During the last decades, the interest in generating clean energy by unconventional methods has grown due to the high electricity demand, especially in urban areas.

In this context, several strategies for the conversion of mechanical energy to electrical energy have been identified.

Saucedo et al. evaluated electricity generation from waves and generated electric power of 20V, from a magnetic wheel, to be used in the public transport system [1]. Besides, other authors proposed the generation of electrical energy using piezoelectric devices [2-3]. Furthermore, Márquez and Tlatelpa designed and built a power generator through a system of mechanical gears, chain drive, and alternator output to be used in the mass transport system in Mexico [4]. Thus, 747.5 W were generated from a single turnstile of the station "Indian Green", for one hour. Moreover, Ahmad et al designed and fabricated a power generator induced by the movement of a door to charge a smartphone; as a result, it generated a voltage of 11.54V, which was enough to meet the target [5].

In another case, Blanchette and Al-Haddad implemented a permanent magnet generator of a double rotor with a magnetic gear to collect wind power directly without mechanical conversions; as a result, a prototype of a compact generator of 2,5kVA was proposed, which could be coupled directly to the source [6]. Finally, Sepulveda designed and manufactured a power generator based on rotary piezoelectric, through a polygon-shaped gear. The power generator was built using the pair of the axle of the pedals of a gym’s static bike. A previous device was developed to feed the control circuit of the bike. Therefore, a power of 16mW was generated, for an average input of 300 revolutions per minute [7].

The methodological design of the research was divided into two phases. The first consists of performing a statistical study about pedestrian entrance and exit at Escuela Colombiana de Ingeniería Julio Garavito. The study was carried out from January 15^th, 2017 to December 15^th, 2017. It allows performing a characterization of people’s influx behavior, time bands, use of turnstiles for pedestrian access, torque value, time, and turning distance. The turning speed was calculated considering the time and turning distance. Afterwards, the power was obtained from the calculated turning speed and force. It is important to remark that the power value is an initial condition for the design of the two energy conversion mechanisms. In this way, the average speed of the turnstile was 15.3846 rpm due to the staff entry/exit, and the exit speed of the dynamo was 1583.07 rpm.

According to eq. (2), to calculate the power, the force applied to the turnstile arm was required to be measured.

Force was measured with a dynamometer. To identify the support point used to move the turnstile, this variable was considered in the statistical study about the entrance and exit of the staff at the Institution.

As a result, it was found that 90% of the users support the central segment D2 (Fig. 8) to push the arm, and an average force of 23.6N was measured on this point.

Figure 8 Source: The Authors.

Subsequently, the generated torque was calculated using eq. (3) [10], this value was 3.27 Nm, and it allowed us to know the dynamic moment exerted by a motor (which in this case is the user) on the power transmission shaft (the turnstile arm).

Where F is the force in Newton, and r is the average turning radius of the tourniquet arm.

Then, the second phase of the methodological design consisted in dimensioning, designing, simulating, constructing, testing, and adjusting the mechanisms using magnetic induction and through gears, for the generation of electrical energy.

The mechanism by magnetic induction was based on the construction of a mobile wheel prototype (40cm diameter and 1cm thickness), with equal magnets of 30mm radius and a height of 3mm. Furthermore, it rotated on a specific axis, in which two fixed square (40cm x 40cm, and thickness of 1cm) structures with multilayer coils (3cm) were located. Coils were included for the electricity production.

The prototype was made from wood, and enameled copper coils; 18-gauge AWG were also produced. Coils had an iron core and a radius of 2.54 cm. The two fixed square structures were coupled to the movable wheel on a common axis. These structures were simulated in SolidWorks.

Fig. 9 shows the moving wheel with the neodymium magnets.

Figure 9 Source: The Authors.

The two fixed square structures shown in Fig. 10 contained the group of coils arranged circularly.

Figure 10 Source: The Authors.

The common axis of the mechanism by magnetic induction consisted of an aluminum tube of length 540mm (Fig. 11).

Figure 11 Source: The Authors.

The design of the mechanism using gears started from the gain that the system must have, considering that the average speed of the turnstile (due to the entrance of users) was V1 = 15.3846 rpm, and the exit speed of the dynamo was V2 = 1583.07 rpm. Considering previous data as input information, and following gear design equations, the results shown in Table 2 were obtained.

Table 2

Source: The Authors.

Eight tests were carried out with each of the mechanisms. For each test, the power difference in RMS values was measured at the output of the mechanism on a load resistance of 20Ω (LED bulb for the gear mechanism) and 100Ω (for the magnetic induction mechanism). The values or scores obtained were recorded, coded, organized, and transferred to a table in Excel. The calculation was made by computer, and the data analysis was done through descriptive statistics. The voltage variable was analyzed, and afterwards its relation to power was obtained for the same outlet point. The mean value was the used measure related to a central tendency because this statistical parameter is the most appropriate for levels of measurement by intervals. Median values were calculated, as the arithmetic average of recorded measurement distribution.

Since this measure is sensitive to extreme changes, atypical values were eliminated. The spread of the data was calculated using measures of variability: standard deviation and variance.

Fig. 12 shows the integration of the parts in the system.

Figure 12 Source: The Authors.

Fig. 13 shows the SolidWorks simulation of the gear mechanism integrated into the tourniquet.

Figure 13 Source: The Author.

Fig. 14 shows the first section of the mechanism and the composed gear train.

Figure 14 Source: The Authors.

