After cleaning the sensor again, I tried again last night and lasted only one sub before the sensor fogged over again. This is very frustrating, since I took great care to seal the camera and purge it with argon. It's becoming clear to me that without a cover glass over the sensor (effectively double glazing) the sensor and the cooling running at 2C - it just becomes a dew generator on humid nights.
All colour sensors in Canon DSLRs, and other brands, have a cover glass over the sensor surface. You can't tell it's there but what you see when you look at the sensor is the coverglass first and the sensor surface below that. It's there to protect the sensor, and supports the 'shaker' dust cleaning devices. The cover glass might have a coating to discourage dust attachment (anti-static coating).
When CentralDS debayered my sensor they had to remove the cover glass to do that work. CentralDS do offer to replace it with the stock (highly reflective cover glass) or to put on a Astronomik MC Clear cover glass, at additional cost of course. I opted to have them just send me back the bare debayered sensor and I was intending to use my Astronomik MC Clear EOS Clip-In filter. Well a MC Clear Clip-In does provide dust protection, it does nothing in terms of condensation double glazing.
So for now I am packing this project away for a couple of months. I have a EOS filter drawer on the way from Teleskop-Express, and I need to buy the small Astronomik MC Clear glass that fits into the sensor filter frame - and allows me to create a double glaze chamber just over the top of the sensor; I believe this may solve the condensation problem. If not I may have to get to a cover glass heater, which Roland has developed.
Just more development problems to deal with, I guess I will get there eventually.
In my case I just added a 2" glass window to the 2" T-Adapter nose piece. I also have an Ha 2" filter I can use instead. Effectively gives a sealed front end for Argon or Silica bags.
Also the distance separation from the chilled sensor means the window doesn't get quite so cold and prone to icing but I can use a warm air feed line from the TEC radiator and fan to add a little warmth across it if necessary.
Still waiting for delivery on the temp sensors from China so in a holding pattern present
I have taken some test shots with the debayered sensor (now cleaned as best I can). Because I don't have the cover glass replacement for it yet, I decided to turn the cooling off and just take these at ambient. Because they are taken through my RC08 (f8) I had to run fairly long sub times and shoot at ISO1600 (hence more noise), and the pesky moon was a factor last night as well, so excuse the quality.
What I was after is to determine if the debayering had worked and that all pixels are functional, and whether there were dead spots that might have been damaged or are masked by something that potentialy could be 'polished off'.
So here are two simple jpg files of well known brighter objects. I over exposed them to see if I could light up everything on the sensor, and being ambient they have some noise.
Let me know what you think of the quality? I won't give you my opinion just yet.
I think I see what you're referring to Glen. Overall though it looks OK, nothing that an aggressive dose of dithering wouldn't cure...
What was your exposure time?
Dunk for M42 the sub was 190" and for Eta Carinae it was 240". I was guiding and it was ok for the one of M42 but the Eta C guide graph was more messy and I suspect a counterweight had slipped down the shaft as I did that one last.
I guess after having shot those targets many times with the colour camera, and produced some nice results, I get frustrated by not be able to replicate those results with the mono sensor.
From a luminance point of view, those subs are not much longer than I've taken recently with my f/5.5 scope, taking into account the extra f-stop or so advantage...I was exposing for 150s.
The difference with the mono sensor is you'll need to take separate colour subs
But maybe from a resolution point of view the seeing just wasn't up to it that night? I've noticed that before with my 1100D even though it has the CFA, so I'd expect a mono sensor of the same resolution to notice it more readily. Of course, this is the same effect we notice when planetary imaging...you've just got to try it on the night and hope for the best.
Yeah, but I am not going to be doing RGB for awhile. Initialy it will be narrowband but I need the filter drawer before trying that. Your right about the seeing, summer heat does not help.
By my rough calculations, that's 0.66"/px which to be fair is probably better than the seeing will allow on most nights. At the same time, on a decent night you should resolve more detail than a sensor with a CFA.
Like I said, on dodgy nights I can see the difference at 550mm focal length with the CFA, let alone 1600mm
I had a response from CentralDS this morning in relation to the first test shots I had sent to them. They claim the sensor was clean when they sent it and that by my attempting to clean it I had buggered up the epoxy coating they used on the contacts, and this had bled onto the surface. Now I have a real problem with that response as:
1. The first test shots that I had sent them were taken with the sensor right out of the box from them and I had not cleaned it, other than a puffer blow bulb. I wanted to illustrate to them what it was like with the evaporative pattern on it when I received it.
2. They had provided absolutely nothing to me in the way of cleaning instructions or other information, just a sensor in the box.
3. When I did clean it for the last set of test shots, it actually improved from the delivered state. And cured epoxy is insoluble to cleaning fluid and I only touched the sensor with a small sensor cleaning swab - never straying onto the contact area.
etc etc.
Anyway, I have offered them three options:
1. I will return the sensor to them so they can clean it and fix it.
2. Or they can replace the sensor with a new one that has been debayered, and properly cleaned.
3. Or they can refund my money.
I will be sending the sensor back to them after the Christmas postal shutdown and they can decide what they do from there. I've decided I am not going to stuff around with it any longer. Grr, it's enough to make me buy a CCD camera to avoid all this stuffing around.
