Harvesting Kinetic Energy Through Linear Alternation

by Seth Woodbury, Lochlyn Reed, and Max Eyster

Abstract

The focus of this research was to harness human kinetics as an alternative energy source. Two body parts (shoulder and calf) of 20 randomly selected subjects were tested with a modified accelerometer that measured movement in all three dimensions which ultimately concluded the calf generates the most movement. A linear alternator was created to fit on the calf and utilize its movement. Designed to employ Faraday’s Law of Electromagnetic Induction, a neodymium magnet was elastically suspended inside of a pipe wrapped with magnet wire. Motion of the calf, such as a step, moved the magnet vertically through the magnetic field generating minute amounts of electricity in the coil that traveled through a simple circuit with a bridge rectifier changing the state of the current from AC to DC before charging a lithium ion battery. The stored electricity was directed into an iPhone X.

The energy generated by the prototype was measured in seconds that the lithium ion battery was able to charge an iPhone X. Forty-six independent trials were performed, where a trial consisted of a subject wearing the prototype around their dominant leg for a set amount of time before the prototype was removed and an iPhone X was plugged into the prototype battery while the time it took for the battery to fully discharge was measured and recorded. Forty trials, 20 trials for two different subjects, were performed using a two hour interval. A 2 sample t-test for difference of means was calculated failing to provide significant evidence (p=.9460864317) that the prototype discharge time differed for each subject providing justification to pool all the data. A median discharge time of 75.85 seconds was calculated from the two hour interval trials, and a strong positive linear relationship (r=.9787701852) was found between hours of prototype use and seconds it takes for the battery to fully discharge into an iPhone X, increasing by an average of 40.63130779 seconds per hour. Electromagnetic Induction as a sole power source driven by human kinetics is a relatively untapped idea. Although its inefficiency is present when given the difficult task of charging an iPhone X, future research could prove it to be quite effective when tasked with powering devices with lower voltage inputs.  

Introduction and Predictions

The focus and purpose of this research was to find a way to store electricity in a battery by the utilization of human motion, an alternative energy source. Subsequently, a prototype was developed that took advantage of electromagnetic induction as well as linear alternation to accomplish those goals. The project proposes a solution to the problem of powering devices during active emergency situations, where there is no available power. Between 2007 and 2013, in a study of 59 national United States parks, 1025 park visitors died, an estimated 160 visitors a year (The Editors, 2016). Considering how many people go without power from various other emergencies such as natural disasters, power outages, or even something as simple as a car breakdown, the possibilities to implement this prototype are vast. The prototype built would have allowed for the people in crises like these to be able to charge something such as a cell phone to make an emergency call.

It was hypothesized that the prototype would generate the most energy positioned on the calf. Speculating that the calf would generate the most energy, it was reasonable to assume that a linear shaped device with vertical magnet movement through a coil would be the best design to generate the largest power output. The project was limited to cost and quantifying the data, which was recorded in the seconds the prototype could charge an iPhone X.

Methodology

In this study, students were randomly chosen and requested to wear an accelerometer which was used to measure the location on the body that was the most prolific in kinetic energy production. This preliminary stage of research focused specifically on the upper arm and calf, recording independent measurements of kinetic energy on each. The accelerometer was configured to count any movement exceeding a specific G-force threshold, set and programmed by us, as a significant movement. This was a unique unit created for the observations to measure an estimation of kinetic energy. In each test, the accelerometer was secured by a strap to one of the limbs being studied. Then, for one minute each, the test subject was asked to walk, and then run. Each subject had their dominant upper arm and dominant calf movements observed. The accelerometer gave a total significant movement number, which was recorded, and then the subject was excused. With this testing strategy, twenty data points were recorded for each location, giving a total of forty observations in total. A statistical summary was calculated for each set of data. The calf had the highest median significant movements and thus, the prototype was designed to fit the calf. Additionally, there was a difference between the median significant movements for running and walking for both the arm and the leg, the running movement generated more significant movements confirming the hypothesis.

Beginning the design process, a variety of magnets and copper wires of different strengths and thicknesses were reviewed because those factors have direct effects on electrical output and resistance according to Faraday’s Law of Electromagnetic Induction. Once the magnet and wire was chosen, a shell that could hold the magnet elastically suspended by the springs was developed using Computed-aided Design. The shell was 3D printed, the magnet was inserted, and springs were attached to each end of the shell before being inserted into the PVC pipe. Based on the previously gathered data, the calf was selected to holster the prototype because it is the most kinetically active limb.

