Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

Debayering a DSLR's Bayer matrix.


RAC

Recommended Posts

Jnp:

1= yes

2= yes

Im not sure the glass and gel would work. There are no scatches to fill its perfectly smooth but at uneven depths.

I ended up getting an atik 383l+ mono

Link to comment
Share on other sites

I just had a thought to help out the sensitivity of the stock colour canon 1000d sensor. And thats just to remove the glass/plastic cover on the sensor. From looking at ccd tests with the class removed you get about 5% more light, its easy so i might do it to my one if it ever gets a chance to get used again.

Link to comment
Share on other sites

I just had a thought to help out the sensitivity of the stock colour canon 1000d sensor. And thats just to remove the glass/plastic cover on the sensor. From looking at ccd tests with the class removed you get about 5% more light, its easy so i might do it to my one if it ever gets a chance to get used again.

The danger of clouds! :grin: :grin: I think that's the only reason I modded my camera, I had nothing else to do!

Link to comment
Share on other sites

I think this sensor has had it :sad: I'm having no luck reattaching the gold wires.

Apart from that the actual filter removal is going well. I think the potential advantages are worth having a go with another sensor, so I'll open a new one up tomorrow (with extra care this time).

More 350Ds than sense...

Link to comment
Share on other sites

Good point :)

We actually have clear skies here now. Overcast earlier but clear at 10:30pm.

I bought a mono 314L+ rather than destroy any more DSLR sensors - it was cheaper! :D 1100D still makes a reasonable large sensor OSC though. I'm running one ATM with 200mm f4 SMC Takumar telephoto lens covering the Flame via horsey to M42. 30s at ISO 3200. Astronomik CLD CCD filter to clean up the LP. I'm also running my 314L+ on my ED80 running 5m subs in 5nm Ha and 3nm OIII covering the Flame and Horsey. I may increase the exposure to 10m or more.

Edited by Gina
Link to comment
Share on other sites

I just had a thought to help out the sensitivity of the stock colour canon 1000d sensor. And thats just to remove the glass/plastic cover on the sensor. From looking at ccd tests with the class removed you get about 5% more light, its easy so i might do it to my one if it ever gets a chance to get used again.

I was pondering whether to put the glass cover back on. Any bit of glass is going to lose you around 5%, and having no glass in the image train at all would be good for Venus images as you'd get more UV reaching the sensor. I suppose the shutter will be closed when the camera is off the telescope so the sensor will be fairly protected without the glass.

Not a DSLR I know, but here's some interim test results with an SPC900NC.

In between the original section and where the filter's been removed you can see a dark patch where just the top microlens layer has gone. You can see it decreases the sensitivity a fair bit, but this is then made up when removing the filter. The net result isn't a huge change in sensitivity in optical, but it'll be noticably improved for IR and UV imaging. However the improvement in resolution is pretty obvious even with my crude test-card setup.

post-970-0-02187800-1357910760_thumb.jpg

post-970-0-63000800-1357910763_thumb.jpg

post-970-0-83806500-1357910770_thumb.jpg

  • Like 2
Link to comment
Share on other sites

Why only 30sec for the 1100d gina? I would lower the iso and go 5 or more min

I did. I went with ISO 1600 and 1 minute. In tests I've found no benefit in going below ISO 1600. I started off with 30s without guiding to make sure of getting something out of the night's imaging. I went over to my standard ISO 1600 when I'd got guiding going.
Link to comment
Share on other sites

I've started on 350D sensor #2, after practicing with the first one. I've swapped the metal dental pick for a scraping tool carved from the plastic end of a paintbrush. The plastic won't scratch through the layer beneath the filter like the metal can do, but does an adequate job of scraping off the filter.

The first pic is a flat(ish) field. The darker area is the top layer removed, you can see again that sensitivity takes quite a hit with the loss of the microlenses. The blueish area is with the filter layer removed.

The second is a crop of some text. The original with bayer filter on the right, filter removed on the left, and just microlenses removed on the strip between them.

The raw image on the top, and then demosaiced on the bottom.

post-970-0-45592700-1358001448_thumb.jpg

post-970-0-04252100-1358002668_thumb.jpg

  • Like 1
Link to comment
Share on other sites

It's just the white balance. You'll see in the top right, where the bayer filter is still there that one in four pixels is much darker, this is the blue one (darker due to sensitivity and the spectrum of light in the room).

When it comes to converting the raw image to colour it will boost the blue pixels to balance the colour. Where the bayer filter has been removed, all pixels will be equal in brightness, but during the colour conversion it still thinks one of them has been filtered blue, and boosts the blue accordingly.

Link to comment
Share on other sites

This is the result of this weekend's work. I'll need to make up a smaller tool to scrape near the edges, already had one close call with the gold wires (ok, I don't need to go right to the edges.. but it's nice to do things properly).

All looks fine so far, no issues with the flat fields as with the polishing method. Once I'm finished it'll need a good clean up to remove all of the filter debris though.

post-970-0-41691200-1358107620_thumb.jpg

  • Like 5
Link to comment
Share on other sites

Here's a flat, f/8, heavily stretched to bring out the lovely details.

This is only after brushing off the loose debris from the filter removal. Once I've done the edges I'll clean it up properly with IPA and compressed air and post another.

post-970-0-98877600-1358635922_thumb.jpg

  • Like 1
Link to comment
Share on other sites

  • 2 weeks later...

Hello all, I've been watching this thread with great excitement - and amusement. 12dstring your experiments are very promissing. Can you post a picture of your paintbrush handle tool? Also what power magnification do you think would be good for this operation? I have an old 20D that I was planning on converting to IR but now I'm thinking "why stop there!" Let's take off all of the filters. I'm very much looking forward to seeing the cleaned up version.

Tristan

Link to comment
Share on other sites

  • 3 weeks later...

Hi Dave,

I can't wait to see some more results on this, great that someone is getting these stunnig results and is breaking some barriers here :)

I'm thinking of getting a second hand 350D body (they are very cheap now) and do the sensor debayer myself, i already have modded mine with the Baader filter and a TEC with cold finger method, it wors beautifully, I can only imagine how it would be to have two identical cameras like that, one with sensor debayred with also TEC cooling and the other with the color sensor for just color captures, would be like having o sensitive mono camera but without the need to buy a 2" filter wheel and RGB filters, what do you think about that? :)

Greets,

Luís

Link to comment
Share on other sites

There are a number of things that might attack the plastic part or glue to make getting the glass out easier. See

rfelektronik.se/manuals/Datasheets/solventguide.pdf

Cement cleaner might be worth trying. It's about 10% or so hydrochloric acid. Nylons are often used for electronics. Other chemicals can sometimes be found by searching the name as it may be more common than you think. The glue is probably a UV cure type so a search for solvents for those might bring something up.

Looks like some help with dcraw could be useful too so I have attached a list of it's options.

John

-

dcrawoptions.txt

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.