Adding cooling and filter wheel to a debayered 1100D mono DSLR

[Gina]

Having successfully debayered an 1100D sensor I will now be working on set point cooling to around -15C and adding a remote controlled filter wheel whilst retaining suitable back focus for use with focal reducers, field flatteners and coma correctors. This thread will describe my experiments and adventures as I progress along this journey.

[Gina]

For the cooling, I need to construct a sealed and dehumidified chamber around the imaging sensor and shutter with as small a volume as possible. Outside of that another enclosure will comtain the rest of the camera parts plus an Arduino to control the cooling. Also in this outer enclosure I propose to include a filter wheel also controlled by the Arduino.

Here are a number of photos showing the main frame and shutter assembly plus the sensor assembly with cold finger and also with the main board added.

1. Main frame and shutter assembly with the battery box separated
   
2. Ditto from a different angle
   
3. Imaging assembly with cold finger added
   
4. Ditto without the battery box
   
5. Side view showing gap between imaging assembly and main frame
   
6. Main board attached to main frame
   

[Gina]

I've been studying a number of options for the sealed and dehumidified chamber.

1. Include most of the camera including main board and battery box in a rectangular chamber
   
2. Include just the imaging side of the sensor and shutter parts
   
3. Add an optical window between sensor and shutter and seal the sensor side (used by CentralDS - thanks to Milos for pointing it out)
   

Option 1. is what I have used in the past. It provides the simplest solution but also quite a large volume to dehumidify. It has the advantage of providing a dry environment for the camera parts regardless of how much dew there is in the atmosphere.

Option 2. is what I have been investigating recently to see if it's viable.

Option 3. provides the smallest dry chamber volume but the optical window will be cold and hence need some means of demisting if atmospheric air is allowed to get to it with it's moisture content.

Two considerations in choosing :-

1. Keeping the dry chamber as small as possible to ease drying
   
2. Keeping the volume and mass cooled by the Peltier TEC small and thermally well insulated to improve efficiency
   

[Gina]

I've decided to go for the KISS approach - Option 1. I'm pretty confident I can make this work. I like the idea of having the camera parts kept dry too. I have ordered some copper sheet to make up a dry chamber. This will enclose the back of the camera with the Peltier TEC hot side in contact with it, helped by some thermal grease. The CPU cooler will go on the outside of the copper box opposite the TEC. I'll draw up a diagram to illustrate this.

[Gina]

Here is the diagram showing the imaging assembly with cold finger, Pelter TEC and copper box.

[Gina]

I'm looking now at the optical window to seal the dry chamber for the light path. Seems to me I might as well combine the sealing properties with a filter for light pollution or at least IR/UV blocking. So... possible options :-

1. IDAS LPR filter - expensive - http://www.firstlightoptics.com/light-pollution-reduction-imaging/hutech-idas-p2-light-pollution-suppression-filter.html
   
2. Baader Neodymium Filter - http://www.firstlightoptics.com/light-pollution-reduction-imaging/baader-neodymium-filter.html
   
3. Baader IR/UV blocking filter - http://www.firstlightoptics.com/uv-ir-filters/baader-uvir-cut-filter.html
   

I'd welcome opinions, please, bearing in mind the camera will be intended for DSO imaging mainly NB and maybe luminance.

[Gina]

I repeated my question about filters in the Discussion - Cameras forum and got a few replies there resulting in my deciding to go for an IR cut filter at a lot less than any of those above.

[Gina]

Here is a photo of the camera with the front cut off ready for mounting the filter wheel. In fact next to the camera body and screwed to it will be an aluminium plate attached and sealed to the copper box and with an optical window (IR cut filter in fact). This plate will extend beyond the copper box to provide a mounting for a stepper motor which will drive the filter wheel directly. Then this whole assembly will fit into another case to keep light and dust out and provide connection to telescope or lens. I'll post a diagram later.

[Gina]

Side view drawing of camera with filter wheel.

[fwm891]

Wish you well with this Gina.

Reading above you were looking at a smaller chamber around the sensor with an optical window, but worried about frosting of the window - couldn't you channel the exhaust air from the tec fan across the front of the OW to keep it frost/moisture free?

