All Sky Camera/Weather Station build

[angryowl]

After printing some prototypes this weekend, focus and aperture control as well as the lens cover mechanism are finished and working surprisingly well. 

Lens cover Printed

I spent more time than would have liked on the lens cover, trying to determine what the proper curvature of the cover would have to be in order to be retractable and at the same time not let the tiniest bit of stray light into the lens. This is not so important when it's protecting the camera from the Sun, but very important when taking a dark frame as even the tiniest amount of stray light would render the frame unusable, even during the night as tested. Despite the fact that it now looks awful it really does the job and lets nothing through. It will later be painted dark on the inside to keep reflections to a minimum.

Aperture gears

 

Focus gears

All gears were printed in 0.05 mm layer height as it really does produce a much nicer finish. They mesh properly with each other and I have no fear they would ever slip. 

The play in the gears is not that large, there's more in the aperture gears than there is in the focus ones. This is good as later on I'd like to try to build an Ascom focuser to work with the focus stepper to automatically and continually focus on a star during the night. So having little play here helps although would not be a deal-breaker.

One thing I noticed was that whilst testing, the steppers were getting REALLY hot, to the point I was afraid the PLA holders would start to melt. I am running them off a separate 12V 1A power supply but I don't think that's the culprit. Anyway this is only when they move so maybe it's not such a big problem as throughout a day/night the cover motor would only stay on for a couple of seconds, and so would the aperture and focus, well focus motor maybe a bit more use but not really concerned at the moment.

I've played with some anemometer designs, some of the electronics and also tested if the MLX90316 works through the arduino through SPI and it does! 

Depending on how I feel I'll either start on the anemometer or wind vane whilst printing the Stevenson Screen...

 

 

[Gina]

That's an amazingly small layer height, what size nozzle are you using?  I'm thinking of going down from 0.4mm maybe to 0.25mm for more accurate small parts.  Oh and what sort of filament?

[angryowl]

Hi Gina

I'm printing with the default 0.4 nozzle that came with the printer and the filament is the Hatchbox White PLA. Very good filament, never failed on me once, too bad I can't seem to find it for sale in the UK as I'm running out.

Yeah, 50 microns is quite darn small and I can barely see the printer laying the plastic when looking closely. A few times I thought the nozzle jammed as I just would not see any plastic flow! Downside to it though it shoots print times through the roof!

[Gina]

Thank you   I thought it was PLA   I've found rigid.ink in the UK do very good filament though I haven't tried their PLA.  I mainly use higher temperature materials.

[angryowl]

It just occurred to me that when assembled and mounted up the pole I would have no idea where all of the stepper motors would be in their travel. 

This is something that completely slipped my mind during design and I need to address this now, as imagine the thing being up and the focus gear is at the end of it's travel in one direction and I issue commands to the stepper to keep rotating the gear in that same direction. The motors are very powerful and I imagine the first thing to give in such a situation would be the three grub screws holding the lens to the camera 

I won't be redesigning the whole assembly, but instead I'll just fix two end switches for each of the focus, aperture and lens cover mechanisms at their end travels. These would then be hooked up into the arduino sketch to both let me know if I'm at one of the extremes and also stop me from issuing commands to turn that way.

[angryowl]

This the design I ended up with for the anemometer, the core of it will consist of two parts which will be boltable to each other.

There will be two bearings (636 6x22x7mm open type) at the top and at the bottom a washer for an 8 mm bolt. Both pieces are identical except one of them has a hole going through its case up to the level where the magnet will be mounted on the 6 mm shaft. The Hall Effect sensor and cable are inserted through this hole and should sit very close to the magnet once fully inserted. 

An 8 mm bolt will mount on the washer which in turn is secured by the two parts snapping together thus giving me a way to mount the anemometer. 

This is the casing that will go through the 6 mm bolt at the top and will shield the bearings and sensor from water and also provide three points where the cups will be inserted and secured via 3 mm bolts. It's got a thread on top to screw the final water barrier for the bolts inside.

The cups simply slide in the holes on the casing...

I think this design should protect the bearings, bolts, and sensor from the rain rather nicely and keep it spinning for years to come. Here's to hoping! 

I went with open type bearings due to their inherently reduced friction compared to rubber or metal seal types and if I find the grease in them is not allowing the anemometer to spin as freely as possible I could easily degrease them for less friction.

[Gina]

If these are ordinary steel bearings they'll rust from damp in the air.  I suggest PTFE plain bearings.  HTH

[angryowl]

13 hours ago, Gina said:

If these are ordinary steel bearings they'll rust from damp in the air.  I suggest PTFE plain bearings.  HTH

Hmm, interesting. They are grade 440c stainless and supposedly rust proof. Apparently 440c grade has greater corrosion resistance compared to simple stainless but only time will tell.

