Gina

Just bought a broken 1100D from ebay auctions for debayering the sensor. Bit dearer than the 1000D but a better model.

Is that random noise or interference patterns?

The 1mL syringes came this morning They look just the job - long and thin and should provide much better control than the 2mL syringes I have which are much thicker and shorter.

TBH I would be afraid to use a sensor swab so close to the gold wires I think I'll just apply enough resin to reach the sensor itself and cover the gold wires and maybe when that's set go round adding a tiny amount of resin to cover the blue border.

The 450D I bought from ebay auctions as not working came this afternoon and is working fine now I've put a battery in it. It is body plus IS zoom lens

Put the 1100D sensor back in a working camera and took a photo. Sensor still working, at least at ambient temperature. Now I need to find my IPA and clean it up ready for debayering. This photo is without any filters in front of the sensor.

I've ordered some 1mL syringes Amazon Prime due for delivery tomorrow - enough to keep me going forever Cheap enough to use and throw away @ just over 10p each

Heres an image from the 1000D sensor. Where I removed the errant blob of resin and pushed it back in line with the rest, it crept back onto the imaging area a bit overnight. This is just a standard L jpeg image saved to SD card and downloaded to the PC then contrast increased and resized in Photoshop.

The problem with whatever you use is that resin will creep underneath and get on the sensor that way. I think the only way is to apply the resin slowly and carefully round the edge and not let it get onto the sensor surface. I know we want the blue edge protected as well as the gold wires but the resin will stop et the edge if you don't apply too much. Maybe the answer is to apply a bit more resin once the first lot has set enough to protect the gold wires - very tricky and needs a steady hand. I found that if I applied enough resin to cover the blue edge it refused to stop there and spread onto the image area. I've been using a 2mL syringe with one of the thin plastic tubes from a cotton bud but the syringe plunger tends to go in fits and starts and drop large blobs of resin gaps between I need to find a better way of extruding resin onto the job. For a start I'll see if I can find a source of thinner syringes.

Used cotton buds to clean the epoxy off the sensors' image areas. Still some loss around the edges but it'll do. 1000D and 1100D sensors.

I've mixed up some more epoxy potting resin and applied it to the 1000D and 1100D sensors. Didn't go as well as the other day though. On the 1000D one I had a twitch as I was dribbling the epoxy around the edge from the syringe and got a blob on the image area. I daren't touch it until it's set or I'll make it worse. With the 1100D I had a bit of a problem with the glass, though not what I had expected. I got the epoxy under the glass alright but the surface tension sucked up resin from the rest of the border and started forming a wider epoxy border. So I have reduced image area on that one. Still, I guess I should still be able to test the effect of CFA and micro lens removal - assuming the epoxy doesn't kill the sensor.

No, I haven't and my eyesight and dexterity would no longer be up to it.

It now seems that all four 350D main boards are faulty in one way or other so I think I'll abandon 350D work for the time being. The two 1000D cameras both work. One is fully debayered but with some image defects. The other is partly debayered with a border of untouched image area round the edges. In both cases the gold wires are still bare so I could test the epoxy on those. Once hard I could then continue right up to the edge of the image area of the one with some CFA left but there seems little point in adding epoxy to the fully debayered sensor. I think the first go at adding the epoxy went well enough that I could try it on the 1100D sensor though the remaining piece of cover glass will make things more difficult.

The 1000D is simple to cool using the same cold finger shape and path as the 450D and 1100D and there are no SMDs on the back of the image processing board and a nice big gap

The connector is beyond repair as a small gold pin has broken off. I have a couple of other main boards though. I think the 350D will be just an experimental camera for debayering though as adding cooling is difficult. I may well see what the 1000Ds can offer and I expect to see how the 1100D goes. There is also the 450D which has the advantage of an easily removed cover glass. AND... I have just bought another 450D from ebay auctions OK, I know I said 1100Ds were what I really wanted but they are in short supply at a decent price. I'm enjoying myself with this series of projects so don't mind spending some money on it ATM I haven't any aspirations for more telescopes

Got the clips back in but bent a pin on another connector which I think may feed the shutter - getting error 99. Something else to fix under high magnification

Resin has set and I've put the sensor back in the working camera. BUT... the plastic clips have come off the sensor main board connectors and without them I'm just getting a blank image. The clips aren't broken - just seem to have come out so I've either got to get them bacl=k in or see if any of the other mains boards work. I won't risk attacking the CFA until the resin has hardened more. Fully cured apparently takes a week at 20C.

350D sensor photos. Cover glass removed Epoxy potting resin added with syringe and tube from cotton bud Epoxy resin touched up with tiny drips from a toothpickThis resin is quite runny when mixed.

Just checked the 1000Ds. Both working and both with debayered sensors One shows some slight defects but the other looks quite good.

I've taken the sensor assembly out of the non-working 350D, dismantled it and removed the cover glass with my Exacto knife. Just took it very carefully, gradually working round and round the glue line with the tip of the knife slowly increasing the white area. Eventually I got it freed all the way round and lifted the glass off. After getting the white right across the glue area there is a rim on the inside and further careful working is needed to break this. Anyway that's that part done Slowly, slowly catchee monkey Wish this would work with the 1100D cover glass

The glue is extremely thin even a sharp knife barely gets under it.

I think the 1200D would suffer from the smaller pixels when used for astrophotography. The 1100D with its 12Mpx is getting on the small side so going up to 18Mpx would mean even smaller pixels. For AP I would like to see fewer pixels while retaining the Digic 4 image processor but Canon are unlikely to go in that direction!

Have taken the case off No. 3 - the silver one - and can see I haven't touched this as there is a screen over the main board covering the sensor connections with original soldering to the metal frame. As this camera shows no sign of life I think I'll just take the sensor out to work on and test with the working camera.

Found the 300D but not looked at it other than to see that I've had the screws out and connected external power wires. I can see a sensor through gap in side but no idea of its state. As for the 350Ds... Working and has a partly debayered sensor - it is capturing mono images to the CF card in daylight. Non-working. The sensor has had its cover glass removed but CFA is intact. Complete with all screws in place (so probably not touched be me). Put in fully charged battery and switched on but nothing - no sign of life. Silver coloured case. In bits without sensor.

I have yet to see if any of these cameras work - I have two different batteries on charge one for 350D and one for 1000D. I may have one or more additional cameras somewhere about the house - I'm sure I had a 300D but can't find it ATM. If anyone has any preferences as to which model they would like to see my work on, please say as I may have a choice derpending on what works.