[Gina]

Thank you Francis

That's an idea, thank you, but I think the OW will be alright as the whole camera is enclosed by the copper box which is heated from the hot side of the Peltier and heat will be conducted to the OW by the aluminium plate.  I expect the filter wheel housing to be slightly warm too.  The cold parts will be thermally insulated from everything else.  This is a change from previous designs in that the dry chamber will be warm except for the cold finger, sensor and imaging PCB.

[Gina]

Copper box bent up ready for soldering.

[Gina]

Filter wheel carousel design.

[Luis Campos]

Cool project Gina,

Yep, if you want to go for really low temps. the only way to go is a sealed box and a sealed sensor chamber

On mine hard as I try when the temp goes under -3 or so I start to have condensation problems, still have to purge the sensor / coverglass chamber though.

Good luck, I'll drop by to see how it goes

Cheers,

[Gina]

Thank you Luis

With a previous cooled 1100D I could get down to -15C without condensation with fresh silica gel desiccant but the sealing wasn't perfect and after a couple of weeks the condensation was back and the silica gel exhausted.  This version will have a much smaller dry chamber and it will be well sealed.  Previously, the whole camera (the needed parts only) were all inside a plastic box and I had no insulation around the cold parts.

[Gina]

Here is the FW carousel plus the outline of the dry chamber (copper box) and the aluminium plate for the top of the dry chamber and bottom of the FW enclosure.

[Gina]

The carousel will now be driven from the edge rather than the centre.  Centre drive would have meant the centre would have had to be several mm outside the dry chamber which would have meant a bigger box (and heavier).  I would have liked direct drive from a stepper motor but it's not really practical.  Instead I'll probably use a modified miniature servo motor and optical sensing for both home position and each filter.

[Gina]

The self-indicating orange silica gel bags I ordered came today but they're a bit bigger than I thought.  Not to worry, I can always take some of the beads out   These are 20g bags (nice perforated nylon bags) and I thought the paper bags I got before were 10g but they couldn't have been as these must be 4 or 5 times as big.  Anyway, I now have enough silica gel beads to last a lifetime   Not a bad idea to have plenty of desiccant about though

[Gina]

Made a mistake in the diagram above - dry chamber and ali plate 10mm too big...  Here's a correction.

[Gina]

Here's an updated cross-section diagram.

[Gina]

The nosepiece IR filter has arrived and I've got the filter out of the carrier.  I couldn't use it in the carrier as it's 12mm long with T2 male and female threads and ther simply isn't room for that.  I'll glue it to the aluminium plate I think.

[Gina]

I've now soldered up the copper box that provides the main part of the dry chamber.  I now need to sort out how I'm going to attach the box to the 2mm aluminium plate that forms the top of the dry chamber and takes the optical window and provides part of the housing for the filter wheel.  It needs to be sealed yet able to be taken apart

[Gina]

It's working a bit with a charged battery and I also found a ribbon cable not all the way into it's connector.  Now the shutter works (by the sound of it) but there's only a faint glow from the LCD display and nothing gets recorded to the SD card.  I'll try it on the netbook next.  All the ribbon cables seem to be well in their connectors now.

[Gina]

On the netbook, some things work.  The flash works, the shutter rattles away but nothing is captured either on SD or the PC, live view doesn't work either.  Looks like I'll have to leave the 450D until I get another sample 450D.

What I've learned so far regarding the 450D :-

1. The cover glass is easy to remove like the 1000D
2. The CFA removal presents no problem - coarse then fine abrasive produces a smooth result.
3. The construction is the same as the 1000D
4. The resolution is slightly higher that the 1000D
5. The data depth is 14bits as opposed to 12bits for the 1000D
6. Max ISO is 1600 same as the 1000D

Work on the 450D is now on hold until I can afford and get another 450D (probably faulty).  Meanwhile, I shall be working on another 1000D sensor and I also have 350D and 300D sensors to play with.

[Gina]

I have a few test shots with the mono camera mounted on the MN190 and lightly cooled with a CPU cooler directly on the cold finger.  I shall be doing more when weather permits.  These are all single subs with Baader 7nm Ha filter.  A few shots of the moon, the Eagle Nebula and finally a flat.