The Impact of Maximum Power Point Tracking Algorithms on Properties of On-chip PV-based Energy Harvester for IoT devices

Citation: Lastname, F.; Lastname, F.; Lastname, F. Title. Journal Not Specified 2025,1, 0. Received: Revised: Accepted: Published: Copyright: © 2025 by the authors. Submitted to Journal Not Specified for possible open access publication under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Article The Impact of Maximum Power Point Tracking Algorithms on Properties of On-chip PV-based Energy Harvester for IoT devices Adam Hudec* , Viera Stopjakova , Robert Ondica , Miroslav Potocny and Lukas Nagy Department of IC Design and Test, Institute of Electronics and Photonics, Faculty of Electrical Engineering and Information Technology, Slovak University of Technology, Ilkovicova 3, 841 04 Bratislava, Slovakia; [email protected] (A.H.); [email protected] (V.S.); r[email protected] (R.O.); miroslav[email protected] (M.P.); [email protected] (L.N.) *Correspondence: [email protected] Abstract: This article presents the analysis of selected maximum power point tracking (MPPT) 1 algorithms and their influence on developed energy harvester (EH) system under uniform conditions. 2 The energy harvester is an electronic system that converts available ambient energy to electrical 3 energy and regulates its distribution to the output. The aim is to design a energy harvester with 4 highest integration rate possible with consideration of area requirements and low power consumption. 5 To improve the overall energy conversion of the developed harvester, we implemented several MPPT 6 algorithms (Pilot Cell, Constant Voltage, Perturb & Observe) into a dedicated MPPT controller 7 that controls the DC-DC converter. Consequently, we experimentally analyzed their impact on 8 the harvester system. Findings shows that even simple algorithms with smaller chip area and 9 lower power consumption can achieve results comparable to more complex ones. The proposed, 10 manufactured and experimentally evaluated EH chip prototype has proven its expected functionality, 11 and is therefore, fully capable of supplying energy for low-power electronics and battery-operated 12 devices. 13 Keywords: Energy Harvester; MPPT algorithms; Conversion Efficiency; Perturb&Observe; Constant 14 Voltage; Pilot-Cell 15 1. Introduction and Motivation 16 The research and development of electronics is mostly driven forward by actual needs 17 of human society but it is limited by materials and technologies of the given time. Materials 18 and technologies are the dominant factors influencing the miniaturization of electronic 19 systems or their level of integration on a chip. Reducing physical dimensions of components 20 on a chip has positive effect on the power consumption, while increasing the density of 21 components brings more computing power [ 1 ]. These properties enable the emergence of 22 smart and wireless electronics such as Internet of Things (IoT) or wearable electronics. 23 Recently, we are approaching a point where to provide enough of computing power 24 in a small area is no longer issue for many applications. On the other hand, using energy 25 sources more effectively and decrease energy consumption are becoming more and more 26 important. Wearable electronics or IoT make extensive use of various sensor systems for 27 which a permanent source of electrical energy is unavailable and therefore, are powered 28 by batteries. Battery-powered systems have a finite lifespan influenced by their own 29 energy consumption and battery capacity. When the battery is discharged, it needs to be 30 recharged or replaced. In case of wearable electronics, this is of course not a problem but 31 in other applications like sensor systems in hard-to-reach places or implantable medical 32 devices, such maintenance is either rather demanding or impossible. Therefore, it would 33 be beneficial to avoid recharging/replacing batteries altogether. 34 Energy harvesters (EH) are the solution to extension of the battery system lifespan 35 [ 2 ]. EH is a system that harvests ambient energy from various sources (e.g. solar, thermal, 36 Version November 25, 2025 submitted to Journal Not Specified https://www.mdpi.com/journal/notspecified Version November 25, 2025 submitted to Journal Not Specified 2 of 18 mechanical, etc.) and converts it into electrical energy. High output power is often used in 37 practice employing energy harvesters that convert solar energy into electrical one. Energy 38 harvesting systems generally consist of an energy converter, a circuit that adapts the output 39 voltage value (most often a DC-DC converter), and an energy storage that serves as the 40 main energy source and a place where the excess converted energy is stored. Due to 41 nonlinear characteristics of Photo-Voltaic (PV) cells or the inability to adapt to load changes, 42 such a harvesting system is not able to function properly and independently or offer the 43 maximum energy conversion efficiency, and therefore, the presence of a control circuit 44 is necessary. In terms of simplicity, this can be either a Pulse-Width Modulated (PWM) 45 regulator that drives a power switch for energy distribution from the solar cells to the 46 load or batteries. More advanced option is to apply DC-DC converter to provide more 47 stable energy distribution using direct PWM/PFM control or enhanced circuit providing 48 the maximum power point tracking (MPPT). 