I only paid less than a tenner for four of them so worst comes to worst I can always replace them.

[Gina]

That should be alright then

[angryowl]

On 7/27/2017 at 19:07, CharlP6 said:

Ah okay, that makes sense. It would have been interesting to see why you would have done FEA, not that it's necessary.. 

On the CAD topic, you may want to give Autodesk's Fusion 360 a go. It is a quite user friend parametric mechanical design package and its quite easy to get a grasp of. A big plus is, as a personal user, you get full functionality for free, and the whole package is a small download, less than 100MB if I'm right.

Anyway, I like your ideas for controlling the lens and would like to see it take form over time.

Made the jump from 123d Design to Fusion 360 and now looking back I should have done this earlier as it's such better software with the added benefit that 123d files can be imported.

Thanks for the tip CharlP6 

[angryowl]

Haven't had much time to work on this lately but managed to print the end stop switch holders.

Also received an ASA sample from rigid.ink today and will be doing some tests to see how well and easy it prints. I think I'll be going with ASA for most parts rather than PLA as I've read mixed opinions to the suitability of PLA when used outdoors.

[angryowl]

I've found the ASA from rigid.ink to be relatively easy to print with and layer adhesion seems good resulting in nice sturdy prints. The surface finish is not nearly as nice and smooth as PLA but this I was expecting and I suppose it doesn't really matter in this case. 

Ordered a roll and just finished printing the Stevenson screen and the bracket for it. The sensors are inside and and the purple cable is Cat5 UTP.

Will update this when I'm done with the anemometer as I'm currently printing the cups for it...

[Gina]

Looks good

[angryowl]

Anemometer is now fully assembled 

Bearing and magnet mount. The two halves for this were printed in some glow in the dark PLA I had leftover, I want to save the ASA for outside-facing prints as the inside of the anemometer won't be seeing a lot of sun!

Hall effect sensor with resistor.

There's a gasket I had laying around which fits perfectly between the case and screw-in cap

Finished anemometer, the cap screws in nicely and should, with the help of the gasket, prevent any water from entering from the top side.

The way I plan on calibrating it is simple; an array of four high speed server fans directed at the cups should give me a nice constant reading on the Arduino, then in the same position and same distance from the fans place a cheap digital handheld anemometer and record the wind speed. Then it's a simple case of lowering the fan voltage and taking measurements of the wind speed for every reduction in voltage. I hope this will then give me a roughly linear correlation between rotations per second as recorded by the Arduino and the real wind speed in mph. All that's then needed to do is modify the Arduino code to do a conversion then output the value in mph on the serial in real time.

Now printing the wind vane and will update in due course...

[angryowl]

Due to unforeseen circumstances I won't be able to continue printing parts for this for the next month or so as I am away from home with intermittent internet access.

However, I hope I'll be able to design some of the remaining parts and will try and keep the thread going...

[angryowl]

Back and with a change of plans 

I will not be printing the main enclosure for the camera, instead I'll be using an IP66 waterproof enclosure which only cost me £35 delivered. The enclosure is ROSE 01.23 20 11 and is 232 width x 202 length x 111 height external dimensions. The height is a bit small but with a printed spacer before the acrylic dome it should be fine. It's all powder coated aluminium and has 4 mm thick walls and a nice soft nitrile gasket round the entire edge making it completely waterproof. A bit on the heavy side but I think my pole should handle it okay.

Reason for this change was mainly that I thought an enclosure printed from ASA would not bee sturdy enough to hold the heavy camera and all other electronics inside and then there's the issue of making it completely watertight. Also mounting the main enclosure will be done with two M10 bolts and these will grab the 4 mm aluminium bottom of the enclosure much better than any printed solution would have worked.

The white plastic bottom part goes into the end of the pole and the enclosure will be drilled for the two M10 rods 

Printed the casing and fin for the wind vane 

The anemometer and wind vane will be mounted underneath the main enclosure whilst the rain gauge and Stevenson screen will be clamped to the main pole further down to make the whole station more compact

 

 

 

[Gina]

This is a very impressive project   Very interesting to see you are mounting your ASC above the wind instruments of the weather station.  I have considered that and I guess it's an option.  Having done a lot of research into the best places and heights for the various parts of a weather station, I have my wind instruments about 10m above ground on an unsupported tubular mast.   Currently strapped to a very solid fence post but it sometimes interferes with my astro imaging and wants moving.  I have a large shed I could mount it on but I think that position would suffer from more wind disturbance.  OTOH I rather wonder how much this matters as everything is very approximate anyway.

