Home arrow Tutorials arrow Hardware arrow Weatherproof Wireless Enclosure Build Tutorial
Weatherproof Wireless Enclosure Build Tutorial | Print |
Written by Akiba   
Tuesday, 10 August 2010

Hi all.

No, I haven't been dead these past few weeks. I've actually been spending a lot of my free time researching how to build a weatherproof enclosure for the rice farm sensing project in the Tokyo Hackerspace. I've mentioned it previously and the project takes place in Kamogawa which is about an hour and a half outside of Tokyo. The place is a complete contrast to Tokyo and has problems with monkeys, boars, poisonous snakes, and some nasty bugs. Overall, quite interesting, although I am a bit scared of poisonous snakes.

Anyways, we'll be going out there this weekend to check out the terrain and I thought it might be good to bring out a prototype of what I have in mind for the outdoor sensor network project. There's not enough time to build a finished device, but I figured that I could at least put together a mock up that's very close to the real thing. It would at least give people in the hackerspace and others involved an idea of what I mean when I'm talking about agricultural sensor networks. 

The real difficulty in keeping anything outdoors for extended periods of time is how to ruggedize it to withstand the elements. Nature has a way of decomposing anything you put in front of her, and since there's going to be issues with wild animals, it would have to withstand being hit, bitten, nibbled on, and trampled as well. 

There were a lot of decisions that had to be made including enclosure type, form factor, I/O, platform, antenna, etc. The list goes on and on, which is why at some point, I needed to stop thinking and just put together a prototype. I figure a lot of the major issues will come out once something physical exists. This build tutorial is mostly a record of the steps to make the first draft of the Tokyo Hackerspace wireless farm sensors, ie: Hatake-Net (hatake=farm in Japanese).

One of the major decisions I made is to make the system based on the Arduino . I've noticed that it's very hard to get people in Tokyo Hackerspace to participate in the software development and I think a lot of that is because the standard embedded toolchain and programming environments are intimidating for people that don't do it regularly. On the other hand, there's such a wide base of knowledge for the Arduino and many tutorials aimed at the beginner level. It's really impressive how much thought was put into the Arduino platform and environment for usability. Hopefully after a few Arduino workshops in the hackerspace, we can build up a good base of people willing to help out on the software development. One of the goals of the hatake-net project is to have everyone participate in creating the sensor network. I'm hoping that people will realize that the simple act of reading a sensor can dramatically impact a farmer's life. Of course, I added a slight twist on the Arduino by adding native wireless communication to it...heh heh heh...

Anyways, I want to mention that this build is just for a mock-up but demonstrates the approach I'm currently taking for designing a ruggedized, weatherproof, and wireless enclosure. The main base enclosure is an industrial grade, IP-65 box. In this case, IP stands for "Ingress Protection" and is an IEC standard that classifies the degree of protection provided by an enclosure. IP-65 specifies that the enclosure is dust-resistant and water-resistant (but not submersible). It's also equivalent to a NEMA rating of 4 for those familiar with the NEMA enclosure code.

Also, this build is just a first draft and the final device will probably be much different based on what we observe in real usage. If you have any comments or questions on the design, please feel free to post a comment. I'd love to hear any feedback and answer any questions.

Well, here ya go. Enjoy! 

Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!
Click image to open!

Hits: 35911
Trackback(0)
Comments (9)Add Comment
Considerable work DONE--Fantastic
written by VinodGanesh, August 13, 2010
Step By Step process for weatherproofing. I appreciate your patience for the whole guide. Its fantastic for noobs to understand the process well. :-)
report abuse
vote down
vote up
Votes: +0
...
written by PhangMoh, August 13, 2010
Very neat piece of work!!! We used similar approach but we used a removable gland because sometimes the sensors are buried deep in the ground. You might want to seal the antenna hole at the top but again maybe there's not much rain over there compared to where I came from.
report abuse
vote down
vote up
Votes: +0
...
written by Akiba, August 13, 2010
Thanks. Actually, I'm doing a special order for an antenna bulkhead with rubber O-ring. I'm also going to add rubber gaskets for the antenna connector.

I'm working on getting it down to a smaller sized enclosure. The one in the picture uses my prototype PCB but the final PCB will have a bottom mount battery connector underneath the PCB which will save a lot of space. I'm also going to install a filtered vent which maintains water resistance but can allow ambient temperature and pressure monitoring.