49 There are several known algorithms that can tune the EH operating point, which 50 approaches or overlaps the maximum power point (MPP). The MPP tuning should take 51 as little time as possible to eliminate the MPP drift, eliminate deformations of the PV cell 52 characteristics under partially shaded conditions or adapt to material degradation due to 53 overheating or aging. The solution to each of the above-mentioned problems increases 54 the complexity of the MPPT algorithm, which leads to robustness of the respective digital 55 circuit and increase in the energy consumption and chip area overhead [ 3 – 5 ]. Digital circuits 56 providing the MPPT control are generally less energy-hungry and occupy significantly 57 less chip area compared to analog circuits, but despite these facts, their increase in total 58 energy consumption can disable the whole energy harvester. Additionally, if they are very 59 complex, the area overhead might be unacceptably large. Based on this information, one 60 may conclude that when developing an energy harvester, it is necessary to take into account 61 several parameters and make a compromise between them. This work therefore focuses on 62 the analysis of selected MPPT algorithms and their impact on the entire energy harvester 63 system in terms of several important aspects. 64 This paper is structured as follows: Section 1brings motivation for the analysis of 65 selected MPPT algorithms and their impact on the developed energy harvester. Section 2 66 describes energy harvesting systems for low-power electronics and offers comparison of 67 selected energy converters in terms of the main parameters. Section 3presents typical 68 types of energy regulation that can be implemented in solar energy harvesting systems. 69 Section 4provides introduction into indirect and direct MPPT algorithms and summa70 rizes advantages and disadvantages of each. General structure of the on-chip EH system 71 (implemented in 65 nm CMOS technology) is described in Section 5, which also explains 72 other used off-chip components to test the whole system. Evaluation of prototyped EH 73 system chip samples using robust measurement setup and achieved results are presented 74 Section 6. Discussion about the measured properties of each implemented MPPT algorithm 75 within the EH system is given in Section 7. The final section concludes the key findings 76 and options for possible future improvements. 77 2. Energy Harvesters 78 Energy harvesters are not a newly emerging concept. On the contrary, they are well79 known and widely utilized in the form of photovoltaic, hydro or wind power plants, which 80 convert solar/mechanical energy into electrical one, typically ranging from kilowatts to 81 megawatts, and this capacity continues to grow. 82 As for devices more accessible for the general public, energy harvesters have appeared 83 in products like calculators and LED lights. However, in recent years, they are increas84 ingly found in more complex low-power electronics and systems-on-chip (SoC). Thus, for 85 purposes of this paper, on-chip EH systems will be considered. As mentioned in Section 86 1, the enhancement of computational power and the reduction of energy consumption 87 open doors for the development of sensor systems, wearable electronics and IoT devices. 88 A critical component of such systems from energy point of view is typically the power 89 Version November 25, 2025 submitted to Journal Not Specified 3 of 18 source, namely the battery. Its ability of supplying the energy to a system is time-limited 90 and depends on the value of electrical load and ambient conditions. By utilizing energy 91 harvesters, it is possible to extend the lifespan of battery-powered systems or in special 92 cases, even completely replace them. 93 A significant advantage of energy harvesters for low-power electronics is the pos94 sibility of applying other types of energy converters (EC), such as piezoelectric or radio 95 frequency (RF) ones, in addition to those already mentioned. It is also useful to combine 96 more converters to create so-called hybrid energy harvesting systems. These systems are 97 more likely to be capable of supplying electrical energy for longer periods of time or they 98 may be able to provide power under conditions where a stand-alone converter would no 99 longer have the necessary efficiency (such as a photovoltaic converter in the dark). 100 Energy harvesters are widely applicable systems but the choice of the appropriate 101 type of harvester depends on the specific application. When selecting parameters such as 102 the final location as well as available surrounding energy sources that can be converted 103 into electrical energy, the value of the voltage generated by the respective converter, its 104 output power density, and last but not least, the physical dimensions must be considered. 