Great development on this tread guys and gals

I promised results from my debayered cooled 450D, but that will have to wait a little. I didn't have the time to swap sensors in my cooled camera, so I went to dark sky location with my colour chip one... Sorry

Pixueto, I tried to remove glass from Nikon D50 CCD sensor that I own. That's the same sensor as in QHY8 camera (Sony ICX453AQ).

In short, I found that to be mission impossible...

Sensor package is made of plastic, and glass cover is glued with some bad ass glue that doesn't react to heat at all. I think that only way to safely remove glass is to cut it out with Dremel and diamond cutting wheel. I tried to pry it out with scalpel blade but glass shattered.

Another problem is that distance from glass to silicon is only around 1mm, and bonding wires are half the thickness of Canon sensors. Nightmare

On the positive side, once you manage to remove glass, scraping the CFA is piece of cake

[flolic]

Picture of poor sensor

[Gina]

I now plan to try to debayer two different DSLR sensors - an 1100D (Rebel T3) and the 350D that is currently winging it's way to me. If I can succeed with the 1100D the plan is to return to my Peltier cooled camera project with many improvements over version 3 as described in my 1100D cooling thread. I may start a new thread for version 4 I'll continue to post my results of debayering in this thread but I think cooling and enclosure are off topic.

[Gina]

Here's a link to my v3 1100d Peltier cooling thread :- http://stargazerslounge.com/topic/171415-1100d-set-point-cooling-with-peltier-tec-and-air-cooling-v3/

[RAC]

I must say its good to see JTW comment on this thread. Thanks.

My dermal method may not have been the best but It got people thinking and it did get a basic first result haha.

Gina you crack me up don't you have enough cameras

[Luis Campos]

I must say its good to see JTW comment on this thread. Thanks.

My dermal method may not have been the best but It got people thinking and it did get a basic first result haha.

Gina you crack me up don't you have enough cameras

I back that up

You were the pioneer Rac for sure

Cheers,

[pixueto]

Well, the dichloroethane arrived today but I'm not going to touch it until I get the protective goggles and mask. The fact that it is a carcinogen agent is freaking me out a bit! Also, after seeing Luis' brilliant results with his debayered camera, I, as Gina, have ordered a cheap working 350D. I might have a go and see if it's really that different compared to the 1000D (btw, I haven't found my old bricked 350D yet Luis). The plan is to apply the dichloroethane and see if I can remove the CFA without any scrapping leaving the silicone bed intact.

I noticed that nobody has removed the cover glass intact in the 350D. Is it that difficult guys? Do you think applying heat as JTW suggested will work?

Thanks

[JamesF]

Well, the dichloroethane arrived today but I'm not going to touch it until I get the protective goggles and mask. The fact that it is a carcinogen agent is freaking me out a bit!

No need to get freaked out. There are probably lots of things that you come into contact with that can be carcinogenic in the right circumstances and you may not even think twice about them. Use a bit of common sense and I'm sure you'll be fine.

James

[pixueto]

Great development on this tread guys and gals

I promised results from my debayered cooled 450D, but that will have to wait a little. I didn't have the time to swap sensors in my cooled camera, so I went to dark sky location with my colour chip one... Sorry

Pixueto, I tried to remove glass from Nikon D50 CCD sensor that I own. That's the same sensor as in QHY8 camera (Sony ICX453AQ).

In short, I found that to be mission impossible...

Sensor package is made of plastic, and glass cover is glued with some bad ass glue that doesn't react to heat at all. I think that only way to safely remove glass is to cut it out with Dremel and diamond cutting wheel. I tried to pry it out with scalpel blade but glass shattered.

Another problem is that distance from glass to silicon is only around 1mm, and bonding wires are half the thickness of Canon sensors. Nightmare

On the positive side, once you manage to remove glass, scraping the CFA is piece of cake

Thank you Filip. That's very interesting. The other thing that needs to be considered is the impossibility of getting 100% raw files from Nikon cameras. There is currently a working group doing reverse engineering trying to hack nikon cameras in a similar way to the Magic Lantern project. Maybe, obtaining real raw files from nikon cameras could be something they can achieve in the near future?