Should be a V2 coming up soon hopefully.
report abuse
vote down
vote up
Votes: +3
...
written by Ignacio Cobos, August 19, 2010
Nice job:
Just some tips.
One enclose also (inside enclosure) batteries. At least to avoid possible drops (if any) from anntena hole.
Think in dew: If you expect really a long exposure, finnaly humidity will be inside and there is always a risk of dew. At least put some dissecant bag (reuse one that are in somoe package products). An try to imagine were gravity will accumulate dew in your final distribution.
Think in hot:
It´s is very difficult a trade-off between water proof and heat dissipation. I don´t know how many heat is generated in your design. Try to imagine the heat dissipation if the box is exposed in a sunny day.
It seems that some analog electronics should be put in place in you design and heat could decalibrate it.

And finnaly the more paranoiac issue. Select carefully the cover of your wires as far as i know some animals like very much to eat plastic. Maybe rice farmers of the final location might give you some tips.
Please
report abuse
vote down
vote up
Votes: +0
...
written by Akiba, August 19, 2010
Thanks for the excellent tips. I especially like the dessicant one. For the enclosure, I'm going to be using rubber gaskets and O-rings on the antenna hole since the whole point of the enclosure is not to allow water to get in. My goal is not to breach the water resistance of the enclosure. I'm also going to be doing a second version that will be vented and will probably end up as the one that will get deployed. I'm looking at an industrial grade filtered vent and the filter is made of Gore-Tex. Its breathable which allows air to pass but moisture can't get through it. I'll be testing it out soon.

For the wiring, I'm not sure how to prevent animals from chomping on them, but when we deploy, we should be visiting the area roughly once a month. We'll be able to see what the animal preferences are and if they abuse the boxes in any way. That does give me an idea though that we should be monitoring the integrity of the lines though.

Me and the rest of the Tokyo Hackerspace will also be taking a lot of data on the enclosures including sensor readings comparing the inside conditions (temp, humidity, pressure) to the outside conditions. We'll be publishing all of this since there seems to be some interest in how to ruggedize devices for outdoor conditions.

Thanks again for the comments. I'm hoping that this is just the start of putting out ruggedized enclosure techniques to the open source community smilies/smiley.gif
report abuse
vote down
vote up
Votes: +2
...
written by Ebara Pumps, August 25, 2010
very well defined every steps. it can be a easy guide for a novice like me. smilies/smiley.gif
report abuse
vote down
vote up
Votes: +0
More info about Hatake-Net?
written by aevin, August 29, 2010
Nice article showing off your prototype. I like that you've included all the pics.

But I'm curious to know more about what type of sensing will these devices do? I.e. what parameters will they measure, and how will they help the farmers? I guess you plan to both have ~real-time access to sensor data for instant monitoring, and the ability to collect logs.

How many sensors do you plan to deploy? How large rice acre area will one sensor device cover? And do you plan to use a mesh network topology (decided on any particular protocol yet?)?

I guess some of my questions may not have a clear answer yet, as you will have to try and see, and design as you go. But I assume you already have an idea on the capacities and features for the final Hatake-net.

Btw, I like the idea that your hackerspace in Tokyo will apply you "magic" tech and skills to help out farmers on the country-side. It sounds like an interesting project to follow, and I bet other hackers & farmers can learn from your experiences as you go along.
report abuse
vote down
vote up
Votes: +2
...
written by Akiba, August 29, 2010
There are a couple different applications that we have in mind. One of them is sensing water level in the rice paddies. The farmers need to check the level every day and the paddies are on steep, stepped hillside terraces. So if we can implement float sensors and transmit the data wirelessly, then it would make their lives a lot easier.

The bigger issue currently is the monkeys, boars, and deer. Japan did a huge government subsidized re-planting of the forest land and put a lot of cedar trees in. They covered whole forests with cedar and now the trees are big and crowding out other foliage. That foliage produced food that many of the forest animals lived on so now they're starving and raiding the farmer's crops. We're talking with some of the guys out there on how to stop the animals from foraging but still keep them fed. I'll keep everyone posted on that, but we're hoping to do some animal population counting which will help the effort.

The project has just started so we're still getting an idea of what they really want/need out there. We actually just took a trip out and posted the pictures on the Tokyo Hackerspace website. You can check it out here:

Link
report abuse
vote down
vote up
Votes: +0
where to buy?
written by t3db0t, May 06, 2011
Howdy! Thanks for the excellent writeup. I am looking for an enclosure just like this, what is the brand, or where did you get it? I'm having a hard time finding a similar one with mounting holes...
report abuse
vote down
vote up
Votes: +1

Write comment

busy
  No Comments.

Discuss...
Next >