Debayering a DSLR's Bayer matrix.

[Gina]

Here is the 10m Eagle Nebula sub taken with the MN190 and mono 1100D processed in Ps subtracting the single flat frame and applying various tweaks to improve the image.  I din't get the flat subtraction exactly right but here's what I've managed.  Firstly cropped to remove CFA and cropped more to show the nebula.  At least the Eagle is recognisable.  Not too bad for a single 10m sub with virtually no cooling.

 

Edited September 27, 2013 by Gina

[Gina]

I really must sort out what I've got - camera models, which work and which don't - sensors ditto.  I have some cameras in bits and some untouched.  I'm stopping bidding for any more cameras until an 1100D non-working one comes up at a sensible starting price.  I want another good 1100D sensor to work on.  I have a lot more experience now and would hope to convert this to mono with a better result than my current working one.  My aim is to get at least one good mono 1100D with set point cooling.  Now I have my heart set on an EQ8 and want to start saving for that.  With the EQ8 I could have MN190 plus 2 ED80s and maybe widefield rig as well on the one mount.

I have just picked out a 350D and tested it - no life in it with fully charged battery.  So dismantled it and took out the sensor.  Laid bare the cover glass and attacked that - came off quite easily but in several pieces even with very careful knife work.  These glasses are pretty fragile.  Then attacked the CFA and have about three quarters of that removed.  Now stopped for tea.

[Gina]

Well... here's another 350D sensor taken as far as I dare risk   Now to find a working 350D I can try it in hoping it works.

[Gina]

Yep - it works   I'll do a flat once I've charged up the battery.

Edited September 28, 2013 by Gina

[Kbramley]

Looking good, I am assuming the "Homemade Orange Marmalade" isnt used in the de-bayering process anywhere

Keith

[Gina]

Thank you   The marmalade provides essential nutrition for the operative

[Gina]

Here's a flat taken with the latest 350D sensor.  Well stretched in Ps as usual.

Edited September 29, 2013 by Gina

[Gina]

Started the stock-taking   Seems I have 3 350Ds, one of which works.  3 350D sensors all of which work.  One has just the cover glass removed another with a fair margin of CFA left and the third with just a small CFA margin.  I guess I'll have to do some fault finding or I could have more working sensors than cameras

[Gina]

A 450D non-working camera arrived this morning, one of my ebay auction wins from a little while back.  It has a completely wrecked shutter.  I've now stripped it down and taken out the sensor and put it in my (mostly) working 450D and the sensor is fine (as expected).  So I now have another 450D sensor to wreck... um... I mean debayer.  Dud 450Ds seem quite common in the ebay auctions unlike 1100Ds which come up quite rarely.

The 450D is much easier to work on than earlier models and looks quite a good candidate for cooling so I might just have a go at a debayered and cooled 450D and see how that goes.  Assuming, of course, that I don't wreck this sensor   At least the cover glass on these comes off reasonably easily.  (Well, it did last time.)

[Gina]

450D cover glass off all in one piece - not so much as a chip   Used craft knife to work gently round the glue edge gradually increasing the white area until it came off.  Fine gold wires untouched.

[Andrew*]

C'mon Gina. I've got a 450D cowering in a corner here. Please tell me the CFA is virtually impossible to remove and it might be safe....

[Gina]

Well, I got a fair bit of the CFA off the first one but went too far and broke a couple of those dreaded fine gold wires.  My intention is that I won't do the same with this one   It's working fine withe the cover glass and filters removed but I haven't attacked the CFA yet.

[Gina]

Started CFA removal and all's well   Took quite a while scrubbing away wuth the car paint polish to get through the tough top layer but proceeded well after that.

Edited October 2, 2013 by Gina

[Gina]

Here's a flat

[Gina]

Flat taken with tissue paper over lens and a cloudy sky.  Converted from CR2 to TIFF with DCRaw and stretched in Ps then resized and saved as JPEG.

[Gina]

More 450D CFA removed - I don't know if I dare go any further, I'm getting very close to the danger zone.  There is some residue on the sensor I can't seem to get rid of even with IPA and cotton buds.  It just seems to dry streaky.

Edited October 3, 2013 by Gina

[Gina]

With the relative ease of cover glass removal and lower prices plus more availability of cheap second hand 450Ds, I've been looking into adding set point Peltier TEC cooling to the 450D and adding a filter wheel. 

