All Sky Camera Revisited

[Gina]

With rain all day, I have been able to assess the effectiveness of the dew heater.  With the failure of the power MOSFET that was driving the dew heater and the latter connected to the TEC, the dew heater has been running off 5v.  This has proved adequate for dry conditions so far but there is now some misting of the dome and it wants full power.  It would seem that 3 power levels might be appropriate for the dew heater viz.  OFF - 5v - 13.8v.  This would require changes to the circuitry and to the Astroberry Board coding to provide the control.  ATM the only way to get full dew heating is to set the cooling to HIGH.

[Gina]

With dew not clearing from the dome and the need to fix the dew heater control I have brought the ASC indoors.

The moisture is on the inside of the dome and yet the colour-change desiccant is still showing orange meaning DRY!!  Well, the dome certainly isn't dry on the inside!!  So I haven't cured the problem of ingress of moisture into the casing!

[Gina]

Looks like I need a complete redesign of the ASC.  I'll list the requirements and see if anything comes to mind.

1. Capable of airtight seal
2. Transparent casing or part of to show colour-change desiccant to confirm dry inside
3. Big enough to take camera with lens and thermal lagging plus the electronics
4. Cooling for the Peltier TEC
5. Ability to take apart for maintenance yet retain airtight seal
6. WiFi transparent casing

[Gina]

The experience of the present design shows that the camera doesn't need cooling as much as I thought so the super efficient cooling is OTT.

[Gina]

I'm thinking a totally different approach is indicated.  The idea of a sealed and dry environment for the cold camera was fine as long as a sealed casing was practical.  Now I think it isn't and I've had a really good try but not succeeded.  Also, the camera I'm using now doesn't need as low a temperature to get rid of noise and hot pixels. 

Edited June 4, 2020 by Gina

[Gina]

I've taken the ASC apart and a few drops of water came out of the camera.  Fortunately, any condensation or ice forms on the case of the camera and not on the PCB or components.  The sensor will always be warmer than the case.

It's clear that if a perfectly sealed enclosure is not possible, there has to be provision for melting ice to drain out of the camera away from any electronics.  Condensation of moisture on the camera case will counteract the cooling slightly and there is always going to be some dampness in the air except in very dry weather.  If the camera case if cooled below freezing, ice will continuously form on the inside and any outside surface not lagged.  It may be sufficient to cool the case to just above freezing and provide a pipe to carry condensate away.

Condensate wants keeping away from the hot side of the TEC or it will evaporate and fog up the inside of the dome and also condense on anything cool.

Edited June 4, 2020 by Gina

[Gina]

I think somehow I need to keep the air round the camera separate from the air under the dome.  Maybe take incoming air and warm it up with the heatsink that's on the hot side of the TEC preferably taken from inside the observatory where it's likely to be drier (warm room perhaps).

[Gina]

I think I would like to use the same mast pipe as it provides a good way of mounting the ASC and simpler than attaching it to the roll off roof.  Makes it easier to take off for maintenance too.  The pipe was used to carry the power cable and water cooling pipes - now it could carry air from the observatory as well as the power cable.  For the air supply I would use a 3D printed adapter to enclose the pipe and provide a side-entry.

I have a 40mm x 40mm x 20mm heatsink that would fit on the hot side of the Peltier TEC.  The incoming air could blow through this and be ducted up and into the dome.

 

Edited June 4, 2020 by Gina

[Gina]

I shall have to set up a test rig to check that this heatsink is suitable.

Edited June 4, 2020 by Gina

[Gina]

That's the air intake but I need to sort out how and where it escapes too without allowing wet to get in if we get horizontal rain!

[Gina]

I'm thinking that heatsink may be too small and I have a bigger one which may be more suitable.

Firming up on the design, relatively dry air from inside the observatory could be blown up the hollow mast and into the ASC casing, firstly cooling the electronics then through the fins of a heatsink on the hot side of the Peltier TEC.  From there warmed air could be ducted up to the dome and down the other side and out down an air duct to the bottom of the casing and out to the atmosphere.  As before, the camera would be lagged to help with the cooling.  A pipe could be used to connect the camera enclosure to the relatively dry observatory and carry any condensate away from the camera.

