Inspiration: Train Arrival Clock

Antique Metro Display

Great project by Rolf Rando, Re-animating Antiques With Arduino:

Living a hundred feet from the San Francisco MUNI ‘J’ line is a great convenience getting to work, but the train’s unpredictability led me to constantly check online to see when the next train would arrive.



Inspiration: LUZ by Marina Mellado


Really impressive student project by Marina Mellado called LUZ: a lamp which helps people who are physically and psychologically affected by the lack of sun or daylight.




Playing with high voltage

Now I’m playing with high voltage.

Well,12v DC which is much higher than the 5v I’ve been dealing with. And to increase my chance of hurting myself I’ve been dealing with higher current in the 2+ amp scale, which is much higher than the 60mA I’ve been working with. So I’ve been careful not to cross the streams like Egon.

The circuit is really simple: power, MOSFET to switch on the current, and a pull-down resistor. Now that I have an idea of how to get higher-intensity lighting with Arduino, I can reevaluate how the main lamp will be constructed.




Logic board update

Lots of movement on the logic board today.

Instead of printing and assembling 16+ LED boards plus a logic board, why not just put it together (like PB&J) onto one board? Then I only need to attach one PCB to the enclosure, and the less things I need to do myself, the better.

Below you can see a Fritzing printout taped to a 1/2″ piece of foamcore inside a prototype enclosure. I expect that the base will be thicker in order to house more stuff, and by stuff I mean at least a battery and charging mechanism, and so that I can hold the Lamp without touching the glass.


As I was showing this around the office, a gauntlet was thrown down: why use an Arduino Uno r3 when I could use a smaller version without any penalty in processing or functionality, with the added gift of miniaturizing the stack? I didn’t have an answer to this, except, “Umm.”

So challenge accepted!

BigRound+LED+MINI-2_pcbI have now integrated an Arduino Pro Mini 5v into the PCB stack. I also added some power flexibility by adding a power mounting block and an external 5v voltage regulator, even though the Pro Mini can accept up to 12VDC and has a built-in voltage regulator. I wanted to split out that function so that the Arduino doesn’t have to do any additional work. Also, I don’t know exactly how much power the main light will take, and I want some headroom to attach the largest battery I can safely source.

You might remember, dear reader, that I originally started this experiment using an Arduino Pro Mini 3v3 board (oh how we go round the bend again). I decided against reusing the 3v3 board because the GPS module which I found to be most stable needs 5v and I don’t want to go around shifting voltage where I don’t need to. I’m all about reducing complexity now.

With that in mind, I am certainly thinking about removing IC3 & IC4; IC1 & 2 run the 16 compass LEDs, while IC3 & 4 are two additional shift registers  which are legacies from my original idea to build my own LED lighting array. I am now considering using an LED string such as this one sold by Adafruit or a single point source I can vary in intensity (yet to be sourced). Moving my main power supply from a 3v3/5v LiPoly source to a high voltage battery would open up more lighting sources I could use efficiently.

Adafruit and Sparkfun orders placed, let’s play with it when they arrive.




Quick Sketchup renderings

Sketchup is a wonderful prototyping tool – it is cheap (mostly free) and has a huge 3d warehouse you can populate your model with. It also has lots of plugins to render the model out to, most notable VRay and Maxwell. I wanted to learn Maxwell, and while the curve appears steep, here are a quick series of renderings I did in less than 10 minutes.

These renderings aren’t good at all, but this blog is about sharing progress and prototypes.

So there.

center hardware 01b

center hardware 02


Compass PCB prototype 02

Building on the Compass PCB prototype 01 I’ve updated the PCB so that I can install the 16 boards easier, along with doing some little things to reduce the amount of wires which will need to run this array.

LED PCB v0.4I created a Ground Bus which I can solder to Board 01 from the Arduino GND and then daisy chain the next 15 together, so I get rid of 15 unnecessary wires. I then moved the +5v feed (which will come from the shift registers) to the top, so I can solder wire, straight, or angled headers depending on what I decide. Lastly, I moved the support holes to the center so that I will be able to flip the board one way or another to install it. This will allow me to manage the 16 wires which go back to the shift registers in a logical way. How I will do that is for another day, at least I know I have options, and I’m giving myself some flexibility here.



A note about power consumption

The Lamp is a portable object which communicates with the satellites in order to find important (or not) places in the world, and then shine a light.

Pretty simple. But the details are important.

The important part of the above sentence is the concept of portability. I can’t tether the lamp to mains power, or require a giant car battery.

Of all the limitations this project has, is the battery. Right now I’m specifying a 3.7v 2000mAh Lithium Ion Battery with an alternate being the 3.7V 6600mAh Lithium Ion Battery Pack, and connecting it to the system through an Adafruit PowerBoost 500 rechargeable 5V power shield. I guess I could design my own charging system, but lithium-ion batteries scare me. They pack a lot of power per cubic centimeter, but don’t play well with others, tending to melt or blow up if they are shorted, bent, crushed or punctured. So, I will have to create a cage for it so it doesn’t hurt me.

Let’s look at what the system draws without lighting the lamp at all:

Component unit
Compass board0.11mA
Compass LED20.00mA
GPS board55.00mA

So I have 2,000 mAh at worst case battery to play with, but let’s assume a 10% reduction due to recharging, storage loss, etc. This gives me 1,800 mAh to work with. I also am assuming worst-case scenario where the lamp is fully lit (for example at a center of population). In reality, most of the time all the LED’s will be lit. I’m sure there I can use calculus to find the slope and areas of illumination to derive the typical power consumption.

