Week 6 (Outside Lab) - Testing Prototype

The team used this time to run tests of the prototype, both to look at the project's robustness, and to get a better idea of how the pump's flow rate interacts with the plastic tubing delivery system. Previously, the tubing attached to the plastic tee was 4' on each side of the tee and 1' from the tee to the pump, but at this meeting the team decided to reduce each side to 2' for ease of transportation and demonstration.

The PlantParent prototype worked very well initially when set to be on for 4 seconds and off for 2 seconds. The pump was able to fill the tubing in a very short time, and once the tubing was full, the output of water with each cycle was fairly consistent, though this was not measured directly. Another success was the continued ability of the system to accurately respond to changes in the Qt application: when tested at different interval settings, the pump acted accordingly.

The video below shows the prototype working. (Note: the setting for "hours" between watering has been translated to seconds in the Arduino code, for demonstration and testing purposes. This is simply the difference of an easily-changed multiplication factor.)

It was the team's plan to measure how long the pump needed to fill a certain volume in order to determine the flow rate at the end of the tubes, but this was not possible due to a last-minute unidentified software error that disallowed proper control of the system. The nature of this software error is still unknown.

Week 6 (Outside Lab) - Fluid Transfer Research Pt. 1

After the success of the last meeting, the team began to move forward towards a more complete assembly of the PlantParent parts. The next problem to tackle is the layout and connections of the plastic tubing, which will be inserted into the pump's output hole at one end, and split into two outputs with a plastic tee at the other - this configuration can be seen in Figure 1.

Figure 1. Plastic tee with tubing inserted

The biggest complication arising from this part of the project is the reality of changing flow rate: the pump supplies an initial flow rate of 180 mL/s (approximately 6 ms/mL), but after traveling a certain distance through the tubing - plus splitting into two paths - the flow rate at the plant end is definitely going to be smaller than it is initially. The team began to do some research to see if we can accurately calculate what this final flow rate will be, so that it can be applied to the time-to-amount calculations in the Arduino code.

Week 6 (In Lab) - First Test

As usual, this week's lab yielded major progress. The team worked on finishing the wiring of the relay system, in the following steps:

1. Parm and Le stripped the neutral wire on the plug-side of the cord, then soldered the two neutral wires (from the plug-side and pump-side) together. This connection was insulated with electrical tape.
2. The hot wires on the plug-side and pump-side of the power cord were connected via alligator clips to the common pin and normally open pin on the relay, respectively. These connections were well-insulated with electrical tape as well, to remove the possibility of a short-circuit occurring.
3. Two jumper wires were connected to the two coil pins on the relay (again via alligator clips, with all connections insulted by electrical tape), then inserted into the breadboard, where other jumper wires completed the connections from the relay to the ground and power output on the Arduino. These jumper wires can be seen connected to the alligator clips in Figure 1, and Figure 2 outlines all connections to the relay.

Figure 1. Jumper wires in alligator clips. The other ends of the alligator clips are connected to the coil pins on the relay.

Figure 2. Schematic of relay circuit.

1. Davina wrote a simple testing code for the Arduino, which turned the power output pin on and off every two seconds. This corresponded to the relay switching on and off every two seconds. 
2. Michelle filled the water container halfway with water and firmly inserted a small piece of plastic tubing into the output hole on the pump, readying it for testing. 
3. The whole apparatus was brought down to the electrical materials department for a last expert check of the wiring before it was plugged in and tested. The apparatus (without water) can be seen in Figure 3. 

Figure 3. First working prototype of PlantParent, with all components connected successfully.

The results of the testing were extremely promising. When everything was connected, and the apparatus was plugged in, the pump turned on and off every two seconds, as expected. 

There appears to be a minor error in the wiring, so that the circuit is normally closed rather than normally open when plugged in. This just means that the pump is on by default, and the activation of the relay turns it off, rather than the other way around. The team could either change the wiring, or, since it is exactly backwards in function, they could simply reverse the programming logic instead, which is the plan going forward from here.