Friday, January 11, 2008

Lanterns and Chargers Finished

Nambeg, Ghana - Summer of 2007:
VIllagers gathering to vote on whether to work
with Cooper Union on the Solar Lantern Project

It's unanimous!

Yoorhim Choi and Nambeg Villagers, Summer 2007

Prof. Cumberbatch's Fall 2006 "EID 101: Engineering Design and Problem Solving" Class
This is the group of freshman that began the Solar Lantern Project

...and the Fall 2007 EID 101 Class

That brings us to the present: January 11, 2008. We -- a collection of students who've been to Ghana, former EID 101 students, former sustainable design students, and general Cooper students (and alumni!) -- worked through the holidays, and built eight lanterns and six chargers, just in time for Professor Cumberbatch to bring with him to Ghana. It's been a long month with no rest. But it was all worth it.

Why we started this Blog: To share photos and to document this project. We hope to add to this as we continue to improve the lanterns, and receive feedback from the villagers in Nambeg, Ghana. This blog is for all the people involved who didn't get to see the finished lanterns in person:
  • the freshman students who got their butts kicked (welcome to Cooper) - see Fall 2007 photo above
  • the students who started us on bamboo and lead-acid batteries last year - see Fall 2006 photo above
  • the students who've been to Ghana and worked with the villagers - see Summer 2007 photos above
  • the Cooper security guards who kept an eye out for our packages and kept the building open EVERY day
  • all the friends and family that put up with our absence throughout holidays and birthdays
  • Wendy Lee and Yunglin Gazes who came after work and on weekends to help out
  • Dr. Matt Foster who took time off from work to help out
  • Sarah Lerner who made sure we got all our last-minute shipments
  • Glenn Gross who gave us circuit advice and unfettered access to the shop materials and tools
  • the folks at Heatron who donated the ultra-bright LEDs (that REALLY saved the project -- thanks again Jason!)
  • the EPA for materials funding
  • the NSF for funding our students to go to Ghana
  • the regular folk who donate money to Cooper to keep these projects alive (by covering costs that grants don't)
This is starting to sound like a list of acknowledgements, so we better include a few more people, even though they did see the finished lanterns:
  • Dave Berger, Nick Wong, Allan Ho, John Tuck, Alex Fazio, and Sara Foley -- Cooper students who helped out, not for credit, but just because it was a good project for a good cause.
  • Dr. Nadia Pervez (the word "we" in this blog = Nadia Pervez + Mike Gazes) without whom I'd fall flat on my face.
  • Prof. Toby Cumberbatch who leaves for Ghana in two days to deliver the lanterns and carry-on with the real work. His commitment to improving living conditions in the third world is both inspiring and infectious.

Five of us drove up (thanks Matt) to Prof. Cumberbatch in Connecticut yesterday. Here we are in his living room with the eight lanterns soon to be in circulation in the village of Nambeg.

(Left to Right): Dave Berger, Mike Gazes,
Dr. Toby Cumberbatch, Allan Ho, Dr. Nadia Pervez
(Matt's taking the photo)

A closer look at the eight lanterns (and the professor's carpet).
The barely visible mess of cables on the bottom right are
the lantern battery chargers, and the parts for making more lanterns in Ghana.

Allan building Blue3 in the lab

"Blue3" is the name of Allan's lantern. The "traditional" cylindrical-looking lanterns were named Blue1 through Blue5. The lanterns with the desk-lamp style gooseneck were labeled Red1 through Red3. These lantern "names" will help us track the performance and popularity of each design once they're in circulation among the Ghanaian villagers.

One thing that all the lantern designs have in common is the controls: a push-button on/off switch, a knob to control brightness, and a jack for recharging the lantern's battery.


Dave and his lanterns
(left to right: Blue4, Blue1, and Blue2)

Dr. Nadia Pervez holding her creation, Blue5
(a.k.a. Purple1: a cross between the Blue and Red designs)

Early versions of Blue2 and Red2
(before we discovered the amazing light distribution properties of paper plates)

Some high-quality photos of the finished versions of these lanterns is posted in the February 22, 2008 blog.

Inside view of Blue#2
We tried several different methods for securing the battery, and mounting the printed circuit board. In this photo, the PCB is screwed to the inner wall of the bamboo. The battery is held in place by a long bolt with washers. The bolt presses the top of the battery down, and the washers clamps the battery's sides.

In Blue#1 and Blue#4, the PCB is mounted to the underside of a clay disc, with the LED mounted to the other side. In Red#3 (the PVC lantern), the battery is suspended in place by fishing line.

For the next phase of lanterns, we plan to mount the PCB directly to the light distribution system, creating a circuit-LED-optics module that is adaptable to any lantern housing and design.

Alex and Nick working on Red3

Alex mounting the indicator lights
(he's not playing with his shirt -- he's putting holes in a small
plastic case to hold the LEDs and switch for the charger plug)

Dave baking his clay discs

The Nambeg villagers make beautiful, durable pottery using a pit-fire to heat the clay. In Ghana last summer, Dave worked with the villagers to make a lantern out of clay (see photo below). Kiln-fired clay shrinks during firing, so it is not suitable for making "parts" that fit with other components -- in our case, mounting holes for the LED and circuit, and grooves for the diffuser and bamboo. Pit-fired clay, however, does not shrink or change shape. Dave felt it was perfect for making lantern parts because it used materials and skills the villagers already possessed. Since we couldn't make our own pit-fire in the lab, we tried air-dry clay, but that was too brittle. Sculpey oven bake clay works very well -- it hardens in under an hour at 130C. (When the villagers make their own lanterns, they'll use their own methods and materials. The sculpey was just for prototyping.)