The cooling looks quite easy - the gap between sensor and PCB is about 1.4 to 1.5mm and unlike the 1100D, no components on the back of the PCB to interfere with a cold finger.  Hence 18swg 1.2mm thick copper strip will fit easily in the gap.  Just like my 1100D cooling, the cold finger can be bent round across the back of the screen of the imaging assembly and Peltier TEC plus CPU cooler can go on the back of the camera, replacing the LCD screen.

What is not so clear is the possibility of reducing the optical back focus to provide room for a filter wheel but I will be investigating this.

[Gina]

Getting ready to test mono DSLRs for AP.

Firstly APT v2.30 didn't work so I've downloaded the lastest v2.33_1 and installed it on my netbook and it's working fine with my totally unmodified 1100D that I use for general photography.  Then I tried my mono 1100D which works fine with EOS Utility.  APT took one exposure (I think) and then no more saying "BUSY".  So this camera will not be having the full AP test as any camera for AP MUST work with APT to take a series of subs automatically/unattended.

Next I connected the 450D that I have been working on and currently has the best debayered sensor installed.  That is working fine with APT and taking images repeatedly.  I therefore plan to make up a cold finger for it and add some cooling.  I'm hoping I might be able to do that today and put the camera on top of my MN190 on the camera mounting point.  Forecast is good for tonight (ATM) and I'm hoping to do some imaging.  So while the MN190 plus Atik 460EX is taking subs of the Heart of the Heart Nebula the 450D can be taking Ha subs of the larger area around the Heart & Soul Nebulae.

Fingers crossed

[Gina]

Just taken some flats and here's a sample sub - fed through DCRaw and stretched in Ps plus the Curves Histogram  showing the stretch

 

[Luis Campos]

Just taken some flats and here's a sample sub - fed through DCRaw and stretched in Ps plus the Curves Histogram  showing the stretch

450D_Flat_2013-10-05_12-34-59_0023_ISO1600_1-4000s__31C.jpg  Flat Histogram.PNG

This one looks really sweet Gina

No doubt, your best one to date, practice leads to perfection

Can't wait to see the result!

BTW, I'm also going to apply a cooling system on a debayered 450D I'm making for a friend, I think I found a way to combat the sensor cover glass dewing problem Will let you know once I have proper tests

Cheers,

Luís

[Gina]

Thank you Luis   I'll be glad to see how you get on

I think I'll just use one of my ready-made cold fingers for the moment and make a thicker one later on.  Also, I'll just use moderate manual cooling for now and stay above the dew point.  I want to get everything set up and ready to go by the time it gets dark.  Still looking fine for tonight

Later on I'll design a dry chamber for the 450D and build a proper cooling unit for it.  I may go for 450Ds rather than 1100Ds.  Not quite as good but could well be viable - 1600 is a viable ISO for imaging I think though I generally went for 3200 on the 1100D before but these CFA free cameras will be that much more sensitive.  I might go for a triple imaging rig with 3 450Ds with a DIY filter drawer rather than a filter wheel.  I already have NB filters in 36mm unmounted size (Baader) and might get a set of RGB filters for galaxies.

I already have 3 sensors in various stages of mono-fying but only one working body ATM though I think I may be able to sort out a second one.  So I may bid on another 450D

Edited October 5, 2013 by Gina

[Gina]

I was hoping to test the mono 450D with the 7nm Ha filter but can't use that with the Canon EF-S lenses I have - the lens projects too far into the camera.  Can't use clip filters either.  I was going to use a Pentax thread M42 lens but can't find an adapter.  I think I may have used all I had in building my triple imaging widefield rig, and I don't want to dismantle that.  So I shall be using the kit lens without any filter tonight.  If the clouds go away

[Gina]

Found an adapter so will be using a Pentax/Takumar lens and can add filter.

[Gina]

Didn't have much joy testing the 450D last night - the humidity was dreadful and after a short time the lens dewed up.  I'll have another go next time we get some clear night sky and arrange to use a lens with dew heater.  The camera sensor also suffered from condensation, even with the cooling off   In fact I reversed the power to the Peltier TEC to warm the sensor   That cured the sensor misting but left some surface marks.

Meanwhile, I'm looking into proper cooling with a dry chamber to stop the sensor misting up.  I'll start a dedicated thread for this when I have something to report.

[themos]

"Reverse the polarity!"