LED Cube Changes for Test

I have decided to scale down the LED Cube to a 5x5x5 in order to test out the timing and construction.  Unlike a lot of LED Cubes that can be found on the internet, I have not had a lot of free time to spend on this endeavor.  Scaling down like this will enable me to at least build a cube with the parts that I have on hand.  Then, as I have some spare time available to me I should be able to test it out fully.  Besides, a 5x5x5 cube only involves soldering 125 LEDs together not the 512 LEDs associated with an 8x8x8 cube.  Aesthetically, it would be more pleasing to have the 8x8x8 or even a 10x10x10 cube but I really don't have the time to devote to it.  This is just for fun anyway.  So a slight change to the connection diagram:




This reduces the MCP23017 count down to 6 or 7, well below the 8 maximum addresses for a single I2C bus.  I will still attempt to do this through one or two Arduinos for control along separate I2C busses.  This structure also means that I will be controlling the Arduinos on one I2C bus from the Raspberry Pi.  So three small computers in the mix.

LED Projector Thoughts

I was contemplating the LED Projector today.  My thoughts from about a year ago was to have some sort of way of projecting LED lights onto a wall or the ground for the purpose of having a Christmas display.  Originally, I was thinking that I could project a series of LED light strings onto a polished cone.  If I were to do that, you could be in a car and pass by the display and see some sort of display in the cone itself.  There would be a radial layout of LED strings on the ground that would light up some sort of scene that would be projected onto the cone and then into the eyes of individuals looking at the cone.  That is a little tricky because of the geometry involved and whether the eye level is in a car, or from someone who is passing on the street (a little higher).

My next thought on the project was to have some sort of hexagonal mirror that could be rotated at a certain speed.  By having a linear string of RGB LEDs you would be able to paint out a scene onto a wall by flashing the LEDs on and off at a specific rate.  There are hexagonal mirrors that are available in laser printers.  I have two old laser printers that could provide these mirrors but I would need to figure out how to control them via a Raspberry Pi or Arduino.

Weather Station Project #6 - Updated to Stream to the Weather Bug Network

Now that I am signed up for WeatherBug, I am going to use the information at http://sourceforge.net/p/weewx/wiki/weatherbug/ to start streaming.  I am also thinking about a web server application and I notice that you can tweek the weewx code to output ephemeris data along with the plots to a website.  I would also like to plot a starchart for the use of individuals who visit this site (some examples of what I want to do at http://wiki.laptop.org/go/StarChart although it's for the One Laptop project).  I would have to modify this code to put the chart into the html files that are developed via weewx.  More updates later.

LED Cube Project Revamp

Ok, just ordered 90 more RGB LEDs to do some more testing of the concepts.  This time I am going to interface the RPi through a couple of Arduino UNOs.  The diagram is as pictured:





This shows a bus structure in which the Raspberry Pi interfaces across the bus to the two Arduino UNOs and in turn, the Arduinos interface to the MCP23017s.  I am hoping that this structure will resolve some of the timing issues that I was experiencing last year.

Weather Station Project #5 - CWOP up - Adding WeatherBug

The weather station has been humming along very nicely over the past several weeks.  Within a day, my data was put up to Weather Underground and I have been able to review what has been happening on my iPhone apps ever since.  I also got a station ID from CWOP and have been able to update the data to them since.  I got a communique from a CWOP handler and my data has passed the first wicket.  Now I need to find out how good the data is - they are doing some analysis on it.  Hopefully I will meet with a thumbs up and my data will be incorporated into the national weather picture.  My only hesitation at this point is noticing that my barometric pressure reading seems to be higher than the nearest two stations to me consistently.  I may have to calibrate it downward.  I am at a loss to determine how I might measure the value at my house to get an accurate reading.  But I will do some more research and make that determination.

Update: I noticed that the quality control site for my CWOP ID is now reporting an analysis line (as of last night).  So looks like I might be able to get a better feel for what is off kilter now.  Also thinking of putting a patch on the weewx software to send updates to WeatherBug network (if they will have me).

Just signed up for WeatherBug Backyard network.  Have a station ID and information.  Will need to modify the weewx software to accommodate http://sourceforge.net/p/weewx/wiki/weatherbug/ which is a patch to get information out to WeatherBug.

Just Purchased an RPi B+

I was at the local Micro Center and just happened to see that they had a few of these in for $35.  I just did install the 6-20-2014 version of Raspbian via a Windows build.  There are a number of differences on the board that would cause havoc with some of the things I own.  For instance, since they are using a microSD card and the HDMI port is in a different location, I will not be easily able to use it with my MediaPi case; bummer.  It appears that some of the cables that I have should be able to fit onto the GPIO header, even though it has extra pins.  The I2C pins that were on the B have been moved to the GPIO header, but none of the boards that I own use this connector.  I have to be careful in the future to make sure that whatever boards I buy, that they work with the B+.  I am interested in finding out if the power supply they added is all that it is put up to be, especially on the USB ports.  I haven't heard anyone chime in about loss of Ethernet when the USB dongles draw too much power; but time will tell.  More later.


Update: it looks like the new power supply is able to deliver up to 1.2 amps to the USB hub.  That in itself is a remarkable improvement - bravo!  That means that I will not have to be saddled with an external usb hub when I do my experiments.  Looks like I will need to buy a couple more of these.