The BOS can become quite complex and needs expertise to develop the most energy-harvesting capability from the PV panels.
ANDY SKUMANICH, IGOR J. MALIK, ELMIRA RYABOVA, SolarVision Consulting, Los Gatos, CA USA
While the PV panel is the iconographic center of the solar installation, it makes up only the capture element for solar-to-electricity conversion. The rest of the materials and installation (Fig. 1) is referred to as the balance of system (BOS). For PV installations, the BOS includes all the hardware and components besides the panel, as well as the design and labor needed to put the installation together. A typical rule of thumb is that the BOS is roughly the same cost as that of the panels (excluding batteries from this calculation). This article outlines the BOS elements, describes them, and provides a representative case study for a 30kW system recently developed by the authors.
The key point is that, contrary to being just a simple wiring hook-up, the BOS can become quite complex and needs expertise to develop the most energy harvesting capability from the PV panels. In addition, we will discuss how the BOS has specific challenges in terms of reducing costs—an essential goal in the PV industry.
Figure 1. From a PV panel to an installed solar PV system—the BOS is the other half.
With the general trends in dropping module prices, there is a comparable need in the BOS segment for price reductions to take PV electricity towards grid parity. But aspects of the BOS are not as amenable to cost reductions. In particular, materials costs may increase, and there is also a need for design expertise. Further, because of the complexity, BOS cannot become “plug-and-play,” which, if it were possible, would indeed help lower costs.
In fact, the improvements will likely come most rapidly in terms of better overall system efficiencies as opposed to materials costs reductions. These improvements rely on expert design.
Even for a given set of panels with a given conversion efficiency (CE), there are a multitude of considerations that impact the overall system CE. The planning, components, and details of the installation will alter the system conversion efficiency. This design element is critical because rebates, subsidies, and feed-in-tariffs (FITs) all depend on the maximum power or maximum energy harvested.
|Figure 2. The general layout and a more detailed view of the BOS - everything between the panels and the load or grid.|
The conclusion is that, using the panels as a starting point, the BOS, design, and installation will alter the energy harvesting. With an ability to improve BOS planning, higher system CEs can lead to effectively lower costs of the PV-generated electricity.
Elements of the BOS
There are multiple elements for the BOS that typically include the hardware, as well as related planning and labor costs. Figure 2 shows a typical hybrid system, illustrating some of the BOS components. In general, the BOS hardware is the set of elements that need maintenance and replacement; so, in fact, operations and maintenance (O&M) can be included as part of the more extended BOS set.
The various hardware BOS elements include:
- Mounting—frames, support elements, ground support structures, and base blocks;
- Connecting wires and conduits;
- The inverter;
- Power interface—breakers, transformer, protective switches, etc.
The other elements for the general BOS include:
- System design (including the feasibility study, and detailed plan);
- System installation (including site preparation and permitting);
- System commissioning;
- Operations and maintenance.
Mounting. The PV arrays need to be mounted on a stable structure that can both provide support for the array as well as withstand the elements, such as wind, hail, and even earthquakes. Typically, flat panel mounting is a fixed structure where the frames tilt the PV array at a given angle determined by the latitude of the site, the requirements of the load, and the availability of sunlight. Among the choices for stationary mounting structures, rack mounting may be the most versatile. It can be constructed fairly easily and installed on the ground, or on flat or slanted roofs.
More complex are tracking systems—single- or dual-axis. In these cases, the argument is that even though tracking increases complexity and cost, the final increase in overall energy harvesting provides enough economic value to compensate. The trackers will lead to better light collection by maintaining an optimal orientation between the panel and the sun.
Tracking racks can increase the output of the array from 10% in winter to 30% in summer. Passive racks have no electronics doing the work, so there is less opportunity for failure; active racks, however, optimize power collection.
Power conditioners. The heart of the BOS, power conditioners process the electricity produced by a PV system so it will meet the specific demands of the load. Although most equipment is standard, it is very important to
Figure 3. An example of the range in inverters (l-r): single panel, home rooftop, and utility (images not to scale).
select equipment that matches the characteristics of the load. The inverter is the core of power conditioning. Power conditioners have several overall functions:
- Manage power conversion (DC to AC or to a battery);
- Limit current and voltage to maximize power output;
- Match the converted AC electricity to a utility’s electrical network;
- Have safeguards that protect utility personnel and the network from harm during repairs.
Inverters. As the most critical elements in the BOS hardware, inverters convert DC power into AC power and may provide data monitoring. Inverters are grouped in two major categories—with or without batteries. An extensive amount of commercial and R&D activity is directed towards inverters. Grid-connected inverters run at ~96-97% efficiency, but this drops to 77% for battery-connected (due to the aging effects of the batteries). Typically, inverters need to be replaced as early as 10 years into a given operating cycle. Inverter designs are optimized for PV installations of different sizes because the system efficiency is pegged to the inverters. Figure 3br />shows three different inverter designs, ranging in maximum power load from ~100W for the mini-inverter, to several MW for the utility-scale model.
Electricity storage. If tapping into the utility grid is not an option, a battery backup system is required. Batteries, however, lower the efficiency of a PV system because only about 80% of the energy that goes into them can be reclaimed. They usually need to be replaced every 5 to 10 years. Also, batteries take up considerable floor space, pose safety problems, and require periodic maintenance.