Finally, the prototype’s electronic circuit was designed and assembled so that the electricity could flow into a lithium-ion battery. Because electromagnetic induction via linear alternation creates an alternating current, a bridge rectifier was soldered onto the circuit to convert the electricity into direct current to maximize efficiency. Trials with the prototype were then performed with three subjects where measurements of electrical outputs were recorded and compiled.

Results

Figure 1:The prototype (Figure 1) operates on the principles of electromagnetic induction and linear alternation.  The outer casing of the device (1c) consists of a copper coil, mounted on a tube of PVC piping capped on both ends.  The mechanical rigging inside the PVC casing (1b), allows for the movement of the magnet through copper coil, creating an electromagnetic charge.  The magnet is elastically suspended at the middle of the pipe by springs at both ends, enabling the magnet to move in a linear fashion through the coil.  The springs are attached to the caps at both ends to provide stability to the magnet. Electronic configuration of the device (1a) allows for the conversion of energy collected from the electromagnetic induction process in coil to be transferred and stored in a rechargeable battery.  Two wires stem from the negative and positive outputs of the copper wiring used in the coil. These wires connect with two opposite ends on the bridge rectifier, which converts the alternating current associated with the magnet’s back and forth (positive and negative) movement to direct current to increase charging efficiency.  Two separate wires extend from the remaining two outlets of the bridge rectifier, and connect with a micro USB adapter which can plug into and charge the rechargeable battery.

Figure 2:
An experiment was conducted using two subjects, having each perform twenty trials with the prototype. A trial consisted of the user wearing the prototype around the calf of their dominant leg (each subject was right leg dominant) while the subject performed moderate movements (a balance of walking, sitting, and standing). After the designated time set for the trial elapsed, the subject stopped and the prototype was retrieved. Then the iPhone Lightning Charger with the Dock Connector was connected to the prototype battery (see Figure 1) and plugged the iPhone X into the Lighting Charger. The process it took for the prototype battery to reach equilibrium with the iPhone X thus ending the charging process was timed; it was found to be most reasonable to measure this in seconds.

Forty of the forty-six total trials were conducted using a two-hour prototype use interval, while the other six trials were conducted after statistically testing whether the mean electrical outputs, measured in time the iPhone X was able to charge from the prototype, differed per subject (Figure 3). These six trials were each conducted with a prototype use interval differing from two-hours so a model could be created from a broader range of data (Figure 5.1).

Figure 3:To see if the electricity generated differs for each person, a 2 sample t-test for difference of means of the data was conducted from the data collected (Figure 2). ?1 refers to the true mean time (in seconds) the battery charged the iPhone X for all two-hour intervals when the prototype was worn by subject X, and ?2 is the true mean time (in seconds) the battery charged the iPhone X for all two hour intervals when the prototype was used by subject Y. After performing the statistic test, it failed to provide sufficient evidence to reject the null hypothesis because the p-value is larger than the designated significance level (?=.05). Therefore, there is not significant evidence that the mean times of charging differ for each person. However, the conditions to perform this test are only partially met; both normal probability plots are approximately linear and thus it is not unreasonable to assume both population distributions to be approximately normal, and the trials were performed independently. However the subjects were not chosen at random and therefore the data must be interpreted with a degree of caution.

Figure 4:Because the statistic test revealed that it is significantly likely that the true mean energy generated per person is the same, it was reasonable to pool the data to create a large 95% confidence interval. A 1-sample t-interval for means was conducted, where ? is the true mean time (in seconds) that the prototype battery is able to charge the iPhone X for all two-hour intervals when the prototype is used by a person. After pooling the data, the sample size was large enough (n=40 ? 30) to justify a confidence interval because normality can be assumed according to the Central Limit Theorem. There is 95% certainty that ?, the true mean time that the prototype can charge the iPhone X after two hours of use, is between 73.968 and 77.192 seconds.

