Gina's DIY All Sky Cam - Complete Redesign

[JamesF]

Probably because you're doing it as your own user rather than root?

James

[Gina]

Yes, seems likely - I though that.  I've been trying to copy it in the GUI maybe I need to use the command line.

[Gina]

Been looking at dew heating again as the dome needs more heating.  I have the following :-

1. 5 x 22 ohm half watt resistors - probably not enough power
2. 2 x 33 ohm 10W resistors - 48mm long x 10mm square - BIG!
3. Nichrome wire @ 2 ohms per foot - a metre would give about 4W on 5v
4. Assorted resistors - some half watt.  Can't be bothered

I think no.3 wins with about 4 or 5 turns inside the dome if run off 5v.  But I will have to do a test to make sure it doesn't get too hot and melt its way out

Does anyone have any idea of how much power would be wanted for this application?  I can only guess.

[Horwig]

Well mine is ok, and I'm running it on the 68 Ohm resistor you recommended on 12v.

Huw

[Gina]

Thank you Huw   One 68 ohm resistor on 12v gives just over 2W. 

I've cut off a metre of nichrome wire and that actually measures 7.4 ohms.  Running off my bench supply it's taking 0.7A at 5v and with 4 turns it's getting quite warm but not too hot to touch.  So that's 3.5W.  Having more than the required heating for dew control would mean I could dry off raindrops.  I can switch this on and off using a power MOSFET running off the Arduino.   I could alter the on/off ratio to maintain a dome temperature a few degrees above ambient.  I think a measurement once a second would be adequate or maybe even less often.

Ideally I guess I could do with measuring ambient temperature and setting the inside dome temperature a couple of degrees above dewpoint to stop the outside dewing up.  I don't think there is any misting up inside the dome.  I might add a DHT22 to measure ambient temperature and humidity in place of the DS18B20 thermometer chip.  In fact I could combine use in the all sky camera with providing outdoor ambient temperature and humidity for my weather station saving a separate Stevenson screen.  I think the indoor temperature and humidity will be that of my observatory scope room.  I could easily add the warm room too if I wanted

[Gina]

Here are some photos of my new dew heater ring.  Nichrome wire in white heatshrink sleeving shrunk onto the wire by running a coil with 13.8v to get a high enough temperature to shrink the sleeving.  Then the sleeved wire was threaded through half a dozen short lengths of black heatshrink sleeving and arranged to be slightly smaller than the dome support tube.  Once formed to shape, heat was applied to the black sleeving to hold everything in place.

I now have the assembly running off 5v continuously to check the highest temperature reached while sitting on the table with minimal cooling.  By touch I would estimate a temperature of around 70-80C - safe for both the heatshrink sleeving and ABS plastic I use for 3D printing.

Sleeved nichrome wire threaded through 6 larger heatshrink sleeves.

Dew heater ring placed in a dome tube.

With ring fixed in shape with heat shrunk short sleeves (6) on test sitting on the table running off 5v and 3.5W input power.

[Gina]

A sky full of contrails

[Starlight 1]

Gina see m8 today , he give up on all the rain things do not work . he use silicone polish  but it you put just one finger on it you have to start all over again.

[Gina]

Thank you for your reply - I was rather afraid of that.

[Gina]

Clear sky ATM   Not dark enough yet to show many stars.

[Gina]

Getting dark now

[Gina]

The dew has started showing on the dome   I plan to add the new dew heater tomorrow.

Cleaned the dew off the dome with a tissue.  All on the outside as I expected.

[Gina]

Project progressing...  I disconnected it, removed from obsy roof and brought the unit indoors yesterday for mods.  I found a dead wasp and several drops of water in the bottom.  I think the wasp got in through the vent I was using to allow the warm air I was using for dew heating to escape.  I forgot to cover it with bug proofing mesh.  Rain must have got in through the holes covered with mesh so I think I need to see if I can stop it.

With the warm air dew heating not working I decided to go for the standard method as mentioned above of a ring of nichrome wire with on/off power control.  In wet weather the cooling fan would draw in damp air and distribute it all round the inside including the dome, lens and electronics.  That may have been how water droplets got in.  It certainly doesn't strike me as such a good idea.  The upshot is that I have abandoned fan blown warm air dew heating and made a new partition with draught proofing to keep the warm damp/wet air away from the electronics, lens and dome.

