Project Development

1. What questions need to be answered?

a. What is the optimal vent configuration in terms of price, time consumption, ease of development, reliability and effectiveness?

b. What is the best spatial configuration of the greenhouse's electronic components? Considerations: moisture, heat dissipation, ease of access, ease of repair.

c. What is the optimum mode of assembly for the shell? Material (price vs. durability,) method of strutcap and hub construction (3D printing vs. casting.)

d. What is the optimal clear covering for the greenhouse shell? Price vs. durability.

2. Tasks to be completed-all of the following have been completed.

a. External shell: 1:2 scale 1V geodesic dome built using this calculator and this strutcap and hub system found on Thingiverse (props to Effalo.)
b. Moving parts: 1 x servo-controlled vent,  1 x water valve, 1 x servo motor.
c. Electronics: 1 x Fab Arduino, 1 x temperature/moisture sensor, solenoid valve, servo, TIP120 transistor, AC-DC inverter power supply for servos and Arduino.

d. Code: an Arduino sketch integrating all of the above.

e. Sundries-plywood mounting board.

3. What has worked?

a. Writing the Arduino sketch has been surprisingly painless-it worked the first time.

b. The servo-controlled vent worked like a charm right off the bat.

c. The solenoid valve took a bit of parts scrounging, but also worked.

d. The domekit worked alright. Casting the hubs was very easy. More on problems below.
4. What hasn't worked?

a. The AM2302 temperature/humidity sensor which I bought from AdaFruit refused to work with the Fab Arduino, though it worked with commercial ones. This turned out to be due to timing differences between the resonators of the two boards. Once Anna changed the library of the sensor to reflect the 8 MHz timing, the sensor worked fine.

b. Due to the weak quality of the MakerBot's extruded strutcaps, they kept snapping when I would push them into the hubs. I solved this temporarily by sanding the hubs a bit, but a more permanent and durable solution would be to cast the caps. This would also speed the most time-consuming part of the construction process-3D printing the strutcaps. Molding may be somewhat difficult due to the caps' intricate internal construction (there is a recess for a nut and a shaft for a bolt.) I think that with high enough quality molding material, it is possible.

5. What will happen when?

This summer I will continue to polish the greenhouse while in Seattle and attempt to sell a few models.