All Sky Camera Revisited

[Gina]

1 hour ago, Dr_Ju_ju said:

There are Arduino libraries for the DHT22, I believe Rob Brown's Arduino dew heater system uses them to measure relative humidity...

Yes, I use such a library in my weather station.

Edited April 5, 2019 by Gina

[Gina]

The hermetical seal has failed.  I noticed this morning that the silica gel was no longer orange.  Powered off and disconnected everything then brought the rig indoors.  Opened up the outer casing and found water in the bottom.  Not a lot - less than a ml but enough to mist everything up inside.  Took the bung out of the camera casing and the silica gel in there was still orange so the camera is dry.  Of course, it was a severe test with continuous rain for a couple of days, some of it heavy and with wind.

I'm wondering now if I'm being too optimistic in trying to hermetically seal the ASC outer casing.  At least one commercial ASC just has drainage holes and a dew heater.  With the volume of air trying to expand and contract with temperature changes there is going to be quite a pressure change in the internal environment so the sealing has to withstand this pressure.  I seem to remember seeing mention of some special vents that allow air through but not moisture.

Edited April 6, 2019 by Gina
typo

[Gina]

I've heard that nothing is entirely waterproof so I guess moisture could seep through the PETG casing particularly as a lot of it is only 1.5mm thick.  OTOH sealing both the top and bottom joints (plus the waterblock inside) gives plenty of scope for leakage.

A dew heater is necessary to keep dew off the outside of the dome so I guess it will keep dew off the inside too (it does for others).  Maybe having the ASC vented to the mast pipe would be fine.

One disadvantage with a hermetically sealed unit is that if water does get in at or near the top, it can't get out again and may eventually build up in the bottom and encroach on the camera sealed chamber.

Opinions please ?

[Dr_Ju_ju]

If there's moisture in the air and a colder surface on the inside, then water will condense out.... it would probably better to have a forced air-flow ??

[Gina]

I guess I could incorporate a small fan to circulate the air in the ASC outer casing.

OTOH what do you think about hermetically sealing and desiccant?

Edited April 6, 2019 by Gina

[Gina]

One reason for trying to achieve a totally dry environment inside the ASC casing was to avoid any condensation on any internal parts.  That means not only the inside of the dome but also the lens which is on top of the cold camera with the camera cooled to as low as -20°C to get rid of noise and hot pixels.

Edited April 6, 2019 by Gina

[Gina]

I'm afraid if the air in the ASC is damp, I shall get condensation in the lens (there is some now but it should dry out).  It would be very difficult to thermally insulate the lens from the camera.  I can't afford a proper cooled astro camera at over £600.

I wonder if a rubber diaphragm to equalise the pressure between inside and outside would help with a sealed outer chamber.

[Gina]

Let's examine what I've learned from this project.

1. Water cooling for the hot side of the Peltier TEC works very well.
2. Cooling the ASI178MC astro camera to -20°C kills hot pixels and reduces the noise considerably.
3. Remote focussing of the 1.55mm f2 fish-eye lens works fine.
4. The 1.55 f2 fish-eye lens and ASI178MC astro camera give just the right sky coverage - a full 180 degrees.
5. Sensitivity is low enough for daytime imaging but a bit lacking for nighttime.
6. Sealing the outer casing and desiccating with colour change silica gel worked fine for a few days after which the seal failed.

[Davey-T]

2 hours ago, Gina said:

I've heard that nothing is entirely waterproof

Submarines ? 

How good a seal can you actually get ? you could purge it with Argon, available in small cans for MIG welding, don't know how pure it is though.

Dave

[Gina]

Yes, though I gather submarines are not entirely waterproof!  Metal is more waterproof than plastic though.  But I don't think ingress through the material is the problem.  As you say "How good a seal can you actually get ?".  No doubt a lot better that I had.  I also want to have a "reservoir" below the camera where any water that does get in can collect without destroying the camera plus an easier way of changing desiccant.  Maybe some forced air circulation would improve the drying of the air.

[Davey-T]

Fan connected to humidistat sounds feasible.

Dave

[Davey-T]

There are a few ASC dotted around the country available on the web, I wonder how they manage ?

Dave

[Dr_Ju_ju]

If you can't get an inert dry gas in, how about producing a vacuum ??

