My solar powered  PiCamera installation took a step closer yesterday with the link options resolved.


Link

The WiFi link for transferring the data requires a directional antenna.

The Antenna design needs to be rugged to withstand the hostile environment, as well as the distances.

 

The links are 350-500m

(click on image for full size)

Link paths.jpg

Google have some neat features that allow you to view the profile of the line. (Save the line, and then right click on it)

Profile of path.jpg

While the height reference data is not 100% it shows that the rise not far from the second camera shouldn't pose too many issues.

The real test will be when its covered in snow/ice as that may change.

 

Since we are only sending the odd image and a bit of data over the link, and its only 100Mbit Ethernet on the Pi, I'm certain bandwidth won't be an issue.

 

External Antenna

I did some searching to see what DIY options were available.

There are a few, and for my application, the mounting and weatherproofing was a cause for concern.

Making the box larger to house the antenna, means greater surface area and less heat to keep the device warm.

 

I had a conversation with our Dale at our local Wifi supplier (www.gowifi.co.nz)

 

I had picked an external antenna that might have worked, but he suggested one of these.

http://www.gowifi.co.nz/backhaul-point-to-point/ubiquiti-nanostation-locom2-802.11n/g-200mw-outdoor-ap/bridge.html

 

They have fitted the larger version at some NZ skifields, and while designed for Outdoors, extra protection in the form of PVC drainpipe and a cap solves the snow/ice issues.

Unlike some Access Points, they run from 12v, and you can control the transmit power to reduce the current consumption.

 

This means there is no USB WiFi required, and a ModelB RPi is necessary, but 12v power is available and therefore the overall design remains the same.

 

Images

When the idea was first suggested, the idea was put a camera looking down the lift line to view the skiers coming up the tow.

A second camera would look back at the lodge (see below).

 

This  is roughly the view that is available, and was taken from the back of a  groomer, so is higher and wider than the Pi will deliver.

(click on image for full size)

Return_Road_view.jpg

 


As we refined the concept and resolved that it was possible, the option to fit the first camera looking back at the lodge had more merit.

These included :-

  • Much easier access
  • Groomer has to pass daily so the panel could be checked/cleared
  • Shorter link path
  • Not subject to falling ice
  • Easier mounting
  • Doesn't require a lens to be fitted to PiCamera
  • Still visible to Cloudy Lift line for future data linking.

 

 

 

Housing

I have a couple of ideas to persue for the housing.

Limiting the number of external holes is a high priority, so this limits the range, but thankfully the electrical industry has a range suitable.

 

Various IP Ratings are available, and a piece of clear acrylic will allow the camera to see if a clear cover isn't available.

 

 

Planning

I hope that viewers get the idea that careful and detailed planning can save a lot of 'issues' further down the track.

These 'issues' can result in the final product NOT delivering what was intended, or that the customer expected xyz, and got abc, due to the two parties having different ideas.

Lack or poor planning is usually the cause of cost or time over runs, and a failure to deliver 'on time and on budget'.

 

I once heard a saying that "unless you have a plan, you can't change it".

It also makes it harder to factor any changes if you strike an problem that requires a different approach.

(If I had have purchased a ModelA then I couldn't use the AP above)

 

There is NO excuse for poor planning on ANY project.

 

Mark

 

 

This was an interesting look into a least one project that the spin doctors made out happened overnight.

Space Dive