Gina's DIY All Sky Cam - Complete Redesign

[Gina]

I've not tried a wireless dongle, no.  I don't think I've even got one to try.

James

Think I've got one out in the observatory warm room on a desktop PC.  I've also got an original Raspberry Pi somewhere that I never got round to trying.  Whether this would be any use considering all the better models developed since I don't know.

[Gina]

If it's kept out of the damp then I see no reason it wouldn't work.  I suspect the fan motor generates sufficient warmth to drive off any moisture that might find its way into the workings in the air.  Some sort of mesh to keep the larger creepy-crawlies out might be a plan though.

James

The warmth generated by the fan motor is a good point - that might well save it from damage.  Anyway, I'm going to try it.  I have some insect proof mesh somewhere too - I'll have a look for it

[Cath]

You don't need to Gina, it's OK, it really is

[Gina]

I now have the green heatsink and fan shown above attached to the hot side of the TEC with a piece of copper bar, 35mm wide x 3mm thick and now testing.  Without the thermometer chips I can't say exactly what the temperatures are but the hot copper bar is not running too hot - I can put my finger on it and hold it there for several seconds.  Above the cooler it's just warm.  This cooler seems to be working well

[Gina]

Here's some photos.

[Gina]

Cooling seems to be working fine - here's a dark with my standard night image parameters.

Later with cooling off.

[Gina]

Working on the new box.  Cut hole for dome and printed new resized dome tube and the flanged part that connects the dome tube to the base.  Using two parts allows for possible change in dome height if required later.  The flanged tube is solvent welded to the box using acetone.

Here are some pics.

Flange ring attached to box/

Dome tube added.

Dome added.

Inside view.

[Stu]

Looking good Gina

[Gina]

Thank you   I appreciate that

Been working on the bottom (normally the top of the box).  Hole positions worked out and drilled plus hole for fan and heatsink.  That's the air intake done but still have the exhaust ports to cut.

[Gina]

Photo showing camera assembly mounted on box lid.  Which will be the bottom of the outer casing of the all sky camera.

[Gina]

The DS18B20 digital thermometer chips arrived today and I've started adding then to the cooling system.  This pic shows one mounted next to the QHY5 image sensor, thermally coupled to the PCB and edge of the sensor with (gold) thermal paste and fastened in place with hot melt glue.  Should at least give me some idea of the sensor temperature.

[Gina]

I've made some fairly large holes in the bottom to vent the hot air.  These will be covered with mesh before I install the unit outside, to keep creepy crawlies out.  I have printed a conical air intake and will add mesh to that too.  The wider bottom of the cone will compensate for the flow restriction of the mesh, I'm hoping.  

I've also added a thermometer chip to the TEC hot side against the copper bar, inside one of the clamp pieces.  I plan to add another just below the dome.  I decided to try the warm air from the cooler to provide dew heating rather than using a copper radiator in the dome tube (easier ).  The thermometer in the dome tube should tell me how well this works - as does a lack of dew on the dome, of course.

[Horwig]

what can I say, WOW

Huw

[Gina]

Thank you Huw

I have been sorting out the Arduino sketch code for the 1-wire thermometers.  Or rather a quick search found the stuff I'd done in the past on this for my weather station and camera cooling.

With multiple sensors the sketch needs to know the individual IDs so the first job is to read these from the devices.  This tutorial tells you how   Then include the ID codes as shown in the sample sketch below.  The codes shown were from one of my cooling projects - I shall be reading the ID codes of my new devices and substituting them in the sketch (lines 22-24).

// This Arduino sketch reads DS18B20 "1-Wire" digital// temperature sensors.// Tutorial:// http://www.hacktronics.com/Tutorials/arduino-1-wire-tutorial.html#include <OneWire.h>#include <DallasTemperature.h>// Data wire is plugged into pin 8 on the Arduino#define ONE_WIRE_BUS 8// Setup a oneWire instance to communicate with any OneWire devicesOneWire oneWire(ONE_WIRE_BUS);// Pass our oneWire reference to Dallas Temperature. DallasTemperature sensors(&oneWire);// Assign the addresses of your 1-Wire temp sensors.// See the tutorial on how to obtain these addresses:// http://www.hacktronics.com/Tutorials/arduino-1-wire-address-finder.htmlDeviceAddress heatsinkThermometer = { 0x28, 0x8C, 0xA4, 0xE0, 0x02, 0x00, 0x00, 0xA3 };DeviceAddress ambientThermometer = { 0x28, 0x96, 0x59, 0xC4, 0x03, 0x00, 0x00, 0xA8 };DeviceAddress coldFingerThermometer = { 0x28, 0xFA, 0x43, 0xC4, 0x03, 0x00, 0x00, 0xD7 };void setup(void){  // start serial port  Serial.begin(9600);  // Start up the library  sensors.begin();  // set the resolution to 10 bit (good enough?)  sensors.setResolution(heatsinkThermometer, 10);  sensors.setResolution(ambientThermometer, 10);  sensors.setResolution(coldFingerThermometer, 10);}void printTemperature(DeviceAddress deviceAddress){  float tempC = sensors.getTempC(deviceAddress);  if (tempC == -127.00) {    Serial.print("Error getting temperature");  } else {    Serial.print(tempC);    Serial.print(" C = ");    Serial.print(DallasTemperature::toFahrenheit(tempC));    Serial.print(" F");  }}void loop(void){   delay(10000);  Serial.print("Getting temperatures...\n\r");  sensors.requestTemperatures();  Serial.print("ColdFinger temperature = ");  printTemperature(coldFingerThermometer);  Serial.print("\n\r");    Serial.print("Heatsink temperature is: ");  printTemperature(heatsinkThermometer);  Serial.print("\n\r");  Serial.print("Ambient temperature is:  ");  printTemperature(ambientThermometer);  Serial.print("\n\r\n\r");}