105 In terms of the mentioned parameters, the most commonly used energy converter in 106 EH systems is a solar cell. The output voltage of a standalone solar cell typically ranges in 107 hundreds of millivolts, and by simply connecting multiple cells in series, one can achieve 108 a voltage in the range of volts, which is suitable for many integrated circuits. The energy 109 conversion efficiency is approximately 25% and in some cases, it can reach up to 50% [?]. 110 Despite the relatively low efficiency, the available power density is in hundreds of mW/m2 , 111 which provides sufficient energy for many low-power electronic systems. A comparison of 112 parameters of selected energy converters with the PV converter is shown in Tab. 1[6–11]. 113 Table 1. Comparison of selected energy converters in terms of main parameters. Photovoltaic Thermoelectric Piezoelectric RF Output voltage ≈600 mV 10 −100 mV 1−100 V1−4V Energy 100 mW 50 −100 µW/m2 15 W0.0002 −1W (outside) (C) Efficiency 15 −25% 1 −13% 50 −90% 1 −90% 3. PV Energy System Regulation 114 The most commonly converted energy using an energy harvester is solar energy, due to 115 the properties mentioned earlier. By connecting solar cells in series, one can achieve a higher 116 output voltage, or by connecting them in parallel, a higher output current. Adjusting the 117 output power from solar cells, whether on the voltage or current side, is a certain adaptation 118 of the input conditions for the load. However, direct connection of PV cells to the load is 119 inefficient and even dangerous, potentially leading to destructive effects. Thus, inserting a 120 regulator between the solar cell and load can significantly prevent potential damage to the 121 load. Furthermore, using a regulator might increase the efficiency of energy harvesting. 122 Just as the input conditions for the energy harvester change over time, so do the load 123 conditions. Therefore, the role of a regulator is to adjust properties of the energy harvester 124 to eliminate changes in conditions, whether at the input or output, to the greatest extent 125 possible, for example, through pulse-width modulation (PWM) or more advanced methods 126 of tracking the maximum power point of the EH system. 127 3.1. PWM Regulator 128 A straightforward method for controlling the distribution of energy from the energy 129 harvester input to load or battery storage, is based on direct connection/disconnection 130 of PV cells to the load. Its block diagram is shown in Fig. 1. PWM regulator adjusts the 131 Version November 25, 2025 submitted to Journal Not Specified 4 of 18 amount of power transmitted to the output of the energy harvester by driving the power 132 switch, however, it does not adapt the output voltage. To achieve the highest possible EH 133 efficiency, the voltage generated at the input of the energy harvester must be equal to (or 134 close to) the voltage of the charged battery storage. 135 VPV IPV Load Energy storage + - PV cell PV cell PV cell PV cell PV cell PV cell PWM controller VLOAD VLOAD Figure 1. Block diagram of PV energy harvester with a PWM controller. Under very high load or when the battery storage is depleted, the PWM regulator 136 (controller) adjusts its output so that the current flowing to the load is unrestricted, meaning 137 it is regulated solely by the load itself. As the load decreases (or as batteries are charged), 138 the duty cycle of the pulse-width modulated signal is adjusted, leading to a gradual 139 disconnection from the input. 140 Simplicity of such a system positively impacts the cost, making it attractive for various 141 applications. It can reliably transfer energy from the energy harvester input to its output 142 and adapt to changing conditions at both the input and output of the energy harvester. 143 During periods of excess solar energy, particularly in the summer, the system operates 144 with high efficiency. At first glance, PWM regulation seems to be a suitable choice for any 145 practical applications where cost is a primary concern. However, it is important to consider 146 also the drawbacks of this regulation method. One of these is the number and configuration 147 of solar cells necessary. Excessively, high charging power with very depleted batteries can 148 negatively affect the lifespan of the battery storage. Last but not least, PWM regulator is 149 unable to ensure full charge of the energy storage or sufficient voltage level for reliable 150 operation of the powered system. 151 3.2. DC-DC Converter 152 An energy harvester with PWM regulator adjusts the output power but cannot control 153 the output voltage. This voltage depends on the number of solar cells, their interconnection, 154 and the intensity of incoming sunlight but generally, is limited mainly by connecting the 155 solar cells in series. 156 To overcome the low voltage limit imposed by the number of solar cells in series, it is 157 necessary to insert a DC-DC converter between the energy harvester and the electrical load 158 or energy storage, as illustrated in Fig. 2. For the DC-DC converter to operate optimally 159 and efficiently, it must be controlled by an additional circuit, for example MPPT controller. 