Optimizing photovoltaic systems

Recent developments in system components make a significant impact on solar efficiency

BY HEATHER ROBERTSON
Avnet Electronics Marketing
Phoenix, AZ
http://www.em.avnet.com/solar

Solar power continues to evolve as a viable energy source, and much attention has been given to lowering the cost of solar panels and improving solar panel efficiencies. However, significant developments have also been made in balance of system (BOS) components such as inverters, chargers, and energy optimizers. This article will provide an overview of new architectures and components that are making an impact on solar BOS efficiency.

In Europe, transformerless dc/ac inverters are widely used. In the U.S., however, transformerless inverters have only recently been allowed in some areas by the NEC electrical codes. There are a variety of transformerless inverter topologies—the patented HERIC topology from the Fraunhofer Institute has shown exceptionally high efficiencies. The traditional full H-bridge inverter is shown in Fig. 1 . In the HERIC topology, shown in Fig. 2 , two new switch/diode pairs are shown. This topology provides additional freewheeling paths to limit switching and inductor losses, enabling efficiencies of greater than 98%.

Fig. 1. Full H-bridge for transformerless inverters.

Fig. 2. Heric topology for transformerless inverters.

Transformerless advantages

There are several benefits to transformerless inverters. The transformer stage, included to provide galvanic isolation, can be heavy, expensive, and lossy. Even high-frequency inverter designs with smaller transformers have significant energy loss associated with them, as high as 1% to 2%. In the ongoing effort to reduce the installation cost of PV systems—leading to increased adoption— every bit of energy is essential. So, the move to transformerless inverters will continue.

Disadvantages

Some disadvantages to transformerless inverters should be noted. As mentioned, transformerless inverters do not contain the galvanic isolation provided by the transformer, which can be a safety consideration. However, designs with integrated safety mechanisms such as isolation resistance tests and residual current measurement make transformerless inverters as safe as transformer based designs. Also, evidence suggests that grounding issues with transformerless inverters may cause permanent damage with thin film panels, particularly some CIGS solar panels.

Common features of inverter topologies are the switches in the H-bridge configuration. As mentioned, inverter designs have been trending toward higher and higher frequencies (>30 kHz) to reduce the size and cost of inductors/capacitors and transformers. High-voltage/high-frequency switches are required in these solar inverter architectures.

Running MOSFETs at high voltage/high frequency can cause significant conduction losses. IGBTs are often used in this application because they have far lower conduction losses than MOSFETs. However, they are subject to issues with tailing current during turn-off — increasing their switching losses.

ESBTs

Emitter-switched bipolar transistors (ESBTs), available from ST Microelectronics, provide an excellent solution to this problem. As shown in Fig. 3 , ESBTs consist of a high-voltage BJT and a power MOSFET in a cascode configuration, with a very low conduction voltage drop across the device.

When an ESBT (which looks like a 4-pin device) is paired with an external MOSFET and diodes/resistors as shown in Fig. 3, the circuit looks like a 3-pin device, and can be driven in a manner similar to an IGBT or a power MOSFET. Turn-off energies for ESTBs are significantly lower than those for IGBTs, resulting in higher-efficiency designs and make ESBTs an excellent choice for high-frequency, high-voltage inverter designs.

Fig. 3. ESBT with MOSFET driver.

The traditional architecture of rooftop solar installations has also been examined in an effort to reduce BOS costs and improve efficiencies. In conventional rooftop systems, solar panels are connected in series/parallel arrays, and are quite susceptible to shading and mismatch. For example, if a panel in a series arrangement is under performing due to failure, shading, or dirt, the output of the entire string associated with that panel will be impacted significantly. One solution to this issue is to have dc/dc conversion and a maximum power point tracker (MPPT) at the panel or string level.

Optimization

Panel-level energy optimization is essentially a power conversion and control task, well suited to integration. The functions that need to be performed are to optimize the energy from the solar panel, to convert the panel energy to a consistent voltage or current, and to communicate status to a central controller. This requires a microcontroller or a state machine, analog sense circuitry, dc/dc power conversion, as well as wired or wireless communications.

Each of these functions is well understood, and is well suited to integration at a module level. Integration in this sense provides cost, reliability and performance benefits. Optimized MPPT output at the panel level increases system-level performance and results in efficiency gains potentially leading to cost reduction at the system level.

An example of an integrated MPPT solution is the ST Microelectronics SPV1020. It consists of an integrated boost converter, a MPPT finite-state machine, analog sense circuitry and PLM (power-line modem). A high-frequency interleaved architecture is used for the converter, leading to smaller inductors and capacitors. This highly integrated solution will be available late in 2010.

A large number of industrial applications are suitable for solar power. Examples are off-grid solar-powered street lighting, signage, collision avoidance lighting, security, data acquisition, and remote communications. Industrial applications often use solar power where grid power is not accessible. However, in areas with available grid power, solar power adoption in these industrial applications has been limited due to cost considerations. As with rooftop solar, off-grid industrial solar-powered systems will see increasing adoption with improved cost and efficiency.

Off-grid systems require significant energy storage, typically in the form of batteries. The circuits required to charge batteries safely and efficiently continue to evolve and increase in sophistication and integration. Cypress Semiconductor, for example, has an integrated solar charger reference design using the PowerPSOC processor. Developed to be powered by a 12-V solar panel and to charge 12-V lead acid batteries, the reference design includes MPPT optimization, and a lead acid battery charger.

The Cypress architecture uses a current controlled buck regulator for MPPT and battery charging (see Fig. 4 ). The MPPT and battery charging algorithm embedded in the PowerPSoC uses voltage and current feedback and operates the panel at its peak power by controlling the switches in the buck regulator. The switches in the synchronous buck regulator circuit are also operated in a way to ensure that the current delivered to the battery is as required by the charge state of the battery.

Fig. 4. MPPT/charge control implementation block diagram.

In another example, ST Microelectronics is developing a highly integrated HBLED solar MPPT charger/driver. This fully integrated solution features a lead acid battery charger with MPPT optimization and integrated HBLED drivers. This high level of integration reduces cost, improves reliability, and simplifies design. Releasing in late 2010, this product is ideal for HBLED street lighting applications. ■