Edited August 7, 2013 by pixueto

[pixueto]

No need to get freaked out. There are probably lots of things that you come into contact with that can be carcinogenic in the right circumstances and you may not even think twice about them. Use a bit of common sense and I'm sure you'll be fine.

James

Having a quick look at wikipedia, it seems that dioxolane or toluene are suggested as possible substitutes as solvents. Maybe someone here wants to have a go with those.

[flolic]

I don't know and honestly, I'm not interested in (old) Nikon camera hacking. It's just happened that I had one lying around, with broken sensor.

So I gave it a try

Possibility of hacking those sensors could be interesting for QHY8 owners.

[flolic]

Having a quick look at wikipedia, it seems that dioxolane or toluene are suggested as possible substitutes as solvents. Maybe someone here wants to have a go with those.

I tried ethyl acetate long time ago on Sony webcam sensors, that stuff does not dissolve CFA, but make it very soft and easy to scrape off with just a gentle rubbing. But Canon sensors are completely different thing...

I would love to try different solvents, but they are hard to find in my country.

There is one more possible chemical that could remove CFA, and that is concentrated Nitric acid under elevated temperature. I have a bottle with concentration of 75% and I will try that on broken sensors. But that stuff is pure evil

I hope it will not eat everything

[Luis Campos]

Well, the dichloroethane arrived today but I'm not going to touch it until I get the protective goggles and mask. The fact that it is a carcinogen agent is freaking me out a bit! Also, after seeing Luis' brilliant results with his debayered camera, I, as Gina, have ordered a cheap working 350D. I might have a go and see if it's really that different compared to the 1000D (btw, I haven't found my old bricked 350D yet Luis). The plan is to apply the dichloroethane and see if I can remove the CFA without any scrapping leaving the silicone bed intact.

I noticed that nobody has removed the cover glass intact in the 350D. Is it that difficult guys? Do you think applying heat as JTW suggested will work?

Thanks

Hi,

Man, those quemicals make me nervous...please be carefull buddy.

I just got another 350D from a friend who wants to do the mono conversion, he is ok in case I ruin the sensor, hope to get it working like my 350D

Will work on that next weekend

Also I will disassemble mine again, to make the nitrogen purge on the sensor, hope this will cure my foggy sensor when using the cooling (it is assembled now).

Cheers guys,

[pixueto]

I tried ethyl acetate long time ago on Sony webcam sensors, that stuff does not dissolve CFA, but make it very soft and easy to scrape off with just a gentle rubbing. But Canon sensors are completely different thing...

I would love to try different solvents, but they are hard to find in my country.

There is one more possible chemical that could remove CFA, and that is concentrated Nitric acid under elevated temperature. I have a bottle with concentration of 75% and I will try that on broken sensors. But that stuff is pure evil

I hope it will not eat everything

How did you apply the ethyl acetate on the sensor? I'm planning to use swabs with the solvent but I also thought of a small paintbrush?

[pixueto]

Hi,

Man, those quemicals make me nervous...please be carefull buddy.

I just got another 350D from a friend who wants to do the mono conversion, he is ok in case I ruin the sensor, hope to get it working like my 350D

Will work on that next weekend

Also I will disassemble mine again, to make the nitrogen purge on the sensor, hope this will cure my foggy sensor when using the cooling (it is assembled now).

Cheers guys,

Luis, looking forward to seeing some results with colour images and H-alpha added to it with your mono 350D -that's the way I would like to use it; with a clip H-alpha filter. BTW, how did you manage to merge those two images from different cameras/sensors?

If you are interested, I can give you an anti-aliasing filter from a canon 1100D (this one, I know where I placed it). I think it's big enough to cover your sensor if you want some protection. The trouble is, the anti-aliasing will blur a bit the images but it'll also provide some protection for star bloating -is this a concern in monochrome sensors too? - I wonder.

