SPOT + EE = SPOTEE: SPOT Hardware Development

Last night Lachlan and I stayed until midnight at the lab working on different things, with Georges who also stayed until 9:30pm (Lachlan is an Australian intern who works with me at the lab and Georges is a Labs visitor from the Netherlands).

What got done:

• The radar reflector is built
• We have a PowerPole to Cigarette Lighter cable for the APRS radio in the car
• We have a mag-mount 5/8 wavelength antenna for the chase car
• The payload bay was designed and a lot of parts of it have been cut out of styrofoam (about 80% of the parts)
  
• The
  MicroTrak radio was tested with the APRS radio and we transmitted my
  callsign (VA2TCV) and the name of the balloon (EDGAR1). It worked.
  However there is some problem in that the GPS coordinates seem to not
  be decoded from the GPS. Maybe due to a bad connection.
• Wires were soldered to the camera so that it could be operated from a spot

We have two short clips:
http://www.youtube.com/watch?v=JB-J96wQfqE
http://www.youtube.com/watch?v=tbGD5YkI0GE


Pictures are here:
http://www.flickr.com/photos/25109749@N03/

BTW, the new official name of the balloon is EDGAR-1.

This weekend, some of the guys will be working on code and on finishing-up the payload. The sky will be ours !
 

Some of us at the Labs have taken-on a project of making a high-altitude weather balloon after-hours. Up until now, we are progressing rapidly and we should be able to meet our launch date of April 5th. The objectives of the project are:

• Have loads of fun working on an hobby engineering project
• Provide examples of custom hardware design and software drivers for the Sun(tm) SPOT platform
• Gather data (pictures, atmospheric pressure, temperature, GPS coordinates and altitude)
• Document the whole endeavor and have an Open Source project around it

We have started a Java.net project for code and hardware and a wiki for info. Our mailing list is also available for up-to-the-minute coverage :)

The payload will consist of the following:

• GPS+GSM Module (Telit GM862-GPS) with antennas
• MicroTrak 8000FA APRS tracking transmitter
  
• SPOT with custom electronics for payload control
• Olympus FE115 digital camera
• Honeywell absolute pressure sensor w/ signal conditionning and LTC2415 24-bit Sigma-Delta ADC
• Diode-based temperature probe connected to an ADT7411 (eDEMO board ADC)
  
• 2x Energizer e² batteries for camera
  
• 3.7V 6000mAh Li Ion battery for the GPS
• 9.6V 2000mAh NiMH battery for the tracking radio
  

We will be using a Kaymount 600g balloon with 450g of free lift. Our preliminary launch site is around Watsonville, CA. We would try to control the payload cut-off point so that with prevailing winds we arrive in the Central valley between Merced,CA and Fresno,CA. According to some initial calculations, this would put us at a max altitude of around 50,000 feet (15.2km) and a travel distance of around 85mi (137km)

Right now, most of the material has been acquired or ordered (balloon, parts, radios, parachute, hardware) and the electronic design is completed. Everything to date will be committed to the SVN as soon as the project gets approved on the java.net site.

In the meantime, here are some pictures of PCBs and parts we have:

 

Batteries and tracking radio 

 

eBALLOON Board and GSM/GPS module carrier board

Olympus FE115 camera body, stripped of its case

If you have interest in this project, feel free to visit the wiki and contact us on the mailing list. We would need some Amateur Radio operators would would be willing to help in tracking the balloon using APRS :) I'll keep you posted on the developments over the next days.

Earlier in the eDAQ development process, I had to rig my EVK1101 AVR32 dev board to support SPI slave operation from the SPI header. It was wired so that the Chip Select line on the header was not the one that is mandatory for use with the slave mode, so the port was stuck as an SPI master. I documented a patch to this problem and it can be found here.

eDAQ REV 1.0 is now alive on the test fixture. It is alive end-to-end meaning that the following are true:

• All power supplies are functional and the daughter card can live even while the SunSPOT is sleeping
• SPI communication between the ARM and eDAQ is functionnal
• An SPI-mode AVR32 bootloader is written and working (both AVR32 and Java sides)
  
• AVR32 bring-up and test code is working

As an added bonus, the UART is also working for test purposes on the AVR32. From now on, everything should start taking shape pretty quickly :) Here is a short video of the beast doing its thing:

Next step is to test every on-board peripheral and start a power manager implementation. 

The eDAQ board prototype is now ready ! The eDAQ board is my main development project here at Sun Labs. The specifications of this new daughterboard are:

• 6 channels of 16-bit ADCs with fully differential inputs in 3 banks:
• Two individual 16-bit ADCs (Analog Devices AD7684) at up to 100ksps
• One quad-muxed 16-bit ADC (Linear Technology LT1867L) at up to 30ksps aggregate
• 4 trigger comparators with programmable references for low-power wakeup on analog events
• AVR32 MCU (AT32UC3B0256) for conversion sequencing and buffering while the Sun(tm) SPOT is asleep
• 40-pins 0.050"-pitch connector for connection to external signals
• External digital I/O
• UART from AVR32
• SPI Master from AVR32 with chip select
• 2 LEDs (Red + Green)
• 2 general-purpose I/Os
• 1 Megabyte of Flash accessible from the SPOT mainboard (main SPI bus) for data storage
• SPI bootloader for field-upgrade of the AVR32 using the SPOT SDK's "ant upgrade" functionality
  

I am currently writing the firmware for the AVR32 and the Java API and will keep you all posted on progress :) A breakout board is in the works with standardized sensor interface circuit blocks that will be adaptable by users for their purposes (ie: easy connection to bridge sensors and current loop output transducers).

 

For those interested, the assembly cost 65US$ per board and the PCB is 4 layers with smallest drill at 8mils.

 Here are some pictures:

Hello there ! My name is Tennessee Carmel-Veilleux and I am an International Student Intern working for Sun Labs in Menlo Park until April 25 2008. I started on January 7th, so this blog is sort of a late start.

I am currently in my senior year of Electrical Engineering at Ecole de technologie supérieure (ETS) in Montreal, Quebec, Canada. I have been part of the SONIA autonomous underwater vehicle project there since my first day at ETS.

At Sun Labs, I work on the Sun(tm) SPOTs project team. My job consists in designing daughter boards for the SPOTs so that new and exciting applications can be developed with the platform. My main project is the eDAQ open-hardware high-resolution data acquisition board (more on that in later posts). I do circuit design, firmware programming and system integration for the project.

I currently reside in San Jose, CA with my fiancée Melissa for the duration of my internship and I am overall trying to enjoy the warmer climate afforded by the Californian winter ;) During my spare time, I like like walking, biking, watching movies and doing embedded systems design.