The statistical analysis of the database related to the number of users, and the time of pedestrian entrance and exit at the Institution, allowed us to identify that on Monday, from 8:00 a.m. to 9:00 a.m. the largest influx of people is presented. Considering previous findings, turnstile 3 was the most used, which generates an average of 3.18 rpm in continuous operation.

By connecting the mechanism based on the mechanical gears to the turnstile and loading it with a 12W LED bulb (Fig. 15), electric power was generated in a range of 12v to 17v. The statistics of the eight tests yielded an average median of 15v, an average voltage of 15v (average power of 12W), an average standard deviation of 1.56, a variance of 2.43, a maximum value of 17v, a minimum value of 12v, and range of 5.

Figure 15 Source: The Authors.

The mechanism by magnetic induction (Fig. 16), connected to tourniquet 3 and loaded with a bank of resistors R = 100Ω, generated a voltage in the range of 7V to 11V. The statistics of the eight tests yielded an average median of 9v, an average voltage of 10v (average power of 1W), an average standard deviation of 1.41, a variance of 1.99, a maximum value of 11v, a minimum value of 7v, and range of 4.

Figure 16 Source: The Authors.

Table 3 compared both mechanisms under some parameters.

Table 3

Source: The Authors.

Due to the magnetic flux equation ф = NBA, the induced voltage depends on the coil core for the mechanism by magnetic induction. The flux is directly proportional to the amount of magnetic induction B = uH; which in turn is directly proportional to the permeability of the coil core material (μ) [12]. Hence, as coil core material increases, the magnetic flux density also increases, as well as the flow, and induced voltage (while working in the unsaturated core region). In the context of this research, and as shown in Fig. 17, the core was constructed in a laminar way to avoid Edy currents, and it was made out of a material with high μ (recycled sheets from the core of a transformer). To concatenate high flux was not possible due to the lack of uniformity in the manual construction of the nucleus.

Figure 17 Source: The Authors.

Another reason why a low induced FEM was obtained by this mechanism was the small size of the Neodymium magnets, and the large separation distance between them, which did not allow efficient use of the high coercivity of the magnets [8] and the area of the wheels (see Fig. 18).

Figure 18 Source: The Authors.

According to Faraday, the direction of the induced current is such that it opposes to the flux variation since the area of the coil does not change, the flux is modified by changing the magnetic field. In this context, there are two determining factors for the correct operation of a mechanism based on magnetic induction: (1) to correctly identify the poles of the magnets and (2) to achieve the interconnection between the coils, since the direction of the induced flow and the winding direction must be considered. In this way, the induced voltage polarity is determined, and the interconnection between the coil terminals is performed to cause voltage overlap [9].

Even though it was all fulfilled, a low induced voltage was obtained. It could be a consequence of the lack of coupling between the magnets and the coils in the magnetic mechanism (equalization or confrontation) and a shorter separation distance between them (see Fig. 19).

Figure 19 Source: The Authors.

The size of the wheels could be re-evaluated, reducing their radius to achieve a more stable system and occupying less space. To increase the speed, and the generated voltage, it is recommended to couple the design mechanisms. Thus, a greater voltage is generated by overlapping (coupled one to the other in the same system), compared to mechanisms working independently.

Regarding the quantification of generated electrical energy, results show an average of generated electrical power of 12W and 1W for the mechanism based on gears and magnetic induction, respectively. Considering the goal of feeding the University access control circuit, the generated energy is enough, and its use is applicable to the isolated generation of electricity for the independent consumption of a circuit. It could be improved by connecting the mechanism in parallel.

Nevertheless, the author acknowledges that the obtained output powers of each mechanism are not high. It is partly because of the mechanical source (turnstile), and the low level of the mechanism efficiency. The difference between generated powers by each mechanism is due to their own features, mechanical loss, due to Foucault, hysteresis, and Joule effect.

For instance, for the mechanism based on magnetic induction, no modification of the coil positions and angles was carried out, nor a reduction of the distance between coils and magnets. Due to the concentration of flux lines, the mentioned modifications could reduce the dispersed flow and increase the magnetic induction, and therefore, a more electromotive force (EMF) could be generated.

As future works, it is important to consider that the improvement of the magnetic flux exploit is possible by performing an analysis of finite elements and modifying different constructive parameters of the designs.

On the other hand, the operational feasibility of this research is based on the statistical study of the staff entrance and exit at the University, collected during the two semesters of 2017. Meanwhile, the gradual increase in the number of new students during these last three years can be translated into more generated electrical energy. The study and experimentation of new alternatives for the generation of clean energy must be remarked, as it will allow reducing the dependence on the use of fossil fuels.

Based on the coupling of the two mechanisms, this research demonstrates the technical and operational feasibility of converting mechanical energy to electrical energy, which was generated from users’ entrance and exit

The geared electrical power generation mechanism generated more power (12W) than its counterpart by magnetic induction (1W).

The power generation can increase, if the multiplication factor of the speed is increased, the above could be achieved by coupling the two mechanisms into one.

The mechanism by magnetic induction can generate more electrical energy if the flow is better coupled. This is achieved by decreasing the separation distance between the magnets and the coils and matching the location of the magnets with respect to the coils (facing faces) in the structures that contain them (wheels).

The obtained energy at the output of the two mechanisms was not regulated, which is an important condition to be considered, for instance, to power the turnstile's electronic circuits.

From this research experience, it can be concluded that the energy conversion based on magnetic induction is not suitable for the generation of electrical energy in large amounts.