4. Discussion

Most of the batteries have a voltage that is sustained for a period of time without dropping much and then suddenly dropping within the span of 5 to 20 minutes, a few exceptions that we noticed were GP tests 1 and 2, as well as Eveready tests 1 and 2. These 2 companies' batteries, have a more linear curve to the discharge rate of their batteries, going from its peak voltage to slowly discharge instead of sustaining that peak voltage.

Some of the key findings that we have discovered from the experiment are that the Matsusho and Eveready batteries lasted the longest before failing to power the motor. The GP and Maxell batteries lasted for the shortest time before stopping the motor.

These findings suggest that the time taken to stop powering the motor is entirely dependent on the internal chemical composition of the battery and that the external factors only have minimal effect on the battery. The bar graph in the data collection section provides a comparison of the different sets of data including the averages of the battery data. The bar graph so to give a more direct comparison of all the times instead of manually comparing all the graphs. From this, there were some anomalies like Eveready 2, although it lasted the longest while powering the motor, its peak voltage was around 0.6 Volts compared to most of the others (namely Energizer, Ikea, Maxell) which sustained a higher voltage of at least 0.65 volts, with Energizer and Maxell peaking at over 1 Volt.

Some of our key findings are that some batteries have an extremely steep decline in voltage after a period of time sustained at higher voltages, comparing our results with the general discharge curve of a battery ( Alkaline Handbook - Panasonic ), we can see that most of our tests on the batteries yield similar results as those other made with a few anomalies being, Matsusho 1, GP 1, GP 2 and Eveready 2.

Concerning our hypothesis, the data extracted from the experiment not only is agreeable but also confirms our hypothesis, as the differences between the brands for the different batteries are extremely different at polar ends of the acceptable spectrum. Though in some areas, for example, we hypothesized that Energizer would last the longest, through this experiment, it seems that that is not the case as Energizer only lasted for a period of time similar to the others and its average discharge timing before it stopped powering the circuit was also not as long as the other brands. The way it is different from the experimental prediction is that the battery graphs are non-linear. The hypothesis graph was linear while the actual graphs mostly dropped abruptly after holding a sustained voltage.

In contrast to the above, another point is that the Energizer batteries we tested were consistent, with both tests yielding similar results and both sets of batteries having a higher starting voltage and voltage output as compared to the other brands we tested.

Some of the experiment's limitations are the circuit used to drain the batteries, as the circuit falls apart easily and is not always accurate. The values may abruptly change if a wire or resistor comes out of its hole unexpectedly due to the circuit design not being sturdy enough to keep them in place, leading to inaccuracies or nonexistence of data collection.

There is no way to determine if the circuit has malfunctioned other than reading the values at the end. One way to improve the experiment is to design a better circuit that can alert when a component does not function as intended. Such a function will save a lot of time and allow more efficiency in experimenting.

Another limitation is with the batteries used, we could get off the shelf batteries but to an extent, not all of the batteries would be of the same “age”, this means when we buy the batteries some may have leaked some of their chemical energy if their production was a long time back. A way to resolve this is to source fresh factory parts to work with, as that way we know that the batteries are less likely to have significant power leakage that can affect the experiment and can give us more accurate results during analysis.

An area for improvement is that we spent a lot of time trying to get the data into a graph, so next time, while the 2nd, 3rd and other experiments are running, we can work with the first set of data to truncate and test our graphs and do things concurrently.

Another area for improvement is that we can graph the data directly in the Arduino software instead of external software, as graphing it directly in the Arduino software makes it easier to organize the results instead of having to scrub through the data and graph it externally.

Our last area for improvement is that we should have conducted more research and been clear of our goal during the December holidays so that we would not need to work intensely during the term and have tight deadlines.