160 Version November 25, 2025 submitted to Journal Not Specified 5 of 18 DC - DC converter VPV IPV MPPT controller IPV VPV Load Energy storage + - PV cell PV cell PV cell PV cell PV cell PV cell Figure 2. Block diagram of energy harvester with a MPPT controller. 3.2.1. PWM/PFM control 161 DC-DC converters are commonly controlled by Pulse-Width Modulated (PWM) or 162 Pulse-Frequency Modulated (PFM) signal. They secure a suitable output voltage by adjust163 ing the duty cycle or frequency. 164 In the first case, the PWM is a technique for generating a signal whose period (T) is 165 constant. The dynamic change of duty cycle (ratio of the duration of Log.1 ( Ton ) to the 166 duration of one period) is obtained by change of Ton.167 In the second case, the PFM generates a signal with fixed duration of Log.1 (Ton) and 168 variable duration of Log.0. This way the frequency (and period) of the control signal is 169 adjusted. For a better idea, the Fig. 3display the waveform of the PWM and PFM signals. 170 PWM T T T Ton1 Ton2 Ton3 PFM T0 Ton Ton Ton T1T2T3T4 Ton Ton Figure 3. PWM and PFM signal waveforms. 3.2.2. MPPT control 171 The main role of the MPPT regulator is to evaluate the measured input parameters, 172 specifically voltage and current, and to adjust or shift the operating point on the power173 voltage (P-V) curve of the solar cell (Fig. 4) to the maximum power point that can be 174 obtained under given conditions, thereby maximizing energy extraction and the efficiency 175 of the energy harvester. 176 Version November 25, 2025 submitted to Journal Not Specified 6 of 18 VMPP VOV 0 PMPP PV voltage [V] PV power [W] MPP Figure 4. P-V curve of a solar cell and MPP tracking. In regulating the output power of the energy harvester, a wide range of algorithms 177 exist, each suitable for a specific application. Simpler algorithms only need to measure 178 one of the input parameters, voltage or current, and achieve high tuning speeds of the 179 operating point, albeit at the cost of never reaching the actual Vmpp value. More complex 180 algorithms for precise regulation and achieving higher efficiency require providing both 181 variables, and it is possible to expand the inputs to include information about the intensity 182 of ambient light irradiation or temperature. 183 Obtaining and processing input data is one thing but adapting the output voltage 184 of the DC-DC converter based on this information is another. Such a voltage adaptation 185 is achieved by generating so-called Pulse Frequency Modulated (PFM) or Pulse Width 186 Modulated (PWM) control signal. In the case of PFM, the output voltage of the DC-DC 187 converter is regulated by changing the frequency of the control signal. To achieve the 188 highest possible voltage level, the higher frequency must be ensured. High frequency 189 negatively impacts power consumption of the system, increases switching losses and 190 electromagnetic interference, and consequently, decreases the overall efficiency. 191 The second option is PWM regulation, which operates with constant control signal 192 frequency, and the duty cycle changes. Since the switching frequency is constant, it is 193 accompanied by constant losses what creates an imaginary boundary between PWM and 194 PFM to choose the right control signal modulation method. 195 A constant switching frequency is accompanied by constant losses, which when small 196 in relation to the transmitted power, make PWM control more efficient than PFM. On the 197 other hand, if the value of the transmitted power is small, the losses when switching at a 198 constant frequency can exceed this value and such control becomes inefficient. 199 The general power ranges and efficiency levels, for which the selected type of modula200 tion of the control signal is suitable, are shown in Fig. 5[12–14]. 201 Version November 25, 2025 submitted to Journal Not Specified 7 of 18 0Load current Efficiency PFM PWM Figure 5. Power efficiency of PFM and PWM vs load current. 4. MPPT Algorithms 202 The most commonly used energy converter is the solar cell due to its relatively high 203 output voltage, provided power density and stability of the output power. Its current204 voltage (I-V) characteristic or power-voltage (P-V) relationship, is non-linear and typically 205 contains only one maximum, as shown in Fig. 4. When the operating point is located 206 at the MPP on the P-V curve, we say that the system is tuned for the maximum energy 207 extraction. The position of the operating point is determined by parameters such as the 208 amount of incoming radiation on PV cells or the size of load. Since these conditions change 209 dynamically, the position of the operating point shifts uncontrollably and may lead to 210 inefficient energy extraction and potentially disabling the entire system. To control and set 211 the operating point to MPP position, the aforementioned PWM regulator is commonly used, 212 due to its simplicity. To obtain higher power conversion efficiency and higher accuracy, the 213 more advanced regulator based on MPPT algorithms[ 15 ] can be implemented. Based on 214 the type of inputs and the method of tuning the operating point, MPPT algorithms can be 215 categorized into two groups: indirect and direct algorithms. 