Turtle Sense/Phase Two

Phase two is starting up.

People

• User:Dave -- Electrical engineering
• User:Sam -- Project manager / Software
• User:Tom -- Engineering support
• User:Eric -- Fundraising / Community liaison
• User:Chuck -- Documentation/fundraising
• VOLUNTEER NEEDED -- Web Design and data display

Overview

The plan for Phase Two is to create about 30 boards for installation in the Outer Banks of North Carolina during the Spring and Summer of 2013. The design is based around creating a custom processing mother board connected to an off-the-shelf telemetry board. This mother board is designed to be mass produced.

A second simpler version of the mother board is also be developed which connects to two off-the-shelf boards (one for low power computing and the other for telemetry). Every thing on the mother board will have through-hole components to make the board as easy as possible to assemble a single unit. The idea is to have a set of instructions on this web-site which will allow anyone who can do basic soldering to order the parts (from the manufacturers) and quickly assemble a board, install it on a nest (with approval of the appropriate agencies) to communicate with a website we will construct to monitor and collect the data from the nests.

Organization

Organization of the Sea Turtle project proceeds through informal discussions between the 5 scientists involved. Decisions as to priorities, division of labor, resources and strategies come from these discussions which may be conference calls involving all members or smaller discussions with only two or three of the scientists involved. The latter may involve topics such as probe design and construction, trouble shooting of current prototypes, and device improvements. Important decisions involve a consensus of the group.

Conference Call Summary Oct 18, 2013

Participants Tom, Sam, Dave, Eric and Chuck. The agenda had 10 items, and the important items are summarized on this page. The complete minutes can be found here. We need at least $15k for hardware for the 30 projected units, based on current designs. It appears that the National Park Service will have $50k/year for this project which might be utilized in 2014. Eric's proposal to NC Coastal Recreational Fishing License Fund (BBAFF proposal) may provide funding in 2015; total funds requested (including some labor) are $100k.

Web support, especially the Nerds Without Borders Wiki (Sam developing and coordinating) received extensive discussion. It is proposed as an open source site which covers our progress, what we've developed and learned, and, ultimately, what we consider to be best practices. It is hoped it will become a reference site for workers in the field. A topic of considerable discussion was whether in addition to the public section there would be need for a private section for those primarily involved in the project. Such a section may be needed for legal interactions with future partners. This item was unresolved.

Engineering design changes for version two were discussed by Sam and Tom. Sam reported he'd found some volunteers who were interested in working on a radio version, including prototypes. Design of a prototype is under discussion.

Funding through a Crowd funding source (CrowdRise) was discussed (Chuck coordinating). These require careful preparation, hopefully with a well-advertised buildup to an event with subsequent follow-through. It is important to have feedback to donors, from "swag" to online stroking. Overall it may be beyond our current resources and progress, but might be reconsidered if the project has successes in predicting hatching dates.

Discussions of division of labor resulted in: Eric (Hatteras Island Ocean Center non profit; interface with National Park Service including on-site actions; interface with other turtle groups). Dave/Sam (engineering/electronic communications). Tom (probe design/implementation). Chuck (admin, funding, searches for other turtle groups).

Engineering

Components

Turtle egg sensor

This would be unchanged from the phase one design that uses plastic ping-pong balls (not good ping-pong balls, but cheap plastic knock offs) and a temperature/motion sensor pre-installed on a small circuit board (Qty 1 Price=$14.95). Assembly involves cutting a small slot in a ball, putting in some silicone caulk, inserting a sensor board attached to a cable into the caulk, and then filling the rest of the ball with caulk. A ball cut in half covers the slot and makes a thorough seal with the silicone.

Mother board

We are creating two prototypes of the mother board. One will have through-hole components and the second has primarily surface-mounted parts. The through-hole hole mother board has headers for attaching a pre-fabricated processing board based on the TI MSP430 CPU and another two rows of headers for a prefabricated telemetry board. The surface-mount board only has headers for the telemetry board. The circuitry on the pre-fabricated processing board is duplicated almost exactly on the surface mount boards. Both boards are functionally equivalent and will operate identically with the same software.

By creating two versions of the mother board, we can make the design useful to more people, and ultimately help protect more turtles. The through-hole board is easy to assemble for anyone with basic electronic soldering experience (think hacker space events and high schoolers). The surface mount board is designed for mass production and is very difficult to assemble without expert soldering skills and magnification because the parts are much smaller, there are many more of them (because it does not use the prefabricated processor board) and surface mounted parts are not designed for hand assembly. The surface-mount board is therefore optimized for automated production in larger quantities. In large quantities the cost of components and assembly is competitive with the cost of just the components of the through-hole board. While the design of the surface mount board is much more complicated than the through-hole board, it solves the problem of having to find volunteer labor for assembly.

While both the telemetry and processing boards already have their own power regulators built in, but there needs to be an additional power regulator for the telemetry board on the mother board to regulate battery power down to the voltage required by the telemetry board. The power regulator for the telemetry board can be totally powered down most of the time, and then powered up when data needs to be uploaded (typically just once a day). This greatly reduces the power requirements of the unit. Depending upon how frequently data is uploaded, the units will operate for many months on a single battery charge. The surface-mount board has two power supplies, one for the telemetry board and the other for the micro-processor.

The custom circuit boards are 2.5 inches by 3.8 inches. Three of these boards can be purchased for as little as $51 plus shipping. For larger production runs the price would be much less per unit (about $5 each for a run of 100 including shipping). Other components on the board total between $15 (surface mount) to $22 (through hole). The through hole board also needs a processing board which adds about $18. The $25 savings in parts for the surface mount board is about the same as the cost of having it assembled. The total cost per board is about $40 either way.

Processing board

For the through-hole board we usea pre-assembled processing board made by Oilmex Ltd, that is based on the the ultra-low-power MSP430FR5739 processor by Texas Instruments. The MSP430-HFR5739 board has all the essential components needed. There is a header on the board for easy programming, and a connector which we can use for plugging in the cable to the turtle egg sensor. The MSP430FR5739 CPU has 16K Bytes program FRAM, 1024 Bytes SRAM, 10-bit A/D converter, 16-channel comparator with voltage reference generation and hysteresis capabilities which we can use for monitoring the battery voltage, three enhanced serial channels (capable of I2C, SPI, or UART protocols), hardware multiplier, real-time clock, five 16-bit timers, and more. The board also has power supply filtering capacitor and a 3.3V voltage regulator. There are two rows of headers for each pin of the microprocessor which we will use to plug this board into the mother board. The dimension of the board is 2.15" × 1.10". The boards cost only $18.13 qty 1 and are available from microcontrollershop.com.

The circuitry on the processing board is duplicated on our surface mount board almost exactly. The only difference is we have omitted one auxiliary button which we do not need.

Telemetry board

We plan on using plug in Terminus boards made by Janus. The Janus GSM865CF v1.1 GSM/GPRS Modem with GPS works with AT&T and T-mobile. For Verizon we can use the CDMA864CF v3.0. The boards are interchangeable and will plug into our motherboard. The cost is roughly $140 qty 1 and are from the manufacturer, Janus.

Other components

The only other components necessary are the cable and connector for the turtle egg sensor (a couple of bucks at most), an antenna for the telemetry board (GSM and GPS), which costs about $20 (perhaps more if reception is a problem). The units should operate for roughly 4 months running on 4 or 8 low self-discharge NiMH Batteries. The batteries and holder (TBD) cost about $25. Also, anyone creating these devices will need the hardware necessary for programming the microprocessor (more about this later).

The hardware cost of the electronics is about $250 per unit after start up expenses, and not counting telephony charges, taxes and shipping.