[Gina]

22 minutes ago, Davey-T said:

There are a few ASC dotted around the country available on the web, I wonder how they manage ?

Dave

I think some use lower resolution and hence more sensitive cameras.  Some use that better Fujinon fish-eye lens that I had earlier before I destroyed it with a bit of carelessness!!

[Gina]

17 minutes ago, Dr_Ju_ju said:

If you can't get an inert dry gas in, how about producing a vacuum ??

Ha ha...

[Gina]

Thought I'd see if the top of the outer casing had a leak so I turned it upside down and filled it with water.  Water poured out from the dome seal!!  Probably where the wet got in.

Looks like there's plenty of time before I get any clear nights to test the ASC again

[Gina]

There are a couple of problems with the 3" dome, the flange is very narrow and there's just barely room to get the dew heater in between that and the focus mechanism.  Also, the dome flange has three slots that go in almost to the hemispherical part.  A crack has occurred from one of these about a cm up the dome but it doesn't seem to leak.  The water came in on the opposite side, presumable lacking silicone sealant. 

The 4" dome has a much wider flange and smaller notches relatively.  It would also give much more room inside.  The only disadvantage I can see is that the dew heating would be around 15mm in from the inside of the dome and may be less effective at stopping dew on the outside of the dome.  It's rated at 2W.  Also, a 4" dome is going to take more heating that a 3" one.  I've seen ASCs with 4" dome and the same model of dew heater though.  Guess it's worth a try - I could add extra heating if the heater module is insufficient.

Edited April 6, 2019 by Gina

[Gina]

This is a redesigned top outer casing to take the 4" dome,  A 3D printed ring will clamp the dome flange to the top of the outer casing.  I think I'll use a large O ring or a ring printed from TPU filament rather than silicone sealant.

[Gina]

The bottom sealed joint can also have a TPU ring as a seal.  I need to think more about whether to have a transparent window to observe the colour change silica gel or a DHT22.  The transparent window could be made from some acrylic sheet with a "window frame" in the outer casing and TPU seal.  This would also allow access to the silica gel bags for inserting and later exchanging for fresh ones.  If I can get good seals the silica gel should last a long time before needing changing.  With a separate window to see the silica gel (or a DHT22), I could make the outer casing quite thick.

Regarding an air circulating fan, this wasn't needed for the couple of days the seals worked so if I can get better seals it shouldn't be needed in the future.

[Gina]

Model of outer casing top with window.  Only problem is that the window is rather small.  I'll sleep on it...

[Starlight 1]

I started work on my skycam again after looking at your post. and put it aside again thinking how can I get round that . so sometime a good night sleep help the next day Gina.

 

[Gina]

The window and silica gel could be brought outwards in a "pod" as an alternative.  OTOH a DHT22 in combination with Arduino could give a warning of moisture ingress.  There is a system combining RPi and Arduino with INDI called INDIDUINO and a project using this is MeteoStation which included a DHT22.  This could solve two problems in one go if I can fathom out how to do it - ASC and weather station (project on hold).  It could also come in with my automated ROR project as I have a motor control shield for an Arduino for controlling the ROR drive motor.  There are lots of things the Arduino can do that the RPi can't, and vice versa.

Think I'll look into the silica gel pod arrangement as the first job and sort out the DHT22 later - I think it could take some time.

[Gina]

New outer casing top design for 4" dome and window pod for silica gel.  I started to print this on my Concorde printer but the printer went wrong (again)

[Gina]

I think I may have found code to read the DHT22 in the RPi - HERE - I'm thinking of just the C code to read the DHT22 to incorporate into an INDI driver.  Whether this will work remains to be seen.  The code is experimental and in C rather than C++ so may require some tweaking to include in a driver.  Could be another option and save an Arduino.

[Gina]

I can see one problem with connecting the DHT22 directly to the RPi.  The DHT22 would be in the ASC on top of the mast and the RPi in the observatory with about 1.5m of cable.  I've read that the DHT22 is critical with timing and I think the cable length could be a problem.  I'm now thinking of having an Arduino Nano in the ASC to use the DHT22 and run a digital connection to the RPi.  Whether this is USB or something else I haven't decided.