216 4.1. Indirect Algorithms 217 Indirect MPPT algorithms can be described as algorithms that primarily work with 218 static variables. They process information from a general model or data obtained from pre219 application characterization of the PV cell. This involves laboratory measurements acquired 220 from the I-V and P-V curves in an unloaded state, from which we can extract important 221 parameters such as open-circuit voltage VOV , short-circuit current ISC , and voltage VMPP 222 and current IMPP at the MPP. 223 Based on the obtained data, it is possible to achieve the proper setting of the operating 224 point of the energy converter for the maximum energy extraction. However, it is not 225 possible to dynamically respond to changes in input conditions caused by factors such 226 as changes in illumination, partial shading of the PV cell, and last but not least, material 227 degradation due to aging. Individual indirect algorithms, by their very nature, will never 228 reach the actual MPP but will approach and oscillate in its vicinity. 229 Despite the aforementioned drawbacks of indirect algorithms, some of them are very 230 interesting for many applications. For example, algorithms like Constant Voltage (CV)[ 16 ], 231 Pilot-cell (PC), Fractional Open Circuit Voltage (FOCV), Fractional Short Circuit Current 232 (FSCC)[ 17 ], and others dominate in speed of tuning the operating point and are hardware233 simple, which means they are not demanding in terms of area size and electrical energy 234 consumption. 235 4.2. Direct Algorithms 236 Direct MPPT algorithms continuously adjust the operational point of the energy 237 harvester based on real-time measurements of voltage VPV and current IPV , and thus 238 Version November 25, 2025 submitted to Journal Not Specified 8 of 18 power PPV , using an analog-to-digital converter. They do not require disconnecting the 239 solar cell from the rest of the circuit, as is the case with some indirect algorithms. 240 Additionally, in contrast to indirect MPPT algorithms, direct MPPT algorithms do not 241 require absolute knowledge of the I-V and P-V curves of the solar cell. Instead, they can 242 dynamically adapt to changes in input conditions such as lighting, partial shading of the 243 PV panel, temperature, and even material degradation. 244 The operation of direct algorithms using relative values—specifically by comparing 245 current and previous measurements—enables the energy harvester to function effectively 246 across a wide range of conditions while maintaining the highest possible extraction effi247 ciency. 248 Characteristics such as high accuracy in tuning the operational point, responsive249 ness to dynamic changes, and high efficiency are typical features of even the simplest 250 direct algorithms, such as Perturb and Observe (P&O)[ 18 ] and Incremental Conductance 251 (IncCond)[ 19 ]. Of course, there are also many other algorithms, particularly those based on 252 fuzzy logic[ 20 ] or artificial intelligence[ 21 ], which can better manage local maximum on 253 the P-V curve generated by partially shaded PV panels or can eliminate MPPT drift[ 22 ]. 254 However, such algorithms may require significantly more complex hardware and higher 255 computational power. 256 A comparison of the properties of indirect and direct algorithms operating under 257 uniform conditions is presented in Tab. 2.258 Table 2. Comparison of selected indirect and direct MPPT algorithms under uniform conditions. Algorithm PV Cell Dependecy Sensor Tracking Speed Efficiency V I Constant Voltage YES •FAST LOW Open Circuit Voltage YES •FAST LOW Short Circuit Current YES •FAST LOW Pilot-Cell YES •FAST LOW Perturb & Observe NO • • SLOW HIGH Incremental Conductance NO • • MEDIUM HIGH 5. Developed On-chip Energy Harvester 259 We have designed an energy harvester that converts solar energy into electrical energy. 260 This is a relatively complex microelectronic system, where the goal was to achieve the 261 highest possible level of integration on a chip using standard 65 nm CMOS technology. Its 262 simplified block diagram is depicted in Fig. 6, This figure illustrates parts that have been 263 integrated onto the chip and also discrete components used externally. PV POWER PV SENSE TRIM FILTER ONESHOTOSCMPPT + - M3M_TUNE CS_TUNE VPV Vref DC-DC CONVERTER SHUNT REGULATOR OFF-CHIP COMPONENTS ON-CHIP COMPONENTS REGISTERS PC Figure 6. Simplified block diagram of the proposed on-chip energy harvester. 264 5.1. On-chip Components 265 Parts of the EH system integrated on a chip are the following. 