CentralDS just advised they have sent a full refund to my Paypal account. No admission of course, but they did not seem to want to get the sensor back to have a look at it.
So I will leave it till the new year before deciding what to do with it.
There are definitely a few dead spots on those images, but they can be smeared away with a clone tool in PS.
Your chip looks heaps better than the ones I have seen on Stargazers Lounge where they have been scraping the sensor to get the bayer print off.
Theirs are all ragged at the edges where the user is too scared to scrape close to the edge in case they damage the wire connectors.
At least you got a very cheap de-bayered sensor now, as your funds have been returned.
I wouldn't give up on it, as it is still very usable even if you have repair the final image later with the clone tool.
I may have worked out a way to salvage the sensor and have ordered some equipment to enable me to tackle it in the new year. I have located some fine glass polish, pointed sensor swabs, and a sensor loupe for magnification, which together should allow me to get right to those spots. As long as the spots are remnants of the debayering process I may be able to remove them with glass polish but it's a risky step as you don't want to cut into the pixel itself. I will use the JTW debayering tutorial as a guide. More to come on this. I'd like to hear from anyone that has done this sort of thing.
Another step completed with the arrival of the EOS filter drawer from Teleskop-Express just before Christmas. I will be using the 36mm unmounted filters so needed that drawer to go with the T2 EOS bayonet attachment. A small S/S spacer is required (also provided by Teleskop-express) to avoid having the standard depth T2 adaptor intrude into the filter drawer area where it could block drawer withdrawal. It seems strong enough as I can pickup the camera/cooler assembly by the adaptor without any movement. Sensor backspacing is unaffected as it takes up the same depth as a standard EOS Bayonet adaptor. I just have ordered a set of unmounted 36mm Baader Narrowband filters, seen here with curves.
I wanted to share some info on what I consider to be an essential tool for DSLR modification - the magnifying work light. The one I use came from Bunnings (haha - where else), and despite that, it has proven to make the whole mod thing much easier than doing without it, or trying to hand hold a standard magnifying glass. The main lense is a 3x diopter, and the little spot on the main lense is a 12x diopter. The diopters and the circular floro light make it ideal for seeing what your doing, whether that's filter removals, ribbon cable attachment, cleaning the sensor, etc. Some photos attached. It's called the Maggi Magnifying Desk Lamp:
Haven't seen them at our Bunnings over here but I'll have another hunt around. Which section did you find them in ?
I struggle with my eyes these days.
Haven't seen them at our Bunnings over here but I'll have another hunt around. Which section did you find them in ?
I struggle with my eyes these days.
Brent they are in the light section, near the normal desk type lamps. Not all Bunnings may stock them, the one near me is pretty big and I rang ahead to make sure they had them. Just ring them with the item number (I/N: 4370313) and check, even if they don't have it they might be able to get them in.
All of my sensor cleaning gear arrived this morning and I have dived right into it. Using the glass polish I believe I have removed the obvious spots, they came off pretty easily. I have inspected and reinspected every square mm of the sensor using the magnifying lamp (3X), the Sensor Loupe (7X), and believe I have it servicable now. My only concern is that I may have 'gone too deep' with the polish and I need to do some test shots to check on the quality of the job. Photos attached. Note that the sensor surface is highly reflective and when moved shows shifting light diffraction off the surface that appears to form a border in the one photo, it's not actually there but a reflective artifact from the light.
Update: The camera is re-assembled, but I have left the cover glass off the sensor in case I have to go back in and do some more work. I am also leaving the cooling off for the moment while I do some test shooting at ambient. I need to be sure it's as good as I can get it before I button it all up into an argon chamber. So attached is a sunny daytime sky shot. I have expanded the RAW file 100% in Photoshop and it looks pretty clean. The dark spots in the lower left are a couple of specks of dust probably on the MC Clear Clip-in filter. I can see some stuff on the very edges of the sensor and in the corners which needs more attention (hard areas to work into) , but looking promising. I will put this camera on a scope tonight and try to get some long subs to see how it goes. Stay tuned.
Update: The night sky test was terrible, it is obvious that the sensor is stuffed beyond recovery. It takes clean subs now but the sensitivity is shot, I must have cleaned it too deep and impacted the pixel surfaces. Note for potential debayers: "Do not go into the Orange/Gold layer below the Bayer Matrix surface", even if it means leaving a little bit there. Lesson learned.
Not one to give up yet, I have ordered a replacement sensor from overseas ($40 USD) and it has now been shipped. Whether I try to debayer this one myself or just assemble it as another colour camera is yet to be decided.
My narrowband filter set arrived today and they fit nicely into the EOS filter drawer. I had to use two of the Teleskop-Express S/S spacer rings to prevent the threads of the T-Adaptor extending into the filter drawer assembly (about 1.5mm), but it clears nicely now.
I will report back when the replacement sensor arrives. In the mean time I will review all the online debayering instructions and work out if I want to try it myself.
