the maple seed project

The geometry testing is very nearly complete. At this time, there is no significant attribute to associate with the what causes autorotation. I am looking up some more thorough analysis methods now. Hopefully, I can find some significant feature. The problem is that there are a ton of parameters that have been tested over a few experiments, with not enough data points to make an obvious statistically significant evaluation. My current research will, hopefully, drive out a positive result or re-affirm there is a lack of information.

Once I identified all problems with the design and fabrication of our first prototype daughter board, I went ahead and submitted my Gerber files (basically just PDF files showing the layout and dimensions of the copper traces and pads). to Advancedcircuits.com. Less than a week later five shiny green circuit boards showed up at my door and I couldn't be happier with how they turned out. Overall I would say my first PCB fab house experience has been a good one. Advancedcircuits.com really came through for us on these boards. I have also finished assembly on three of them and have begun to do some preliminary testing. The MCU is working like a champ with the sensors but I had some anomalous problems with the battery charger circuit (the Linear LTC1734 + a power transistor and some capacitors) at first. The charger wouldn't sink any current from the lab power supply so I figured I must have damaged it during soldering. I soldered on another chip using the lowest possible temperature and now the charger seems to be working fine. Horray! All that is left now for our project is to put together the complete device. Good thing, since we have to demo it in just a couple weeks. I don't anticipate any problems but judging by how things have gone so far with this project, "assembly" will probably wind up being something like duct taping the whole thing together at the last minute.

The power supply for the Maple Seed sensor is very close to completion. The printed circuit board with the battery charger has been completed and functions as desired. All that is left is implementing the solar panel and a fully functioning power supply prototype will be completed. Once this has been done, two more power supplies are to be built for the remaining two seeds. An exhaustive power budget is also complete with extensive calculations regarding the solar panel and how much power will be produced at certain times of the day and year. Combining this with TI’s power management algorithm allows for a very good idea of exactly how long the Maple Seed sensor will last in the field.

The first manufacturing runs for the second prototype has been finished. This exciting new prototype is about half the footprint area as the last prototype. The features of this prototype make it easy to assemble. The wing will have the solar panel attached with a built in conduit for routing the solar panel wires, which lead into the head shell. The rest of the components will be housed in the head shell which will be attached directly to the wing. The infrared sensors and antenna will be the only protrusion from the head. This configuration will make it easy for sealing the maple seed from moisture and dust intrusion. As a secondary sealing and binding mechanism, industrial strength shrink wrap which will completely encase the entire device. Next up will be testing the prototype.

The first prototype of the daughter board (or motherboard, depending on how you look at it) has been fabricated in the PCB lab at the college. This board includes the power supply and sensor interfacing circuit to charge the battery, power the CC2430EM module and connect it to peripherals. Though the capabilities of the PCB milling equipment available at the CoE are limited, we though it would be best to produce the first prototype (or number of prototypes until all bugs can be worked out) on-site. A more advanced prototype is to come soon and will incorporate the CC2430 chip on-board, which allows for a decreased footprint and, of course, eliminates the need for a separate module. This board design will need to be sent off to a fab house due to the requisite wave soldering on the QFN packages for the system-on-chip.

After the first prototype was made, we were able to look back and see how thing could be done differently. We are taking our knowledge of the carbon fiber resin induction method and appling to make a more simplified model which is also going to be approximately half the size as the first prototype. Approximately 85% of the work in make the previous prototype was in making the form. By removing some of the unnecessary & complex geometries, the form making process can be done simply and quickly. The main feature that will require the a special for to be made is the head which only need to be a modified hemisphere shape. This hemisphere can be placed open side down onto the flat wing portion. This flat portion doesn't need any special form, just a flat surface. However, in actual maple seeds there is a slightly thicker leading edge. This thicker portion serves as a stiffining mechanism for the wing, and for this project it also doubled as a conduit for wires going down the length of the wing. In the next prototype this feature can be made by placing a flexible (but slightly stiff) rubber tube inbetween the carbon fiber layers before induction. Capping the ends of the tube with plugs will prevent resin from infiltrating the tube. Afterwords the plugs can be removed, and the tube has caused the carbon fiber to be the proper shape for the leading edge, as well as left a tunnel for wiring, and all without requiring any special molding for this feature. This design also lead the the ability to make a large variety of wing foot prints via modifying the flat portion, that is needed is a place for the hemisphere to attach and the appropriate place for the leading edge conduit tube.

Of the three major portions of the power supply, all three have been purchased. Both the battery charger and the solar panel are both on hand while the batteries are being shipped. Testing has begun on the solar panels. They have been fashioned in a way that makes them fit the contours of the wing. The battery charging circuit is currently mounted on the printed circuit board and will be tested along with all other on board components asap. Battery testing will begin as soon as they are received to verify the optimal charging/discharging current. A prototype of all the electronic components of the maple seed sensor is to be assembled soon for testing purposes.

Welcome to the Maple Seed Sensor Network project site. This is your source for all information, documentation, software releases and progress updates for the 2010 Maple Seed project, a collaborative senior design project of ME and EE students at Florida State University. On the documentation page, you'll find design reviews, presentations and other supporting documents. On the download page, you'll find the current alpha release of SamaraNet for TI CC2430/CC2530 ZigBee/802.15.4 Z-Stack enabled sensor devices.