My Mark 6 ASC has been working for about 9 months but now the acrylic dome has yellowed and become less transparent in spite of being an expensive CCTV camera dome designed for outdoor use and guaranteed not to yellow..  This means either getting a replacement every year (or less), using a cheaper acrylic dome or dispensing with a dome altogether and adding a motorised cover for the lens which is only removed when the camera is in use.  I have not yet decided on which option to use or the design of this ASC.

These photos show the yellowing of the dome, a cheap smaller dome from one of my earlier ASC versions and the internal parts of the Mark 6.

These images show the reduction in quality and the lights reflected in the dome.  Discarding the dome would get rid of these but the lens would want protecting from bad weather and wind blown dust and therefore would want a cover over it when the ASC is not in use.

Here's a photo of the Mark 6 ASC with lens.  The yellow bracket holds the RPi board and locates the unit within the casing.  This will require a redesign, as will the casing either to use no dome or to use a different dome.  I tend to favour no dome. 

Now then...  Motorised cover...  I guess 28BYJ-48 stepper motor and A4988 driver module to drive it.  That isn't a problem.  More of a problem is the cover itself and how the drive is arranged bearing in mind that the stepper motor wants keeping dry.  The cover either wants to seal to the lens or have a skirt that goes down over it sufficiently far that the horizontal rain we get here sometimes, can't blow up under.

I don't think sealing would work well enough with a remote controlled cover so I shall concentrate on the skirted version.  The idea I have is to provide a tube to go round the lens and sealed to it and have the cover like a lid with curved top going over that.

Here is a cross-section of the lens tube and cover together with a pivot about which the cover might move.  The problem with any cover with a long skirt is the cover catching on the tube as it opens or closes.  This is shown in the two diagrams below for a typical case - shown to scale with a 15mm skirt.

If the pivot is further out the problem is lessened.

Skirt and lens tube extended and taper increased slightly to improve clearance.

Coming now to the rest of the ASC, here are two photos of the insides.  With and without the lens support.

The lens support being aluminium will conduct heat from the camera to the lens which might be useful if the camera was allowed to warm up as it would warm the lens and help prevent dew.  However, a warm camera produces more noise and this was the reason for adding the Peltier TEC cooling.  The cooling pad can be seen between the camera and the lower aluminium heat sink, which dissipates heat to the aluminium pipe that supports the whole unit. 

If cooling is off the camera can warm up when in use but if the camera were mounted directly to the aluminium block it would be cooled to near ambient.  In view of the sensitivity of the camera and good light throughput of the lens, this would probably be adequate cooling.  But if the lens were at ambient temperature in the winter there is a strong probability of dew settling on the lens.  I really think the lens could do with being heated.  The upshot of this is that the lens wants to be thermally insulated from the camera.  Then the camera can be cool to reduce noise and the lens can be heated to prevent misting up.  In other words the lens support is a nuisance.

I think I shall try the ASC without the lens support.  I'm not sure about the Peltier TEC cooling but the cooler pad is already there and powering it with control from the RPi is no problem.  Similarly heating the lens would be no problem - I just need to decide the best method of heating it.  Heating the air in a dome is easy - just a string of resistors.  For the lens it might be better to use Nichrome wire insulated with Kapton tape (high temperature thin insulation).

Checking the materials used between camera and lens I find they are all black anodised aluminium but at least there would be less aluminium conducting the heat.  The only way round this would be to attach lens to camera with plastic but this has its own problems.  Maybe the anodising on the threads of the various adapters will help with thermal insulation.  Just have to suck it and see but it probably means the Peltier TEC cooling will be needed.  Maybe I could arrange a heat conducting copper strip from the hot side of the TEC to the lens   Copper conducts heat much better than aluminium.

Here's another diagram.  If I print the lens cover in black filament I can use it closed for taking darks and eliminate hot pixels.  I shall probably print the main casing in white ASA filament as that is UV safe and will stay cooler in bright sunshine.

Designed and now printing a test lens tube (to see if it fits).

Designed and now printing the main part of the domed cover.

I think there might be problem printing this unless I reduce the support spacing as the filament is drooping inbetween but I'll eave it printing and see if I have enough clearance from the lens tube.  Whilst a dome is esthetically nice a flat top is much easier to print.

As I thought the dome was rubbish.  It needs finer printing but that takes a lot longer.  The aesthetics are not important - it just needs to work!  The flat "roof" could have a slight slope on it.  I don't really want to spend too long on this project but I want an ASC working again.  I can be too fussy at times

Going for my smaller high resolution printer, "GinaRep Pilot", for these parts and I'll leave my bigger "GinaRep Titan" for the bigger casing parts.

A bit later :- Hmmm... Pilot printer is not working properly.  Oh well, have to put up with fast but coarser printing

The cover and its operating mechanism will need making in several parts which could either be made of ABS and glued together or made of PETG and fastened together with nuts and bolts.  Think I favour the former - my Titan printer is equipped with fume extraction so fine with ABS.  Fumes and nano-particles are blown into the loft.

I'm seeing if a dome top could be printed on its edge.

That went fairly well and I think I shall be able to produce a usable print.  I shall print it in black ABS and thicker.  That print is just 2mm thick.

A couple of photos.

Here is a design for the lens tube with bracket for the cover axle added.  Now printing.

Now to the drive mechanism for the cover.

Changed the design of the cog part for more strength and printed it, then glued it to the cover part.  The final version will want a bit of cleaning up of course but this is just a test of the design.

Seems alright so far but there's the drive from stepper motor to the gear to sort out with particular attention to sealing the motor from damp.  The only way I can think of getting the drive from the inside to the outside is by a shaft.  And not out of the top or it would need a waterproof gland.

I have most of the parts to complete the rest including an RPi HAT, stepper motor and driver module so I should be able to design a casing to suit.