Efficient front-end interface design in solar systems

Efficient front-end interface design in solar systems

In order to maximize the potential of solar energy, the front-end interface (located between the battery and the energy mining circuit) must take into account the characteristics of these batteries; through different algorithms and various hardware/software, this can be achieved to maximize the potential of solar energy.

Solar energy seems to be a free, infinitely renewable energy source. In fact, converting the impacting electrons in the sun into usable resources requires rigorous design solutions, advanced electronic equipment, and sophisticated battery charge/discharge management. Solar energy is widely used and can be divided into three main categories:

• Harvesting energy for IoT for data logging (power range is milliwatts, output DC, low voltage.)

• Backup, supplemental energy for residential or remote installations. (Transportable, power between hundreds of watts and kilowatts, output AC, line voltage)

• As part of the power grid, fixed in place, with power reaching tens of thousands of kilowatts and output of thousands of volts of alternating current.

In Figure 1, the solar device (wireless circuit) has a large number of sub-functions, although practical applications may ignore the inconvenience faced by the user. At a high level, the power subsystem is only a small, but actually important, part of the design: it consists of a front end that interfaces with and collects energy from the solar cells; it feeds the energy into the storage element (battery or super capacitor), the system also includes an electrical load management block responsible for extracting energy from the storage elements. The energy (joules) collected by the system is available, but it is released as power (watts) to meet the load requirements. [Power is the conversion product of energy to operate a load; but it must first be collected as energy. ]

Figure 1: In this case, for IoT, a complete solar powered system consists of many functional blocks; functional blocks (such as sensors and RF links) are of no use when used as backup or standby power supply . (Source: Mouser)

In fact, we have to understand exactly how much power can be extracted from the sun. The average amount of solar radiation reaching the upper layers of the Earth's atmosphere is about 1 kilowatt per square meter, or 0.1 watts per square centimeter.

It can be seen from this that, especially in MW-range harvesting applications (where there is no need to worry about I2R losses), any small loss caused by the solar power supply system is crucial. The most challenging aspect of this optimization is at the front end, where the power output of the solar cells must be extracted and harvested. This is because any losses or inefficiencies cannot be made up after this point, and the impact solar energy is lost forever.

1. Efficiency starts with the power socket

Unlike most conventional power sources, such as internal resistors, whose current or voltage sources have fixed parameters, solar cells have unusual properties. Ignoring the output voltage and current, the designer's goal is to obtain maximum power from the solar cell, but both voltage and current will change as operating conditions change.

Based on a given set of operating conditions, there is a unique "operating point" called the maximum power point (MPP) where the battery provides maximum power (V × I) output. To extract power, the resistance of the connected circuit - must match the characteristics of the resistance of the battery.

This matching situation is similar to matching any power supply to a load for maximum power transfer, such as between the output impedance of a power amplifier and the load antenna, or from an antenna to an RF front-end. In most such cases, the parameters of the source and load impedance are relatively constant, so a fixed circuit is formed (in some applications, especially high-performance RF, it is also considered that due to self-heating and environmental conditions, some parameters do not vary with the changes with temperature).

However, the operating conditions of solar cells are not constant and shift repeatedly due to changes in lighting, temperature of the cell, age of the cell, and other factors. As a result, to achieve maximum efficiency, the solar system must dynamically change the load on the battery, which is called maximum power point tracking (hereinafter referred to as MPPT). MPPT reflects the relationship between load line, maximum power line and current voltage, the relationship between load line, maximum power line and power voltage

MPPT can be achieved in several ways: In the "disturb and observe" technique, the front-end resistance is "dithered" while the output of the board is monitored; if it continues to increase in this direction, it is detected within a certain range Where the output is the maximum, this is a standard approach to optimization problems. Other methods include manipulating cell transconductance, driving with swept current or voltage, to determine the cell's internal parameters. Each approach has pros and cons: potential excessive oscillation or "wobble" in the search for the maximum power point, or suboptimal performance in the search for the maximum power point.

2. MPPT selection

In addition to the MPPT algorithm, maximum power point tracking can actually be implemented through an application-specific integrated circuit in hardware or through software (also called software, as part of the programming of the system's microcontroller). Choosing firmware to fine-tune or even change the maximum power point (MPPT) tracking algorithm to the greatest extent may also become a burden on the system; compared with fixed-function integrated circuits, higher speed and more power-consuming processors are required. . As with all engineering decisions, there are bound to be pros and cons to the decision, and at the same time, major cost or power increments cross thresholds.

For small collection systems, single maximum power point tracking via an application-specific integrated circuit is often the most cost-effective and efficient; although as individual cells and zones have different characteristics, because multi-cell arrays are distributed over a larger area, Some are even only a few square meters, so it is necessary to provide separate MPPT pair partitions. Choose between a front-end IC with dedicated MPPT, a collection subsystem IC with embedded MPPT, and a firmware-based MPPT processor, depending on the size of the solar array, power levels, and flexibility needs.

An example of a front-end IC embedded with MPPT is the SPV1020 from STMicroelectronics, Figure 3. This IC integrates a four-phase interleaved direct current/direct current (DC/DC) boost converter to maximize the power generated by the photovoltaic panel. The level of power radiation produced. The SPV1020 controls pulse width modulation (PWM) and is a fixed frequency converter that implements loop control via a perturb-and-observe algorithm run by embedded logic. The switching frequency of the power supply internally defaults to 100 kHz, generating the converter, but can be externally varied from 50 kHz to 200 kHz, while the duty cycle range can be varied from 5% to 90% (tight step increments of 0.2%) .

Figure 3: STMicroelectronics’ SPV1020 is a DC/DC boost converter (embedded with MPPT algorithm) that requires no processor management for its basic operation.

The supplier believes that since the maximum power point calculation is localized (hereinafter referred to as MPP), the system efficiency is higher than when using the topology, where the MPP calculation is implemented through a centralized inverter topology. Since the integrated circuit consists of a field effect transistor and a synchronous rectifier power supply, the number of external devices is minimized. As for long-term reliability, the four-phase interleaved topology of direct current/direct current (DC/DC) converters eliminates the need for electrolytic capacitors, which often limits system longevity. Designers can use multiple SPV1020s in a panel array, one device per panel; the panels can be connected in series, in parallel, or in a series/parallel combination. Vendors also offer evaluation boards of the integrated circuits that demonstrate use in different power levels and architectures.

For electronic circuits, from small IoT to large backups and even mains power, solar energy has huge appeal due to the fact that it is free and never runs out. However, only a small part of the available solar energy is available, so any design must focus on its front-end efficiency (such as MPPT issues) to make this approach economically reasonable and technically feasible. Whether a designer chooses a dedicated front-end integrated circuit or a fully programmable integrated circuit, the design of the processor must consider several factors: solar array size, physical layout, required modularity and cost.