[aruangra]

How do you seal the chamber to prevent condensation?

[Gina]

How do you seal the chamber to prevent condensation?

Holes where wires go in are sealed with hot melt glue (also holdes the wires in place) other places will be sealed with silicone grease if needed.

[Gina]

Been trying to open the 1-wire ID reading sketch which has .PDE extension and is not compatible with the later Arduino IDE versions.  The normal way round this is to remame the .PDE to .INO but Win 7 Home Premium won't seem to let me   I guess I'll have to transfer it to one of my Linux boxes to change the extension.  How ridiculous is that!!??    But I guess I shouldn't be surprised with a Microsoft OS

[Gina]

Should have thought of that rather than struggling to get Win 7 to do the job   Anyway, started up my Dell laptop running Linux Mint transferred file with TeamViewer, renamed it and transferred it back into my Win Arduino folder.  Job done

Here's the sketch :-

// This sketch looks for 1-wire devices and// prints their addresses (serial number) to// the UART, in a format that is useful in Arduino sketches// Tutorial: // http://www.hacktronics.com/Tutorials/arduino-1-wire-address-finder.html#include <OneWire.h>OneWire  ds(3);  // Connect your 1-wire device to pin 3void setup(void) {  Serial.begin(9600);  discoverOneWireDevices();}void discoverOneWireDevices(void) {  byte i;  byte present = 0;  byte data[12];  byte addr[8];    Serial.print("Looking for 1-Wire devices...\n\r");  while(ds.search(addr)) {    Serial.print("\n\rFound \'1-Wire\' device with address:\n\r");    for( i = 0; i < 8; i++) {      Serial.print("0x");      if (addr[i] < 16) {        Serial.print('0');      }      Serial.print(addr[i], HEX);      if (i < 7) {        Serial.print(", ");      }    }    if ( OneWire::crc8( addr, 7) != addr[7]) {        Serial.print("CRC is not valid!\n");        return;    }  }  Serial.print("\n\r\n\rThat's it.\r\n");  ds.reset_search();  return;}void loop(void) {  // nothing to see here}

[Gina]

Results :-

Image Sensor.

Hot side of Peltier TEC.

Dome.

[Gina]

Unfortunately, the codes won't Copy from the Serial Display so I have to copy from the screenshots...

Image sensor - 28 FF 6A 3F 78 04 00 C7

Hot side - - - -   28 FF BA 2E 77 04 00 71

Dome - - - - - -  28 FF F0 3E 78 04 00 42

[Gina]

Had the sketch working on a Uno for testing and reading the three temperatures though the dome thermometer is not inside the casing and is showing ambient temperature.  I connected up the cooling and watched the readings.  All looked fine at first - the hot side temperature increased then after a while the image sensor showed cooling.  However, after several minutes the hot side reached 55C and the cold had dropped 3.5C then the hot side started decreasing and after a while the cold side started getting warmer.  Then I checked the supply current and it had dropped to 50mA    Seems the Peltier TEC has failed

DRAT and DOUBLE DRAT!!! :mad:

Here's part of the cooling run with readings every 10s.  I've changed the sketch to reduce the output.  Columns are image sensor, hot side and dome/ambient.

[Gina]

Been searching for a replacement Peltier TEC without any success whatever   ebay have higher power ones from China but nothing in the UK.  All the ones on Amazon UK seem to come from China or suchlike too

I had a break from this project and did some watercolour painting yesterday maybe I should do that again - for a lot longer!!!

Anyone any other suggestions, please?

[Gina]

Farnell have them for £23.61 plus VAT = £28.33.  I think this may have been where I got the one I have - I remember buying a Peltier TEC from them as I couldn't find what I wanted anywhere else. 

Searching for the manufacturers part number as supplied by Farnell I've found a couple of other sources, one at £33 and also on ebay a bit cheaper but an extra day or two delivery time.  Here - £26.55

[JamesF]

Do you have any idea why it might have failed?

James

[Gina]

Unfortunately, no, I haven't   I guess it's possible that it might have got too hot in the past - it certainly didn't get too hot on this last run.  The only other thing I can think of is that the clamp squeezed it too hard though I think that's unlikely.  And it isn't that a wire has come off as it reads a bit over 1K rather than 3-4 ohms.