Using that solvent makes me nervous too! I'll do it in the garden No ideal but safety is my first concern. we'll see how good it is with the CFA remnants on the 1000D sensor first.

Edited August 7, 2013 by pixueto

[flolic]

How did you apply the ethyl acetate on the sensor?

I desoldered sensor and put it in a jar with a lid. And just fill it until sensor is submerged. Ethyl acetate evaporate extremely fast, you have to work outside or vapors will make you dizzy.

Ethyl acetate attacks many kind of plastics very fast, you don't want to spill it onto PCB or camera housing.

[Luis Campos]

Luis, looking forward to seeing some results with colour images and H-alpha added to it with your mono 350D -that's the way I would like to use it; with a clip H-alpha filter. BTW, how did you manage to merge those two images from different cameras/sensors?

If you are interested, I can give you an anti-aliasing filter from a canon 1100D (this one, I know where I placed it). I think it's big enough to cover your sensor if you want some protection. The trouble is, the anti-aliasing will blur a bit the images but it'll also provide some protection for star bloating -is this a concern in monochrome sensors too? - I wonder.

Using that solvent makes me nervous too! I'll do it in the garden No ideal but safety is my first concern. we'll see how good it is with the CFA remnants on the 1000D sensor first.

Trust me...I'm dying to get a 2" Baader 7 nm filter or maybe just the glass, I think they sell the filter with no frame and is a bit cheaper )

I merged the two images by simply resing the color image from the 450D it's bigger because of pixel density (smaller pixels) but working with two layer in "difference mode" under Photoshop you can see what you are doing, it takes a wile by trail and error but eventully they will align perfectly

The only thing is you can not move the two cameras orientation when exchanging cameras or you will be in trouble, I just detached the 450D from the T2 adaptor while keeping it on the focuser and reasembled the mono 350D on the exact same orientation, worked good!

Many thanks for the filter but my camera has the Baader on it used with the stock color sensor, it's a great filter

Yep, monochrome cameras will aslo need an IR filter to prevent from star bloat, especially if there's any glass element on the optical train (refractors, coma correctors, field flatteners, etc...)

Cheers,

Edited August 7, 2013 by Luis Campos

[Gina]

I think I would be worried about high-power solvents and other chemicals too. I tried IPA but that didn't touch it. Cleans off heat conducting grease quite well though.

I have ordered a tiny hot air gun from Amazon - designed for craft use, but may be sufficient for heating the edge of the cover glass. Don't know it it will be powerful enough but have to see I could make use of it for other things though. http://www.amazon.co.uk/Papermania-Pink-Basics-Heat-Tool/dp/B002XECJS2/ref=sr_1_21?ie=UTF8&qid=1375882379&sr=8-21&keywords=heat+gun

My full size DeWalt gun was way OTT

[pixueto]

I think I would be worried about high-power solvents and other chemicals too. I tried IPA but that didn't touch it. Cleans off heat conducting grease quite well though.

I have ordered a tiny hot air gun from Amazon - designed for craft use, but may be sufficient for heating the edge of the cover glass. Don't know it it will be powerful enough but have to see I could make use of it for other things though. http://www.amazon.co...ywords=heat gun

My full size DeWalt gun was way OTT

Let us know how you get on with removing that glass with your girlly hot air gun (I looked at it ). I'll be trying the 350D with a butane pen torch.

I also feel apprehensive about using dichloroethanebut that's what the people at JTW suggested it works so I would like to give it a try. I find surprising you can buy this staff online; particularly when it's been banned from use in USA due to its toxicity and carcinogenity.

Edited August 7, 2013 by pixueto

[Gina]

I'll certainly let everyone know how it goes I have a butane pen torch somewhere but I can't find it I really need to do a BIG clear up

[earth titan]

Out of all the solvents mentioned, the most effective is Dichloromethane (DCM or Methylene Chloride). Not too much of a hassle to work with as long as you use small amounts in a well ventilated area and wear nitrile gloves. Used to be the base solvent in Nitromors paint stripper but has recently been banned for domestic / trade use.