266 Version November 25, 2025 submitted to Journal Not Specified 9 of 18 • DC-DC Converter: Proposed voltage converter specifically for operation with a 267 custom designed and fully-integrated inductor on a chip [ 23 – 26 ] is based on the 268 Conventional Boost Converter (CBC) topology [ 27 , 28 ]. Its core components, such as 269 power switches, inductor and part of the input capacitor are fully integrated on the 270 chip. The low-side power switch is controlled by PFM signal generated by control 271 loop and the high-side power switch is driven as ideal diode with zero threshold 272 voltage by Zero Current Crossing Detector (ZCCD) [29] that is part of the converter. 273 • Control Loop: The control loop adjusts the input impedance of the DC-DC converter 274 and in this way, effectively shifts the EH operating point to a position that ensures the 275 maximum energy extraction from the input under given conditions. This energy is 276 transferred to the output of the energy harvester. Main components of the control loop 277 are fully integrated on a chip. The core of the control loop consists of the following 278 circuits: 279 – Registers: The energy harvester is a complex electronic system integrated on 280 a chip with targeted parameters. During the IC manufacturing, fluctuation of 281 process occurs that may affect the EH system parameters. For this reason, we 282 have added the capability for certain corrections/modifications of parameters 283 using compensation banks, tuning banks and switches that can be controlled 284 externally through register memory, i.e. via a computer. In the case of the MPPT 285 block, it is possible to adjust the initialization parameters, change the type of 286 internal MPPT algorithm or activate the option of connecting an external MPPT 287 circuit implemented in a Field Programmable Gate Array (FPGA) form. The energy 288 harvester contains a relatively large number of registers. For better control and 289 interaction with registers during measurement, a custom Graphical User Interface 290 (GUI) was developed in Python programming language using the Tkinter tool. 291 The developed GUI is shown in Fig. 7. Figure 7. Graphical User Interface for manipulating the registers content. 292 – MPPT: The regulation of energy transfer from the energy converter input to 293 its output is managed by a control signal, which can be of the PWM type or 294 PFM type. Since the entire system was designed for low-power applications, 295 we implemented a circuit for generating a PFM signal. The generation of the 296 proper frequency for the control signal is determined by a circuit operating on the 297 MPPT algorithm. Depending on the algorithm type, various input information 298 is evaluated, such as a 1-bit comparison result of voltages on the loaded and 299 reference solar cells for Pilot-Cell (PC) algorithm or in the case of Constant Voltage 300 (CV) and Perturb & Observe (P&O) algorithms, it involves 10-bit numbers from 301 Version November 25, 2025 submitted to Journal Not Specified 16 of 18 Figure 18. Spiderchart representing key properties of each inspected MPPT algorithm. cycles). It occupies the smallest area of 1258 µm2 and its error rate ranges from 3.05 % to 460 11.52 %. 461 8. Conclusion 462 We have designed an energy harvester ASIC intended for low-power electronics that 463 was prototyped using 65 nm CMOS technology by UMC. Most of the system components, 464 such as the DC-DC converter, MPPT controller, relaxation oscillator, and one-shot oscil465 lator have been fully integrated on the chip. Other components such as shunt regulator, 466 evaluation and measurement circuits (i.e. a voltage comparator and ADC converter), and 467 PV cells were used in discrete form. Different algorithms were implemented in the MPPT 468 controller and their properties compared to find the most proper solution for increasing the 469 overall EH efficiency. Indirect MPPT algorithms require information about the voltage of 470 loaded solar cell ( VPV ) and the voltage of unloaded solar cell ( VOV ). In practice, this means 471 that to obtain the necessary VOV , it is essential to disconnect the solar cells from the rest 472 of the circuit, preventing energy transfer for a brief moment. To avoid interrupting the 473 supply of converted energy, we installed another identical pair of solar cells next to the 474 existing one. This configuration provides VPV from one pair of solar cells and VMPP from 475 the other. To manage the energy extraction from the aforementioned solar cell assembly, the 476 Pilot-Cell MPPT algorithm was used. Its selected features can be configured using registers 477 controllable via a computer. It is possible to choose whether the algorithm will operate 478 with an adaptive or static step. In the case of a static step, the size of convergence step 479 ranging from 1 to 31 can be selected. 480 From performed measurements of ASIC samples and results obtained by applying 481 individual MPPT algorithms, we evaluated the main parameters such as area, power 482 consumption, error from the MPP, oscillation range and speed of convergence. Based on 483 these data, comparison of MPPT algorithms was made. 484 Firstly, the Pilot-Cell algorithm was analyzed. It processes 1-bit information from the 485 comparator. In terms of energy consumption and oscillation range, it achieves the lowest 486 values among the others, thus dominating in two key parameters out of five. The second 487 analyzed algorithm was Constant Voltage, which is similar to PC in its operating principle. 