[Gina]

I think I'm changing my mind on blowing warm air from the heatsink up to the dome.  Even relatively dry air from the obsy is going to have a certain amount of dampness in it (particularly in wet weather) and it would be better if the air in the dome was as dry as absolutely possible.  I could still use air cooling of the heatsink but vent the warmed air straight out and just use a small positive pressure of dry air in the camera and dome area.

[Gina]

Now I'm wondering if piping air from inside the observatory up to the ASC is actually more complicated than water cooling.

[Gina]

Screenshot of camera, lens, TEC and 60mm heatsink crudely modelled.

[Gina]

The enclosure will be in several parts.

1. Bottom - conical section attaches to mast pipe and contains the electronics.
2. Middle - rectangular hole in bottom to fit round the heatsink.  Contains the heatsink and has air ducts at the side venting warmed air downwards.
3. Top - contains the Peltier TEC, camera and lens.  Topped with the dome and dew heater.

May further split parts depending on how the detailed design goes.

Whilst it seemed a good idea to use the waste heat from the Peltier TEC to warm the dome, I haven't been able to think of a way to get this heat to the dome without blowing damp air onto it.  Decided on the simpler approach of a separate dew heater.

Edited June 6, 2020 by Gina

[Gina]

I still think this is too complicated.  Particularly piping air from the observatory up to the ASC.

[Gina]

Here's a list of things I would like :-

1. Keep the mast pipe.
2. Sealed enclosure for camera and dome - OK almost sealed with positive internal pressure.
3. Simplest possible effective cooling of Peltier TEC hot side.
4. Reasonable appearance.

[Gina]

As I shall have to run the USB cable through the sealed part of the casing anyway, I think I'll put the electronics inside the observatory.  At least then if I get a fault in that I wont need to take the ASC down.

[Gina]

The only simpler arrangement I can think of is to put a fan directly on the heatsink.  I'm not really happy with that as it will be in damp air when it rains and if it fails it would need the ASC taking down.  Also, it might be difficult to tell if it failed - the TEC would overheat and fail and that is not a cheap item!  The less there is up in the ASC itself the better really.  A fan inside the observatory could provide air pressure for the sealed compartment via the desiccant pot.

Another alternative is back to water cooling which may actually be a bit simpler and even if the circulating pump fails there is still water in the waterblock (until it boils).

Edited June 6, 2020 by Gina

[Gina]

Water cooling is indeed simpler and I have most of that done.  Makes for the most compact ASC body too.

This is the design making use of the previous parts ie. the conical connection to the mast but it would be better with a new conical part.  Dome not shown in this model.

[Gina]

With new conical connector.  Somewhat more elegant I fancy! 😀

[Gina]

I'm hoping I have the ASC design sorted out.  3D printed casing with water cooling and a small positive pressure applied to the camera enclosure in case the casing isn't completely watertight.  Minimal kit up top with the electronics in the observatory.  USB cable and a 4 core round cable plus 2 water cooling pipes and an air pressure pipe will go from the observatory up into the ASC casing.  The air pressure pipe will also serve as a condensate pipe if any damp does get into the ASC enclosure.  The air pressure will be provided by a small radial fan.

[Gina]

Designed a 150mm (6") long lever to adjust the focus with the camera on a fixed mounting, pointing out of the living room window to a far hill & trees some 5 miles away.  Even so focussing was critical and a mm at the end of the lever made a difference.  Remains to be seen how good the focus is when I get the ASC viewing the stars.  I would prefer not to add a remote focussing system but I will if I think it necessary.

[Gina]

It turned out that the dome was around 7mm too low with the present camera enclosure so rather than producing a new camera casing I designed and printed a 7mm spacer.  This is glued and screwed to the camera casing and the dome is glued to the spacer.  The top of the lens is now at the centre of the dome curvature for minimal distortion.

[Gina]

Tested the camera casing and dome for leaking and it failed!!  Seemed to be the dome seal so I shall need to improve this.  I could use epoxy resin but that would prevent removing the dome unless it were attached to another ring that was itself attached to the camera casing but then the seal would move to the junction between ring and camera casing.  Might make sealing less difficult though.