The weather in Ann Arbor this past Saturday was idyllic: a warm summer day with blue skies, save for the large puffy clouds being carried by a gentle breeze. While most of the team enjoyed the conditions while they could, the members of the Strategy division set up in the parking lot were battling the elements. Between checking on upcoming clouds and shifting weights around to prevent their workpieces from flying off with the wind , a delicate and important job was underway. This was I-V testing, and was the last step in processing the solar cells before installation.
Santosh and Jordan position cells on the I-V testing table with large, annoying clouds watching over them.
In electrical power, there are two major factors that we consider: Voltage (“V”, measured in Volts) and Current (“I”, measured in Amps). A third measurement, Power (“P”, in Watts) is equal to V multiplied by I. Strategists need to maximize the Power produced by the solar cells to reduce Quantum’s total racing time, and to do that they need to maximize “V” and “I”. However, the two measures are related – you can’t increase one without decreasing the other.
In regular (or “Ohmic”) power sources – like batteries – I and V are related linearly, and it’s very simple to figure out the V and I to run in order to produce maximum power. Solar cells, on the other hand, are non-linear. You have to experiment with them to find the relationship between I and V. This makes finding the optimum I and V difficult.
A sample IV curve for a solar cell. Notice that as voltage (x-axis) goes up the current (y-axis) goes down, and that the maximum power point is neither at the maximum voltage nor the maximum current.
During the race, tracking the maximum power point is handled by a series of circuit boards called the MPPTs (Maximum Power Point Trackers). However, umsolar’s strategists need a reading before the cells are installed on the car to ensure that the modules were not damaged during tab installation and to verify the cell efficiency advertised by our manufacturer.
David Benson-Putnins and Aaresh Bilimoria examine the results of a test.
The first piece of equipment used in this operation is the tracing unit. Ours is made by IVy Tracer, and is so easily misplaced that we keep it in a special box (below). This unit hooks up to the solar cell modules with alligator clips and to a computer with a USB cable. At the computer’s command, the IVy Tracer runs the cells at various voltages, records the currents that come out at those voltages and produces a graph for the strategists. As long as the shape looks like the example above, the cells have passed their first test – if there was a break, the graph would be straight.
This is the box in which the IVy Tracer is kept. The overly dramatic label is a holdover from previous teams, and helps prevent this surprisingly small (yet vital) piece of hardware from being misplaced.
The second part of the testing involves measuring the efficiency of the solar cells, and requires two pieces of information: the power being produced by each module, and the amount of sunlight power being radiated onto each module. The first piece of information comes from the IVy Tracer graph; the second comes from a piece of hardware called a pyranometer.
The pyranometer reads out the intensity of the sun beating down on it in Watts per square Meter. If our cells were 100% efficient, they would produce this amount of power (about 1000 Watts per square Meter on a very bright day). However, no current solar cells are even close to 100% efficient (a typical rooftop Silicon panel might have an efficiency in the low teens, and current technologies are theoretically limited well below 50%). By dividing the actual power output (from the IVy Tracer) by the maximum possible output (from the pyranometer), we get the actual efficiency of the solar cells. This number will allow the Strategy division to make accurate models of Quantum’s power output under all conditions.
These particular tests both work better under bright sunlight, which proved to be a problem on Saturday. Partly cloudy conditions meant that, in order to stay on schedule, the testers had to set up modules while shaded then spring into action when a clearing appeared in the clouds. All the while, an intermittent breeze kept trying to blow the pieces of protective foam that covered the queued modules all over the parking lot (It occasionally succeeded).
Director of Engineering Santosh Kumar watches for an opening in the cloud cover.
Over half of the modules were tested on Saturday, and the rest were tested on Sunday (a day which featured much bluer skies and calmer winds). We are happy to report that there were no broken modules, and that the Strategists have ample information for their calculations. The modules, cleared for duty, are currently being installed on Quantum and will soon be out on the road for real-world testing!
Visit umsolar.com
Tweet @umsolarcarteam
Like facebook.com/umsolar
Watch youtube.com/umsolar
See flickr.com/umsolar