I will be adding a DHT22 digital humidity and temperature sensor on the outside of the bottom of the box, shielded from rain rather than in the air intake cone as originally thought.  The data sheet for the DHT22 says it shouldn't get wet.  This will enable me to use the dew point calculation in the Arduino sketch and provide dome temperature control to maintain it a couple of degrees above dew point temperature.

One dead wasp.

The dome air vent.  No bug mesh.

Water droplets in bottom of unit.

[Gina]

New partition and dome extension ring plus dew heater.

[Gina]

Here is the latest circuit diagram.  This All Sky Camera seems almost to have turned into half a weather station   I'm now adding the light level sensor from my weather station as sensing the light level could be arranged to reduce the aperture as it gets light in the morning.  No point in duplicating things so some of the data from the camera unit will be included in my weather station.  The 4 cell photo-voltaic battery gives an almost linear voltage versus light level and the output varies little with incident light angle.  (Assuming it's still working after several years ).

[Kropster]

The vent holes look quite large. Is it possible for wind blown rain to get in?

You might need to put some baffles in place outside the box to stop wind blowing directly on the holes, maybe.

Great project by the way. Is your daytime job with NASA? 

[Gina]

No, I don't work for NASA I'm afraid - that would have been very interesting   I'm retired and only have myself, one cat and three goats to look after.

I think it would be quite likely for wind blown rain to have got in through the large holes and I am planning to add some protection.

[Gina]

Here are some photos of the 3D printed rain shields.  They are attached to the box cover with hot melt glue which not only holds them in place but seals the edges to the box.

[Gina]

Working on this in parallel with the weather station...  I think I have now connected up the new components and ready to put it all back together.  I'll recheck all the connections first.  I've added a DHT22 digital humidity and temperature sensor glued to the underside of the casing where it won't get wet.  Wired into the Arduino - 0v, 5v and signal on pin D10 plus pull-up resistor.  The other part is the new dew heater.  Connected to the incoming +5v wire and a power MOSFET IRFZ44N connected to the strip board and wired to 0v, heater and pin D12 on the Arduino Nano.  1-wire is on pin D13.  D2-D5 focus, D6 TEC cooling power MOSFET. 

I'll update the circuit diagram with the pin numbers shortly and take some more photos.

[Gina]

Circuit diagram.

[Gina]

Photos.

Top of box showing added photo-voltaic battery (4 cells)

Circuit board and wiring (could be tidier )  The lens control mechanisms can also be seen.  Aperture with micro servo on the right and focus with stepper motor on the left.

Bottom of unit showing the DHT22.

The unit assembled except for the dome.

Showing the dew heater ring just below the dome position, supported on two stiff wires.

[Gina]

Now testing.  I have the 5v input running off my bench PSU and testing the dome heating.  Strange thing is the current being drawn is only about half what it should be   Current is reading 0.33A rather than 0.7A.  The dome heater is working and the dome is warming up but not as fast as it should do if the power was getting tom the heater coil.

[Gina]

Taken the top off to get to the electronics for fault-finding.  The volts are getting lost in the power MOSFET but this isn't getting hot, in fact it'ts not even warm!  Checked the gate voltage and it's just over 4v so should be switched on (logic threshold is 2v).  It's as if I'm using PWM at near equal mark/space ratio but then the input would show an average of around 2 or 2.5v (half the logic 1 level of the Arduino).  Anyway I'm NOT using PWM - just outputting logic 1.  This is just MAGIC!!!

Later...

Checked the cooling MOSFET and current drawn and it's doing the same as the dew heater one.  Dropping half the voltage and the current is 1.8A rather than 3A.  Clearly I've done something wrong with the latest mods

I was hoping for a night run tonight as the sky is pretty clear and forecast to be clear until well after midnight when mist and fog is due to arrive.  I guess I'll "knock it on the head for tonight" and see what i can find tomorrow.

[pete_l]

 It's as if I'm using PWM at near equal mark/space ratio but then the input would show an average of around 2 or 2.5v (half the logic 1 level of the Arduino).  Anyway I'm NOT using PWM - just outputting logic 1.  This is just MAGIC!!!

You haven't got a parasitic/self oscillation a the MOSFET gate, have you?

[Gina]

I think that's quite a possibility - thought of that myself.  As I've added extra bits the circuit board has got quite crowded.  I'm thinking maybe I should rebuild it taking account of earth loops and adding extra decoupling.