aramok

aramok

Thursday, August 13, 2015

Sun Computer


Click HERE for pics
Components:
1. Teensy 3.1
2. Solar Sensing device, SolarMEMS, ISSD60
3. GPS, Adafruit Ultimate GPS Breakout, MTK333
4. LCD, 20x4
5. SD Card
6. Potentiometer

1. Teensys are great. They don’t have the USB data pins “broken out” so i had to make a little extension out of the main port so i could have a durable USB-B port on the outside of the box.

2. The Solar Sensor itself was trouble. It says it uses the Modbus standard. NOPE. First the differential A and B lines were swapped AND labeled + and -. The sensor data registers began at 1 instead of 40001 (at least i got an error response). The first bits of the response were always corrupted, but I knew is was static data about the model number so i modified the modbus arduino library to return the correct bytes. phew!

3. I used the TinyGPS library for easily decoding NEMA messages.
ARDUINO GPS WARING. NEMA messages are often more than 64 characters long. Most Arduinos and in my case, Teensys, have a Serial buffer only 64 chars long and need to be configured to be longer. On a Teensy, the Serial core files need to be modified.

Originally the box Is supposed to mounted on the hood of a car, possibly with a suction cup. The box is water resistant in case it rains.

Originally The box was to have internet connectivity and serve its data to a MySQL server. I worked with both WIFI and Ethernet modules but ran into issues with both.

I tried to use a stock ESP8266. After a week of work, I determined the AT commands were too buggy to be reliable. For example doing a AT? type read would change some seemingly unrelated setting. Other people confirmed to have these types of problems, good for a quick project but the ESP8266 needs better firmware, updates may be out by now and more custom firmware exist everyday.

I tried to use my Ethernet and had luck! it was reliable to any type of data stream interruption. however my particular Arduino Ethernet shield knockoff must of had its reset pin zapped! because neither high or low pulses reset the device, only touching the reset pin with my finger made it reset. :)

The protocol called Telnet is the standard for shuttling ascii over internet, since 1973.

The box is powered externally through the USB port. I find this is the most compatible method for any situation. A wall charger, PC USB port, Car charger or USB backup battery can all be used to power the device.

There is a TERRIBLE bug i cannot track down, it only happens when other people are looking at it! The LCD will become completely corrupted. it seems to happen more often in a car where there is a lot of vibration. The only way to fix it is to reset the box. In professional product design for anything that will be in a vibration environment, vibration tests are done using a machine like this.

4. The Fact about LCDs: When displaying things on standard character LCD, the screen needs to be cleared between every update to erase extra chars if a displayed number shrinks in number of chars. However clearing and redrawing a 20x4 screen can take long enough that it flashes annoyingly. The easiest remedy is to increase the data rate. A direct hook up instead of a I2C backpack can increase transfer rate and reduce the flicker. Somebody needs to write a library to manage the screen to avoid having to clear and redraw the whole thing, I can't be bothered to do so though, i'll just will use an OLED screen next time. ¯\_(ツ)_/¯

5. For a microSD adapter slot, you can solder directly to a generic microSD->SD adapter.

ARDUINO SD LIBRARY WARNING. A frustrating “bug” i ran into with the default Arduino library is that it only accepts TEN character filenames. An example is “data.txt” which is 8 characters! I recommend using a different library. I don’t think you need a file name extension as long as you open it with a text editor, example “datasheet” will make a file with no extension which is ok.

6. The Potentiometer changes screens. The code includes historesis to prevent jittering between screens.

This device has proven useful in getting quick measurements of The Phenoix's performance before we began tearing it down and refurbishing it. Most prominently doing coast-down tests with the GPS and testing the Array output power. Measuring 964 W/m^2 one day gives us 760 W from the array, we can extrapolate the max output power if the day was 1050 W/m^2, 828W max from the array

Monday, August 10, 2015

Phoenix Solar Racing

Phoenix Solar Racing Website - Note: I may or may-not endorse anything our site says :)

---->>>  CLICK  HERE  FOR  PICTURES  <<<----

This post is just a few words about my current job and life. Without this post, the future projects I post won’t make much sense. :)  My first semester At Oregon State University, I joined to the OSU Solar Vehicle Team. The team's solar car was “out of town” at the Abu Dhabi Solar Challenge . The next semester I tried to become more involved. When the car returned, I came to realize what a monumentally challenging and exciting team project solar car racing was! I took on any responsibility they would give me and began learning everything I could about solar racing.


I built a little Solar Sensor for the team. It was for detecting solar power(W/m^2) and XY sun angle, ahead of the solar car while it was on the road.  The box had an GPS to couple the solar measurements with position data. The solar sensor module itself was donated to the team as apart of Solar Mems sponsorship of the team. It was a huge pain to get working as it used RS-485, and did not follow the RS-485 standards very well. I really don’t recommend it unless you get it for free like we did. :(

Midway through the project, the Electrical Lead I was making it for stepped down. I offered to lead the electrical team and I was elected!

At the time, the returning car, “The Phoenix” was going to be put on a shelf, and an all new Vehicle built. Designs were for the body were finished and tested but ultimately the ending school year and lack of engineers on the team led the New Vehicle project to fall too far behind schedule.

Our deadline is the American Solar Challenge 2016 in July of that year and instead we decided to refurbish the ageing “Phoenix” vehicle. I am designing an all new electrical system for this Phoenix 2.0. It is my hope that this electrical system will also be used in any future vehicle the team makes. I am making sure there is plenty of documentation, flexibility and foresight to do so.

Some things Solar Racing dwells in:
  1. High Power Density Batteries
  2. Super High efficiency motors and controllers
  3. High efficiency Solar Cells and Maximum power point trackers
  4. Sensors for energy statistics and Computer number crunching
  5. Lightweight designs with carbon fiber, Aluminium and Titanium.
  6. Aerodynamics
I’ll make a post about our Phoenix 2.0 “Spec Sheet”

Some other things that have happened:
The Oregon State University Solar Vehicle Team was mysteriously let go by the Oregon State University: College Of Engineering. Our team is now called “Phoenix Solar Racing”. We lost our workshop space, ~10 grand annual sponsorship, legal support, Insurance support, and the nonprofit status that allows sponsors of our team to receive Tax Deductibility on their donations.

The new workshop is now the two car garage at the house I am living at. :)

The Mechanical Lead of our team dragged his feet for months and nobody objected when i took over as Mechanical lead. Now I share Mechanical Lead responsibilities with another teammate.

That's my full time Job, I get payed in the chance to do crazy mad hax with a team of crazy mad people!