Item1 hr ops2hr ops3hr ops4hr ops
System power110.11220.22330.33440.44
Lamp power remaining1689.891579.781469.671359.56
WS2812 LED @ 60mA
(round down)
28 lamps26 lamps24 lamps22 lamps
Luminous Intensity @ 233 (mcd)36524 mcd6058 mcd5592 mcd5126 mcd
R20WHT-F-0160 @20mA
(round down)
84 lamps78 lamps73 lamps67 lamps
Luminous Intensity @ 600 (mcd)350400 mcd46800 mcd43800 mcd40200 mcd

Besides the total power available, the real limiting factor of the battery is the standard discharge of 500mA which can peak at 1000mA. I don’t want to go toward peaking, because see exploding, above.

I have some options: I could source alternative batteries, use multiple batteries to run different systems or parts of systems, or find a better illumination system so I can get the maximum luminance per mA.


Compass Prototype 1

Building on yesterday’s Compass LED PCB prototype, I wanted to get the compass actually working so I could drive some LED’s.

Using Adafruit triple-axis accelerometer & magnetometer LSM303 and NeoPixel Ring – 24 x WS2812 RGB LED, I was able to get a functioning digital compass fairly quickly utilizing the Adafruit_LSM303 library.

I don’t know if I will use off-the-shelf LED’s for the compass as shown in the Compass PCB prototype 01 or use a Neopixel version via SparkFun, the WS2812 RGB LED Breakout. I save about 1.3 mA per LED versus my chosen through-hole LED if I use the WS2812 with a single channel of the RGB on, but to get white it costs me 60 mA! If I use this breakout board I won’t have to assemble or test out the components and boards or test out a self-designed board, which is a huge time saver and reduces the change of me messing up, but I don’t think I can afford the power for the compass.

If I use the Neopixel for the compass, I’m stuck using the Neopixel library for the compass and my home-brew shift registers to drive the lamp subsystem. Which might not be a problem since they both do different things. In the end it might come down to aesthetics: which system do I want more control over the color of light, the lamp or the compass? I think I would want more control over the lamp colors via the Neopixel than control the color of the compass.

Compass Prototype 01

Until I get a handle on my power consumption, I won’t be able to choose one system over the other.



Compass PCB prototype 01

I’ve been thinking a lot about the Lamp’s compass. Creating a Lamp with built-in dowsing rod capability will both increase complexity – a major headache is dealing with all the shift registers and the accompanying 16 LED’s – not to mention the added power load required to drive an accelerometer and magnetometer continuously. But it will be cool. And that’s the whole point, right?

Magnetic Declination of NE USAWhy do I need both an accelerometer and a magnetometer. The magnetometer is self explanatory: I need to know which way is magnetic north and compare that to the local Lamp heading so I can turn on the correct LED to point me in the right direction. I don’t have perfect balance. The accelerometer is essential to compensate for a non-flat magnetometer reading. I can also get fancy and begin to compensate for magnetic declination by both reading the accelerometer and the GPS location. If I want to get fancy.

I’ll deal with the power consumption and compass later, but right now the thought of custom making then soldering 16 LED housings to make a hexakaidecagon (hexadecagon) makes me want to run down the street. Luckily, I’ve been messing with Fritzing Fab and designed my own little LED PCB which will hold the LED, a resister, pins to go to ground and to the proper shift register, and mounting holes. At US$2.15 per board each one isn’t exactly cheap, but I will have a consistent and rugged board to mount the electronics to and then to an internal housing. Kit-bashing my own would both be more time consuming and I would have 16 times to foul it up.

I tried to fabricate my own LED harness with MakerBot, but it just was a pain in the butt to fab and then I still had all the electronics to deal with anyway. I might have to fab one anyway to hold the PCB’s but that will just give me another factor of safety with the electronics.


LED_part2_pcbAs it is, this board might win as the most simple board produced by Fritzing, but you try to create a run of 16 (or 32 in case I mess up) boards yourself. I’m going to rely on mass production on this one.



Location Scouting

It is amazing what you can find online these days.

I remember one of the first sites I went when the Internet was young was the USGS to look at maps. This was prior to Google Maps and you had to find the correct Quad maps from their less-than-perfect search system. This was amazing, and the USGS still creates amazing maps (many which are free digital downloads):

USGS example

Then there is Google Maps, which is wondrous by itself: an almost complete map of the world along with aerial views. Not to sound old, but even 10 years ago this was unheard of. For example, below is the 2000 Mean Center of Population in Google Maps:

loading map - please wait...

2000 37.696990, -91.809570 2000 mean center of population of the United States Phelps County, MO, 2.8 miles east of Edgar Springs


But it goes beyond just Google Maps.

In the United States, most property-related matters are handled at the County level, a very Jeffersonian adventure in bookkeeping.

Out of 25 centers of population, I was able to find the owners and mailing addresses for all but five of the properties – an astounding 80% success rate; all from the comfort of my computer. Many counties have full GIS which make finding the location’s owner very easy – sometimes all you need is the Lat/Lon, sometimes you have to find landmarks and orienteer to the proper parcel. Below is an example of a particularly nice GIS view from a county in Indiana – every Indiana county I needed to research were using one of two systems, a state mandate to digitize property and parcel information must be in effect. You can see an example below, I edited out the sensitive items (you can find this on your own if you want):


In the end all of this data is essential for this project. The whole reason I went through this is to send letters to each property holder in order to gain legal access to their property so I can photograph the Lamp at the center of population. By my reckoning, none of the 24 population centers are on public land (while some might be in public forests, which makes entering that land ambiguous) so gaining permission to photograph is important.

So now I get to write a letter that says, “Hi, I’m a stranger from New York, I have this funny lamp, and I want to come on our property to film and photograph it.”

Not strange at all. Just another step in a slightly strange art project.