Figure 5.1:

Figure 5.2a:

Figure 5.2b:

Using a 3rd subject, Z, six additional independent trials were performed where the x variable was manipulated so a reliable model, functioning seconds of iPhone X charging over hours of prototype use, could be created using a variety of data points. Because the 2-sample t-test for difference of means failed to prove that the true mean seconds of iPhone X charging from the prototype battery differed for each subject (Figure 3) it was reasonable to pool all 46 data points (Figure 2 & 5.1). Graphing the data displayed a strong positive linear relationship between the two variables (Figure 5.2a) and further statistical evidence proves this hypothesis.The coefficient of determination suggests that 95.7991055% of the variation in the values of time the iPhone X was charged by the prototype can be explained by this least squares linear regression equation. Similarly the high positive correlation value confirms the graph’s strong positive linear relationship between the two variables. In addition, the residual plot (Figure 5.2b) displays a random scatter proving that a linear function best models the data and that these data have no pattern that may suggest an underlying variable or pattern that would deem this regression equation unreliable for sizably larger or smaller x values not incorporated in the sampled data.

Discussion and Conclusion

After considering the data, it is evident that electromagnetic induction is fairly inefficient and probably not the best way to harness human kinetics. A mean of 75.58 seconds of iPhone X charging (Figure 2) for 40 trials of two hour intervals of light to moderate prototype use (balance of walking, sitting, and standing) is a considerably large period of time that yields a relatively low result.  And it can be stated with 95% certainty that the true mean seconds of iPhone X charging for all two hour intervals done by all people only lies between a mere 73.968 and 77.192 seconds (Figure 4). However, the prototype is quite reliable due to the low pooled standard deviation in Figure 2 and the hypothesis testing suggests that mean electrical output, measured in the time the prototype charged the iPhone X, does not differ by recognizable margins for any user. Furthermore, data in Figure 5.1, performed by a subject separate to those from Figure 2, seems to agree with the forty points taken at a two hour interval because they all fit almost perfectly on a linear regression (r=.9787701852) thus proving time of prototype use and seconds of iPhone X charge to be strongly and positively correlated.

Nonetheless, the prototype could provide the desperately needed power to any distressed persons that are in need of it. Hikers in particular would probably reap the most benefits from this prototype seeing as they often walk vigorously for many hours with minimal breaks. Hypothetically, if a hiker walked at a moderate pace, taking many extended breaks, for eight hours he/she would generate (? = 40.63130779(8) – 5.929062087) on average 319.12 seconds of charging power for the iPhone X. This statistic does not take into account that many hikers move above a moderate pace with little rest time, nor the fact that most serious hikers go for more than eight hours of hiking a day. Therefore, it is not unreasonable to suggest that this statistic would likely be higher than the one modeled by the linear regression equation. If this hypothetical hiker were to get in a traumatic accident and require aid, their iPhone X could instantly be connected to the prototype battery and be able to charge for over five minutes, which would more than likely provide them the sufficient time to make a descriptive distress call. It is important to note that the prototype can power any device with a USB connector on the charging cord but the iPhone X is what was specifically measured in these experiments. This scenario is a concern among many hikers and it happens more than commonly thought. Therefore the prototype could certainly be marketed toward, but not limited to, a hiking based audience.  

Electromagnetic Induction and general technologies harnessing human kinetics for energy is a relatively new and untapped field. “Everyday activities such as walking can generate significant power. Therefore, several harvesters are under development,” but which method for harvesting the kinetics is superior has yet to be discovered (Gorlatova, Sarik, Cong, Kymissis, & Zussman, 2013). A group took a very in depth look at how much movement humans take a day considering all body parts, similar to our group’s initial testing phase measuring which body part generates the most kinetics using a modified accelerometer. We created similar linear alternators utilizing elastic magnet suspension, but where our prototype was designed and tested on the calf the other group fit the device in the shirt pocket, trouser pocket, and waist belt. However, this group did not release the harvested electricity into a device such as a cell phone. In a separate experiment by another group, they successfully and efficiently utilized human kinetics for the exact purpose of powering a cellular device (Ruellan, Turri, Ben, Hamid, & Bernard, 2005). Similarly, they used electromagnetic induction, but with a modified linear alternator. They used the hip movement of walking, contrary to our group’s use of the calf, to mechanically force the magnet in a linear motion through the coil rather than letting the forces of walking push the elastically suspended magnet linearly through the coil. Three different designs applied to different regions of the body, one cannot rule which is superior to the others because the same parameters were not measured but what can be learned from these experiments is that Faraday’s principle of Electromagnetic Induction appears to be the most common way of harnessing human kinetics and possibly the most efficient.