Your camera looks a lot more expensive than my ZWO ASI178MM and I see it's a colour version.  I would like a colour camera for daytime ASC and thinking of that as a separate project.  My latest ASC is All Sky Camera Mark 7

[angryowl]

A while ago I saw this http://www.weatherstations.co.uk/gooddata.htm detailing the international height standards for different types of sensors. 1250mm for temperature seems doable as I can move the Stevenson screen down the pipe. Same goes for the humidity sensor and the pressure one doesn't care much where it sits. As for the rest they'll just have to do sitting lower to the ground.

My main pipe is 3500mm but could be extended to 5500mm with another one slightly larger in diameter so would be a perfect fit. I still haven't decided on the height yet as I have some low branches in the way so would like to go as high as possible to avoid them but at the same time I have to consider the final total weight of the head with everything mounted on it. The pipe will be strapped to the face of my extended shed.

I am aware the weather station will be eating away from my already tiny unobstructed sky but you can't have it both ways 

Gina - 10 meters above ground level for wind is apparently the official sensor height for speed and direction. And yes colour ASC cameras, despite the decreased sensitivity compared to mono, are great for daytime as they offer superb time lapses of cloud formations.

[angryowl]

Drilled the dome and enclosure lid for the 6mm bolts used for mounting, with a 3D printed template of course

This is the printed cutout to help me with alignment when cutting out the aluminium enclosure lid. I want to cut as little as possible as the heater will sit in the dome and I don't want much of the heat escaping into the main enclosure and heating up the electronics, instead it should stay in the dome heating it thus preventing formation of dew.

Guess what I'll be doing tomorrow!

[angryowl]

Hmmm, looking at your Mark 7 build Gina I can't stop but thinking whether I made the right choice by going with acrylic.  I'm now almost certain mine will yellow as I don't think it's for outdoor use. Yours yellowed a lot in 9 months and it was outdoor grade! The dome cost me £15 so replacing twice every year would work. Is the other solution you're proposing using a large cover over the camera and other exposed components and removing it when the camera is in use? What about rain? You'd need to ensure the cover is over the camera even when it's in use and it starts raining. The other thing would be dewing of the lens as you wouldn't be able to heat the air around it, but a mini dew strip could potentially be used. I'm intrigued by this idea and if changing the dome twice a year or even more frequently proves to be impractical then this could be the solution.

[Gina]

I'm planning to heat the lens to keep the dew off.  A rain shower coming in when imaging could be a problem but maybe a rain detector to close the cover would work.  Probably get a few raindrops on the lens though.  Whether these would leave marks once dry I don't know.  Raindrops on the dome seemed to dry off without leaving a mark.  I'll admit, I would be happier using a dome and may still consider it.  I was disgusted at the way that dome yellowed and became unusable in such a short time, particularly as it was some £30 odd, so far from cheap!  They are no longer available so maybe the firm selling them found they were rubbish and discontinued them.  OTOH there are ASCs around with domes that would seem to survive.  This was described as a replacement dome for a very expensive commercial ASC.

As you say, maybe a different source of cheaper domes and change every six months could be an option.  Carry on with yours as you're going and let's see how it goes.  I shall be very interested   I too will continue with mine and likewise see how it goes (unless I change my mind - which has been known! ).

BTW the lens will be sealed to the casing so water shouldn't get into the electronics.

[angryowl]

13 minutes ago, Gina said:

They are no longer available so maybe the firm selling them found they were rubbish and discontinued them

Hehe, yeah and the fact that it was a replacement for a commercial ASC makes you wonder even more. 

I suppose only time will tell how long it takes for my dome to degrade to the point it affects the image enough that it's unusable and needs replacing. I'll be following your Mark 7 build closely and will be very interesting to see how it turn out.

[Gina]

I see what you mean - replacements needed because the domes soon "wore out"!

One possibility would be to arrange an opaque outer cover for the dome (motorised) and cover during the daytime.  Would mean the ASC couldn't do daytime imaging but if only wanted for night sky views it would work.

[Gina]

I've ordered a cheap 4" dome from ebay due to arrive in a week or so to look at :-
4'' Indoor/Outdoor CCTV Clear Acrylic Monitoring Camera Dome Cover Replacement

[Gina]

Amazon are selling 4" domes with a specification which looks pretty good but coming from China.  £11.72 including delivery to UK.
JMX 4 Inch Acrylic Indoor / Outdoor Cctv Replacement Clear Camera Dome Housing

Quote

Product description

Transparency: ≥94%
Vandal-proof, ultraviolet resistance and aging resistance
Distortion factor: ≤0.16%
Material: Acrylic (Made by ARKEMA and Mitsubishi Chemical)
Through ISO14000, SGS and RoHS Certification

Optically tested for true high definition camera use. 1080P and 10 Mega pixel compatible. Injection molded for consistent optical and dimensional performance.

I'm wondering whether to order one of these.  Pity it's slow boat from China