488 However, it works with multi-bit information from the ADC converter. Attributes such as 489 energy consumption and oscillation range acquire higher values than PC, but as for error, 490 speed of convergence and area, it achieves lower values. This means that it dominates 491 in 3 out of 5 evaluated aspects. The last investigated algorithm is P&O. It is classified 492 as a direct algorithm and uses voltage and current information from the ADC converter. 493 This type of algorithm achieves the worst results in energy consumption, size of area and 494 Version November 25, 2025 submitted to Journal Not Specified 17 of 18 convergence speed. As for the error of the MPP and oscillation range, the achieved values 495 are comparable to other MPPT algorithms. 496 According to the analysis described above, the CV algorithm achieves the best results 497 among the evaluated MPPT algorithms. The drawback is energy consumption and its 498 reduction will be one of the goals of the future work. The PC algorithm also achieves 499 excellent results, but despite the fact that it processes 1-bit information from the comparator, 500 it requires larger area on the chip than CV. In this regard, it will also be necessary to perform 501 optimization steps per area. The most complex analyzed algorithm P&O lags behind in 502 several parameters. Despite the more unfavorable properties, we see a great potential for 503 future improvements. The effort will be to get as close as possible to indirect algorithms 504 properties, especially as for speed of convergence and power consumption. 505 Author Contributions: Investigation A.H. and R.O.; conceptualization V.S. and A.H.; writing—original 506 draft preparation, A.H. and V.S.; writing—review and editing, A.H., R.O. and V.S.; methodology, 507 A.H., R.O., M.P. and L.N.; validation, R.O., M.P. and L.N. All authors have read and agreed to the 508 published version of the manuscript. 509 Funding: This work was supported by grants VEGA 1/0705/24 and VEGA 1/0572/25, and the 510 Slovak Research and Development Agency under grants APVV-23-0071 and VV-MVP-24-0313. This 511 work has also received funding from the Chips Joint Undertaking under grant agreement number 512 101139790 and its members, including the top-up funding by Germany, Italy, Slovakia, Spain and The 513 Netherlands. This research was done in the framework of "Matching“ Grant No.09I01-03-V04-00030 514 from Recovery and Resilience Plan for the Slovak Republic. 515 References 516 1. Randall, J.N.; Owen, J.H.; Fuchs, E.; Lake, J.; Von Ehr, J.R.; Ballard, J.; Henriksen, E. Digital atomic scale fabrication an inverse 517 Moore’s Law – A path to atomically precise manufacturing. Micro and Nano Engineering 2018,1, 1–14. 518 2. Othman, A.; Maga, D. Indoor Photovoltaic Energy Harvester with Rechargeable Battery for Wireless Sensor Node. In Proceedings 519 of the 2018 18th International Conference on Mechatronics - Mechatronika (ME), 2018, pp. 1–6. 520 3. Yang, F.; Wu, H.; Li, X.; Gao, Z.; Alkanderi, M. Optimized Maximum Power Point Tracking for PV System Under Partial Shading 521 Conditions. In Proceedings of the 2023 IEEE 2nd International Power Electronics and Application Symposium (PEAS), 2023, pp. 522 697–703. 523 4. Sen, A.; Pradhan, S.; Kumar, A. A Novel Curve Scanning Based Maximum Power Point Tracking Algorithm Under Partial 524 Shading Conditions. In Proceedings of the 2020 IEEE First International Conference on Smart Technologies for Power, Energy 525 and Control (STPEC), 2020, pp. 1–6. 526 5. Khanam, J.; Foo, S.Y. Neural Networks Technique for Maximum Power Point Tracking of Photovoltaic Array. In Proceedings of 527 the SoutheastCon 2018, 2018, pp. 1–4. 528 6. Bizon, N.; Tabatabaei, N.M.; Blaabjerg, F.; Kurt, E. Energy harvesting and energy efficiency. Technology, Methods, and Applications 529 2017,37.530 7. Cho, K.Y.; Kim, J.; Lee, E.J.; Hong, K.k.; Lee, H.W.; Shim, J.M.; Oh, D.J.; Shin, J.E.; Kim, J.S.; Seo, J.K.; et al. Shinsung Solar Energy 531 high efficiency commercial crystalline Si solar cells. In Proceedings of the 2012 38th IEEE Photovoltaic Specialists Conference, 532 2012, pp. 002145–002148. 533 8. Ryu, B.; Chung, J.; Kumagai, M.; Mato, T.; Ando, Y.; Gunji, S.; Tanaka, A.; Yana, D.; Fujimoto, M.; Imai, Y.; et al. Best thermoelectric 534 efficiency of ever-explored materials. Iscience 2023,26.535 9. Vikhor, L.; Kotsur, M. Evaluation of efficiency for miniscale thermoelectric converter under the influence of electrical and thermal 536 resistance of contacts. Energies 2023,16, 4082. 537 10. Seo, G.S.; Shin, J.W.; Cho, B.H. A magnetic component-less series resonant converter using a piezoelectric transducer for 538 low profile application. In Proceedings of the The 2010 International Power Electronics Conference - ECCE ASIA -, 2010, pp. 539 2810–2814. 540 11. Hasanuddin, M.O.; Hafizh, A.Z.; Prakoso, G.A.; Izzuddin, A. Design and Implementation of RF to DC Converter for Low Power 541 IoT Device Using RF Energy Harvesting. In Proceedings of the 2022 International Symposium on Electronics and Smart Devices 542 (ISESD), 2022, pp. 1–5. 