Typically, batteries are not included in the BOS since they are complicating (both in themselves and for the system) as well as costly. Presently, batteries are usually not included in residential and commercial PV systems and are treated as an adder for off-grid systems.
Case study: 30kW installation
A recent project under the guidance of the authors serves as a useful example. This project was for a 30kW ground-mounted, grid-connected system. The main point was the extent of the customization needed to design the project, highlighting the fact that the PV BOS still requires substantial expertise. Multiple initial considerations included: fixed mounting, power to an adjacent mobile office as well as grid connection, how to include a battery system, and even the aesthetic value and layout.
|Figure 4. Typical BOS cost breakdown for an installed PV system|
The various decision points impact the specifics of the BOS and costs of the system. For this installation, the variations in the BOS were manifested by details such as the grouping to the panels to shorten the lines to the inverters, distance to the mobile office and to the closest grid connection, and considerations for off-grid hybrid configurations. There were significant costs in the design phase associated with multiple layers of planning and schematics.
Because each installation has different requirements, the particulars of the BOS and the planning are still determined very much on a case-by-case basis. As a result, there are varying costs associated with different levels of complexity for the electrical and physical layout and planning.
Some of the particulars for the 30kW project included the fact that the earthquake activity for this project must be accounted for as it is in a Zone 4 (high risk area). Wind loading was another necessary design parameter. In general, some of the design considerations associated with the BOS components are:
- Stand alone inverters—Load compatibility, power rating, power quality, battery up-keep;
- Grid-tied inverters—Islanding, susceptibility to line disturbances, RFI;
- Hybrid inverters—Load compatibility, generator compatibility, battery up-keep;
- Batteries—Type, depth of discharge, rate of charge, lifetime (in PV applications).
The main point from this case study is that the BOS is dependent on the design details that play a key, but less obvious, role. Consequently, even for equally sized installations, both the design and the hardware will vary in details and, of course, in cost.
As indicated previously, the PV industry as a whole needs to follow the decreasing cost/price trend to get to grid-parity. For the PV module, there are advantages from technology improvements and for economies of scale with the manufacturing process. The BOS needs comparable reductions, and indeed, some of this has been achieved . But there are elements of the BOS that may be less amenable to cost reductions, and may, in fact, be susceptible to increases. A typical breakdown of the costs is shown in Fig. 4.
In particular, there are the possibilities of increased costs in the coming years for the frame materials; the prices had actually increased in the last 2-3 years by roughly a factor of two because of the construction demands prior to the global economic slow-down. Additionally, the wiring connections use copper, and the price of copper had increased by roughly 80% in the previous two years. Indeed, the data (from ref. 1) indicate that the BOS costs for the last couple of prior years held flat, whereas the PV panels showed price drops. It is likely that the costs of these raw materials will not drop further.
Figure 5. Andalay Panel stack and parts (l) versus a more conventional stack (r). A single tool is required for assembly of the Andalay.
There is the possibility for improvements in labor costs with more training and standardization between the panels and BOS hardware elements. The labor can be partially amortized for the larger installations. A way to reduce BOS costs for the smaller residential applications is seen in the approach taken by Akeena Solar, which has developed a type of jig-saw puzzle connector in its Andaley product (Fig. 5). Akeena calculates that the labor reduction can be 50% .
The BOS makes up the second half of the PV installation and comes between the panel and the load or grid feed-in point. Although it is common to focus on the particulars of the panels, almost half of the overall PV system cost is in the BOS and installation. Advances are being developed to have more plug-and-play types of modules for installations, but the PV panels are potentially dangerous electrical elements and need adequate competency to install. Additionally, optimization of the energy harvest to the specific attributes of an application requires expertise to determine the best components and layout.
As with the modules, there may be opportunity for decreased costs for the BOS, improving component lifetime and reliability. The inverter is the key element for the BOS, and better functionality and lower costs are already evident. At present, one important factor is to increase the lifetime for inverters to match that of the panels (20 years). The battery part of a BOS is a segment unto itself in terms of complexity and cost, and there is major opportunity for improvements.
There are significant differences in the BOS for different installation sizes, generally with the larger >1/2MW installation benefiting from economies of scale. The PV industry is beginning to enter a more expanded phase, and the design and execution of PV systems will benefit from the increasing scales in the industry. It is likely that as the PV panels reduce in price, the focus on the BOS will accelerate and lead to real improvements in costs. In addition, the ability to squeeze out the maximum potential from a given installation with sophisticated planning will also lead to overall system efficiency improvements. The overall $/W will continue to drop and help achieve the necessary lowered costs for the PV industry to rapidly expand into the energy market.
- R. Wiser, G. Barbose, C. Peterman, N. Dargouth, LBNL Publication LBNL-2674E. Tracking the Sun II: The Installed Cost of Photovoltaics in the U.S. from 1998-2008. Published Sept 2009. http://eetd.lbl.gov/ea/emp/reports/lbnl-2674e.pdf
- B. Cinnamon, correspondence.
Andy Skumanich received his PhD in physics at the U. of Calif. at Berkeley, and is founder and CEO at SolarVision Consulting, 412 LG Almaden Road, Los Gatos, CA 95032 USA; [email protected]
Igor J. Malik received his PhD in chemistry at the U. of Illinois at Chicago, and is a principal consultant at SolarVision Consulting.
Elmira Ryabova received her PhD in Physics at Ioffe Physical Technical Institute, and is Sr. Technologist at SolarVision Consulting.