Introducing a new, modular, open-source, open-hardware wireless sensor-actuator network toolkit for environmental and agricultural projects. The main idea is: may people in environmental and agricultural fields have need for (or could greatly benefit from) deploying sensor-actuator systems in their practice. However, the exact system of sensors for each individual is idiosyncratic to their unique environment, constraints, and line of inquiry, requiring customs systems be designed from scratch for each new project. This results in a barrier to entry for those who don’t have the background or time to learn and develop their own sensor-actuator systems. OPEnS WSAN proposes a plug and play, modular, reconfigurable architecture that would enable those with little-to-no expertise in engineering to quickly map out and assemble a unique system to fit their needs in relatively little time.
A single hub will collect all transmitted data from nodes and translate data streams into WiFi for uploading onto a cloud storage service.
Will be comprised of parents and children. There can be up to 6 unique parents directly connected to the hub. Children placed further away can communicate to the hub by passing messages along to their parent.
These nodes will receive commands from the hub to drive relays, motors, servos, valves, and other commonly used actuators. Messages may be sent in real time by the user sending commands to the hub, or automated through a cloud service like IFTTT (If This Then That). In this manner, hub may be programmed to alter the behavior of sensors or trigger events of actuators based on the behavior or events from other nodes.
These simply extend the range of communication of nodes to the hub by relaying messages to their target destination.
I’m using the RF24Network library by maniacbug, which can be accessed on GitHub, or through the Arduino IDE Library manager. I’m also using the i2cdev library by jrowberg to manage communication with the MPU6050 IMU sensor (see below).
Progress to Date
I’ve begun preliminary work on evaluating the Nordic nRF24L01+ wireless transceiver radios with the Arduino Uno and Adafruit Pro Trinket 3V. These ultra-low power radios give significant range and adjustable data rate down to 250kbps. The datasheet is here. 2 sensor transmitters and one hub now work, will add more when more nRF radios arrive in the mail.
A few changes have been done to the Soil Vapor Sampler. The design has been flip upside down. This was done to accommodate the new cap. The angle at which the tubing came into the sampler was also modified. The hole is now directly extruded towards the center, this facilitates the insertion of the tubing.
This new cap goes on the bottom soil sampler. This is the piece that gets inserted into the ground first.
The gray long tubes are the pathways where the tubing goes through. With the direct path, the tubes have no troubles going in.
This is the model of a vapor sampler with all the pieces put together. The white piece is the microporous tubing that would go between every sampler and between the last sampler and the cap.
At the OPEnS lab we prototype many projects and need to document them. Pictures are essential to our reports and publications. In order to have the best quality pictures we decided to make a photo booth. Here is the CAD of our booth.
This photo booth is a light box, a tool to take professional looking photos. Our is a DIY version of the ones we could buy. The idea is to have all of the openings covered with tissue paper and a light source on each side of the box. The paper will serve as a filter, giving the object inside an even lighting. Here is an example:
To take the CAD into the laser cutter I converted separate sides of the box into DXF files. These files are then processed by the laser cutter software and then loaded to the laser cutter. Here is what our photo booth looks like so far:
Some materials are still needed but it is almost done.
The foam finally arrived and we have started to experiment with it. I cut various small 10 x 10 mm squares at different power levels, speeds and heights to determine our best settings. The settings that I observed were cutting the best: Power 17, Speed 23 and Height 3315.9mm.
After doing some testing I intended to cut the foam inserts for the drawers. When loading the DXF file from Fusion360 on to MetalCut, the file appeared to be changed. There were some extra lines that were not supposed to be there. After analyzing the MetalCut file, I noticed that the program was making some discontinuities in the design outlines and created lines towards the center to compensate for it. I believe it was doing this because the file had some sharp edges and I think it doesn’t precess these edges very well.
The Fusion360 design is being fixed. Only some pieces seemed to be affected by these sharp edges. After all the changes have been made, the cutting process should go smoothly.
Today I set up a new experiment using the old HP tablet and OHAUS usb scales. The setup was three scales connected to the tablet, which was running Jim Wagner’s custom logging software, ScaleWatcher v3. A beaker filled nearly to the top, a wickless validator filled nearly to the top, and a validator with wick saturated and filled nearly to the top with water will have their masses logged every 10 minutes for a month or until the validators are dry. The wicked validator has a dry weight of 191 grams, while the wickless validator has a dry weight of 75 grams. The beaker and wicked validator are on 4000g scales and the wickless validator is on a 600g scale.