A ceramic lantern made by the Dave Berger, Yoorhim Choi, and the people of Nambeg
(this photo was taken last summer)

The three lanterns distributed in the summer of 2007.
(this photo was taken last summer)

The juice can and ceramic lanterns were made by Dave Berger, Yoorhim Choi, and the people of Nambeg. The bamboo lantern in the middle was made by Prof. Cumberbatch's Spring 2007 sustainability class.

Yunglin soldering components onto the printed circuit boards

The finished printed circuit board for the lantern, with all parts soldered in place.

This circuit drives the LED using pulse width modulation to control the LED's brightness. The circuit also automatically disconnects the battery from the LED to prevent the battery from discharging more than about 50%. (When we originally designed this circuit, Fall 2007, we thought we were only discharging the battery 25%. But after additional experiments in the lab this semester, Spring 2008, we are finding that depth of discharge for our circuit is somewhere between 45% and 55%).

Why It's Important to Limit the Battery's Depth of Discharge....
Limiting the battery's depth of discharge is important for increasing the life of lead-acid batteries. Discharging a lead-acid battery more than 100% (a.k.a. "deep discharge") severely shortens the life of the battery. One measure of "battery life" is the number of charge-discharge cycles before charge retention capacity drops to 60%. When a battery can only hold 60% of its charge, it's considered "dead" (and hopefully gets recycled).

Limiting discharge to less than 100% (and assuming optimal temperature and optimal charging methods), you get about 250 cycles for 100% discharge, compared with 1100 cycles for 30% discharge. In other words, thinking about the total run-time (the amount of time the battery lasted each time you used it, multiplied by the number of charge-discharge cycles), you can squeeze out about 30% additional total run-time from a lead-acid battery, discharging it 30% instead of 100%. So, yes, there is a bit of advantage to limiting depth of discharge to MUCH less than 100%, but it gets a bit more complicated...

In between the battery being dead (i.e. 60% charge retention capacity), and being fresh (i.e. 100% charge retention capacity), there is a period of time where the battery is getting old -- not completely dead, but still not able to hold it's full charge. As you might expect, this "aging" effect is accelerated the closer you get to 100% discharge. For example, discharging the battery 30%, it can still hold a full-charge, even after a little over 400 charge-discharge cycles. Discharging 100%, you lose full-charge capacity in less than 100 cycles. Once again, there is a tiny advantage in limiting depth of discharge to MUCH less than 100%...

On top of this depth-of-discharge-effect on battery life, however, there is an upper limit on how long a lead-acid battery will last. In other words, we cannot discharge the battery 100% once a year, and expect it to last 250 years! Even under the best of conditions -- never using the battery, but maintaining it in a "float" state (meaning that it is kept perpetually charged) -- a lead-acid battery will last less than a decade. And as temperature increases, this upper limit decreases exponentially.

Even without knowing exactly what that upper limit is, we can still try to squeeze every last bit of energy out of the battery before it is ready to be recycled. We plan to improve the cut-off circuit so that depth of discharge begins small, but increases as the battery ages to track the drop in the battery's charge retention capacity. Not only does this increase the battery's usable life, it also maintains a consistent lantern-run-time for (almost) the entire life of the battery.

John soldering the charging jacks

Closeup of the battery charging circuit PCB (printed circuit board).
The PCB is mounted on a piece of sheet metal that also
serves as the heat sink for the series-pass transistor

Description of the Lantern's Battery Usage:

A 6-hour charge provides about 20 hours of run-time at full brightness. A battery disconnect circuit in the lantern prevents its battery from discharging more than 30%, resulting in an expected battery life of over eight years (assuming the lantern is used for 6 hours every night).

We used standard outdoor junction boxes to house the charging circuitry. Each box (there's three in total) contains two charging circuits. A single cable gland on each box provides strain relief for the cables, and moisture protection for the circuitry. To avoid making unnecessary holes in the box, each circuit is mounted on a piece of sheet metal, made to fit snuggly in the box. This same piece of sheet metal is also the heat sink for the circuit's series-pass transistor (see previous photo above).

Big thanks to Sara Foley for drawing the cartoon "Charging Instructions" on the cover plates.

The charging plug

Charging Instructions:
The charging plug is inserted into the charging jack on the lantern. Press the button, and the red LED lights up, indicating that the lantern is charging. When the green LED lights up, the lantern is finished charging.

In a few days, Prof. Cumberbatch will install these chargers at the village's base station (see photos below, from summer 2007). The base station was set up last summer by Dave Berger, Yoorhim Choi, and the Nambeg villagers. The chargers will receive power from the base station's car battery, which is re-charged by two Uni-Solar US-64 solar panels.

Villagers on the roof with the solar panels

Side view of the house, the roof of which is shown in the previous photo.

If you look closely, you can see the electrical cable running from the roof into the house, right above the door.

This is where the cable terminates in the house.

On the floor is a car battery for storing energy from the two solar panels on the roof. The white box on the table contains a charge controller, an automotive fuse block for distributing power, and an inverter (DC to AC converter) for powering maintenance tools, such as a soldering iron or a portable drill.

For more information, please contact Professor Toby Cumberbatch of Cooper Union:
tcumberbatch (at)