543 12. Chen, C.L.; Hsieh, W.L.; Lai, W.J.; Chen, K.H.; Wang, C.S. A new PWM/PFM control technique for improving efficiency over 544 wide load range. In Proceedings of the 2008 15th IEEE International Conference on Electronics, Circuits and Systems, 2008, pp. 545 962–965. 546 13. Yu, J.; Hwang, I.; Kim, N. High performance CMOS integrated PWM/PFM dual-mode DC-DC buck converter. In Proceedings of 547 the 2017 18th International Scientific Conference on Electric Power Engineering (EPE), 2017, pp. 1–4. 548 14. Lee, T.H.; Kim, J.G.; So, J.W.; Yoon, K.S. Mode Control Design of Dual Buck Converter Using Variable Frequency to Voltage 549 Converter. The Journal of Korean Institute of Communications and Information Sciences 2017,42, 864–870. 550 Version November 25, 2025 submitted to Journal Not Specified 18 of 18 15. Gawande, M.K.; Ghulaxe, S.G.; Mahatme, T.R.; Salvi, A.S.; Bagewadi, M.D. Modern approach for hybridization of PSO-INC 551 MPPT methods for efficient solar power tracking. In Proceedings of the 2021 2nd Global Conference for Advancement in 552 Technology (GCAT), 2021, pp. 1–6. 553 16. Hudec, A.; Ondica, R.; Ravasz, R.; Stopjakova, V. Constant Voltage Maximum Power Point Tracking Method for Fully Integrated 554 Solar-Powered Energy Harvester. In Proceedings of the 2024 27th International Symposium on Design Diagnostics of Electronic 555 Circuits Systems (DDECS), 2024, pp. 152–155. 556 17. Ostadrahimi, A.; Mahmoud, Y. Curve Computation MPPT Method Based on Simple Modeling of the Photovoltaic Modules. In 557 Proceedings of the 2020 International Conference on Smart Grids and Energy Systems (SGES), 2020, pp. 819–825. 558 18. Michalka, T.; Hudec, A.; Ondica, R.; Stopjakova, V. Inspection of Maximum Power Point Tracking Methods Suitable for Low559 Power Electronics. In Proceedings of the 2024 International Conference on Emerging eLearning Technologies and Applications 560 (ICETA), 2024, pp. 459–464. 561 19. Hasan, H.; Lubis, A.; Arif, T.Y. Application and Analysis of Incremental Conductance Algorithm for A Photovoltaic System with 562 Variable DC Output Voltage. In Proceedings of the 2021 International Conference on Computer System, Information Technology, 563 and Electrical Engineering (COSITE), 2021, pp. 124–128. 564 20. Zeng, G.; Liu, Q. An Intelligent Fuzzy Method for MPPT of Photovoltaic Arrays. In Proceedings of the 2009 Second International 565 Symposium on Computational Intelligence and Design, 2009, Vol. 2, pp. 356–359. 566 21. Rana, Y.; Pratush, P.; Ansari, S.; Jaiswal, S.; Singh, G.; Virmani, R. Performance Comparison of Conventional and AI based MPPT 567 in Solar PV System. In Proceedings of the 2022 IEEE 10th Power India International Conference (PIICON), 2022, pp. 1–6. 568 22. Killi, M.; Samanta, S. Modified Perturb and Observe MPPT Algorithm for Drift Avoidance in Photovoltaic Systems. IEEE 569 Transactions on Industrial Electronics 2015,62, 5549–5559. 570 23. Ondica, R.; Kovàˇc, M.; Maljar, D.; Stopjaková, V. Fully Integrated Multi-layer Stacked Structure of Integrated Inductor with 571 Patterned Ground Shield. In Proceedings of the 2022 18th Biennial Baltic Electronics Conference (BEC), 2022, pp. 1–6. 572 24. Ondica, R.; Arbet, D.; Kováˇc, M.; Stopjaková, V. Feasibility study towards increasing efficiency of fully on-chip DC-DC boost 573 converter. In Proceedings of the 2020 International Conference on Applied Electronics (AE), 2020, pp. 1–4. 574 25. Ondica, R.; Arbet, D.; Kováˇc, M.; Stopjaková, V. Investigation of Inductor-based Fully On-chip Boost Converter. In Proceedings 575 of the 2021 28th International Conference on Mixed Design of Integrated Circuits and System, 2021, pp. 115–119. 576 26. Ondica, R.; Hudec, A.; Maljar, D.; Stopjaková, V. Efficiency study of MPPT algorithms in HDL for full integration of solar-powered 577 voltage converter. In Proceedings of the 2023 21st International Conference on Emerging eLearning Technologies and Applications 578 (ICETA), 2023, pp. 389–394. 579 27. U, C.W.; Zeng, W.L.; Lam, C.S. Review and Comparison of Integrated Inductive-Based Hybrid Step-Down DC-DC Converter 580 Under CCM Operation. In Proceedings of the 2019 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), 2019, pp. 581 49–52. 582 28. Gholizadeh, H.; Totonchi, N.; Shahrivar, R.S.; Mahdizadeh, S.; Afjei, E.; Abbasi, A. Design and Implementation of A Transformer583 less High Step-Up DC-DC Converter Based on Conventional Boost Converter and Voltage Multiplier Cells. In Proceedings of the 584 2021 12th Power Electronics, Drive Systems, and Technologies Conference (PEDSTC), 2021, pp. 1–5. 585 29. Michal, V. Inductor Current Zero-Crossing Detector and CCM/DCM Boundary Detector for Integrated High-Current Switched586 Mode DC–DC Converters. IEEE Transactions on Power Electronics 2014,29, 5384–5391. 587 Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual 588 author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to 589 people or property resulting from any ideas, methods, instructions or products referred to in the content. 590