It is important to note the subject selection for trial testing could not be entirely random because three subjects who were willing to spend many hours of trials wearing the device had to be chosen, a rather large and important commitment to this research. This research project also had limited access to monetary aid and electrical measuring tools leading the electrical output to be measured in seconds that the prototype battery could charge an iPhone X. Future groups researching energy harvesting from human kinetics should look at other types of ways to harness the kinetic energy such as employing piezoelectric fibers in a design. For any group looking into electromagnetic induction, they should perform various experiments altering magnet size, thickness of wire in coil, and number of wraps of wire in coil since all these factors alter electrical output in electromagnetic induction. Groups should also experiment with a variety of designs besides linear alternators, such as the mechanically driven hip prototype used by a previous group to power a cell phone. In addition, groups should perform more in-depth testing to decide which body part generates the most movement and should conduct more specific experiments on the subjects wearing the prototype such as a ten minute jog while wearing the prototype to figure out how dramatically the intensity of movement alters the electrical output. Finally, groups should consider testing the prototypes on powering other electronic devices, likely devices requiring minimal electrical input voltages, such as LED lights because kinetic harvesting techniques on humans is not as efficient or advanced as most other electricity generating techniques.

Keywords: Electromagnetic Induction, Linear Alternation; Direct Current, Alternating Current; G-Force, Accelerometer; Kinetic

Bibliography

Ashley, D. (2015). “Analysis of portable charging systems for mobile devices utilizing today’s technology.” University of Arkansas, Fayetteville. Retrieved from: http://scholarworks.uark.edu/cgi/viewcontent.cgi?article=1031&context=eleguht

Cottone, F. Mincigrucci, R. Igor, N. Orfei, F. Travasso, F. Vocca, H, Gammaitoni, L. (2011). “Nonlinear Kinetic Energy Harvesting.” The European  Future Technologies Conference and Exhibition 2011. Retrieved from: http://www.sciencedirect.com/science/article/pii/S1877050911006089

De Pasquale, G. Soma, A., Fraccarollo, F. (2013). “Comparison between piezoelectric and magnetic strategies for wearable energy harvesting.” J. Physics: Conference. Retrieved from: http://iopscience.iop.org/article/10.1088/1742-6596/476/1/012097/meta

“Electromagnetic Induction.” Pardon Our Interruption, Electronics Tutorials.  Retrieved from: www.electronics-tutorials.ws/electromagnetism/electromagnetic-induction.html.

Fox Lang, G., Synder, D. (2001). “Understanding the Physics of Electrodynamic Shaker Performance.” Data Physics Corporation, Dynamic Testing Reference Issue. Retrieved from: http://www.dataphysics.com/downloads/technical/Understanding-the-Physics-of-Elec trodynamic-Shaker-Performance-by-G.F.-Lang-and-D.-Snyder.pdf

Gorlatova, M., Sarik, J., Cong, M., Kymissis, I., Zussman, G. (2013). “Movers and Shakers: Kinetic Energy Harvesting for the Internet of Things.” Technical Report. Retrieved from: http://enhants.ee.columbia.edu/images/papers/MoversAndShakers_1307.0044v1.pdf

Khaligh, A., Zeng, P., Zheng, C. (2010). “Kinetic Energy Harvesting Using Piezoelectric and Electromagnetic Technologies-State of the Art.” Transactions on Industrial Electronics. Retrieved from: https://wenku.baidu.com/view/34e4d5235901020207409c2c.html

Mitcheson, D., Yeatman, M., Rao, K., Holmes, S., and Green, C. (2008). “Energy Harvesting From Human and Machine Motion for Wireless Electronic Devices.” Contributed Paper.  Retrieved from: http://www.mouser.co/pdfDocs/Maxim-Energy-Harvesting-From-Human-and-Machine Motion-forWireless-Electronic-Devices.pdf

Ruellan, M., Turri, S., Ben, A., Hamid, M., Bernard, G. (2005). “Electromagnetic Resonant Generator.” HAL archives. Retrieved from: https://hal.archives-ouvertes.fr/hal-00676119/file/ResonantGenerator_IEEE-IAS_2005.pdf