PHOTOVOLTAIC SYSTEMS

chapter 9
From : Renewable and Efficient Electric Power Systems
Gilbert M. Masters
Stanford University
STAND-ALONE PV SYSTEMS
• The design process for stand-alone systems begins with an estimate of the
loads.
• trade-offs are made between more expensive, but more efficient, appliances
and devices in exchange for fewer PVs and batteries.
• Lifestyle adjustments need to be considered in which some loads are treated as
essentials that must be provided for, and others are luxuries to be used only
when conditions allow.
• A key decision involves whether to use all dc loads to avoid the inefficiencies
associated with inverters, or whether the convenience of an all ac system is
worth the extra cost, or perhaps a combination of the two is best.
• Another important decision is whether to include a generator back-up system
and, if so, what fraction of the load it will have to supply.
• The simplest of systems will incorporate only devices that run directly on dc.
• Power In the simplest case, energy (watt-hours or kilowatt hours) is just the
product of some nominal power rating of the device multiplied by the hours that
it is in use.
• An especially important consideration for household electronic devices—TVs,
VCRs, computers, portable phones, and so on—is the power consumed when the
device is in its standby or charging modes. Many devices, such as TVs, use power
even while they are turned off since some circuits remain energized awaiting the
turn-on signal from the remote.
• Consumer electronics now account for about 10% of all U.S. residential electricity,
conclude that almost two-thirds of this energy occurs when these devices are not
actually being used (Rosen and Meier, 2000).
• Major appliances and shop tools have another complication caused by the surge
of power required to start their electric motors. While that large initial spike
doesn’t add much to the energy used by a motor, it has important implications
for sizing inverters, wires, fuses, and other ancillary electrical components in the
system.
 1-Estimating the Load
• In Table 9.10, consumer electronics category show power while they are being
used (active) and power consumed the rest of the time (standby), both of which
must be considered when determining energy consumption.
• Refrigerators are also unusual since they are always turned on, but their power
demand varies throughout the day. This means that they are likely to overstate
actual demand in someone’s home—perhaps by as much as 20%.
• Power on some of the devices shows maximum rating tend to overstate power
since they are meant to describe maximum demand rather than the likely
average.
• Some nameplates provide only ampere and voltage; and while it is tempting to
multiply the two to get power, this can also be an overestimate since it ignores
the phase angle, or power factor, between current and voltage.
2-The Inverter and the System Voltage
• To figure out how much power the batteries must supply, the calculation needs to be modified to
account for losses in the dc-to-ac inverter.
• Inverter’s efficiency is a function of the magnitude of the load it happens to be supplying at that
particular instant.
• Most inverters now operate at around 90% efficiency over most of their range.
• When no load is present, a good inverter will power down to less than 1 watt of standby power while it
runs uses on the order of 5–20 W of its own.
• Swapping out the ac refrigerator in the above example for a dc one reduces the
total dc load that the batteries have to supply by about 15%.
• This can translate into a 15% reduction in the size and cost of the photovoltaic
array as well as the batteries.
• Moreover, the inverter itself can be smaller and cheaper since it doesn’t have to
supply as much ac power.
• Additional cost of a dc refrigerator plus the added complexity and cost of wiring a
house to provide for some ac and some dc loads.
• An economic analysis would be needed to determine the right decision.
 System Voltage
• Inverters are specified by their dc input voltage as well as by their ac output
voltage, continuous power handling capability, and the amount of surge
power they can supply for brief periods of time.
• The inverter’s dc input voltage, which is the same as the voltage of the
battery bank and the PV array, is called the system voltage.
• The system voltage is usually 12 V, 24 V, or 48 V.
• Higher voltages need less current, making it easier to minimize wire losses.
• On the other hand, higher voltage means more batteries wired in series, which impacts the number of batteries that may be
needed to supply the load.
• The most important specification for an inverter is the amount of ac power that it can supply on a continuous basis.
• But it is also critically important that the inverter be able to supply surges of current that occur when electric motors are
started.
Implementation of a DC distribution system could result in a more efficient use of
electrical energy because of the following reasons:
• For many electrical appliances a DC voltage is required. Some appliances are
connected to the utility grid by means of a switching mode power supply. Other
appliances are provided with a transformer to reduce the volt-age level before the
AC/DC conversion. With both systems on-line losses occur. Only with the latter,
stand-by losses occur as well.
• Some appliances (e.g. washing machine) are fitted with an AC drive. An AC drive is
usually fed with a DC volt-age that is obtained through rectification of the grid
voltage. The conversion from AC to DC and to AC again involves some losses.
• In a DC distribution system the many smaller rectifiers can be replaced by one
large, more efficient rectifier.
• There is no reactive power in DC systems. Reactive power results in increased
losses in AC systems due to a larger current magnitude for an equal amount of
transferred power.
• In the predictable future, an average residence might contain one or more local
generators. These generators have to be interfaced with the power grid as
efficiently as possible. A DC system could prove useful in this respect.
• There  are  no frequency  stability  and   reactive  power   issues,  
• and   no  skin   effect  and   ac   losses.
DC electronic appliances

• Desktop PCs require +12 Vdc, –12 Vdc, and +5 Vdc at several
amperes.
• Laptops need 16 Vdc for charging battery packs, also at several
amperes.
• Digital TVs and monitors start with a conversion to 24 Vdc, then up-
convert (boost) for the backlight and down-convert (buck) for the
logic.
• Printers and peripherals may require 28 Vdc to 32 Vdc,
• modems 12 Vdc,
• portable stereo systems 9Vdc, and
• common small battery chargers (for cell phones, digital cameras, and
MP3 play -ers) 5 Vdc, at varying currents.
• A problem is developing one (or several) universal dc voltage and
current specification(s) that can power all devices in a typical residence.
• Kitchens are usually equipped with a couple of high-power appliances
for cooking. Almost all cooking appliances may be considered as
resistive loads, such as stoves, toasters, electric rice cookers, electric
kettles, and coffee makers, and they can be energized by DC. Other
equipment, such as microwave ovens, dishwashers and refrigerators,
are DC friendly appliances, and may be energized directly by DC in the
future. So, DC voltage is quite suitable for kitchens.
• Since laundry rooms are usually equipped with high power appliances,
such as washing machines, clothes driers and irons, a separate
distribution circuit needs to be provided. It is not difficult and may be
more efficient to adapt these appliances to a DC supply in the future.
3-Batteries
• Stand-alone systems obviously need some method to store energy gathered during good times to be able to
use it during the bad.
• Various energy storing technologies are possible e.g., batteries, super capacitors, flywheels, compressed air
etc.
• Batteries provide surges of current that are much higher than the instantaneous current available from the
array.
• as well as the inherent and automatic property of controlling the output voltage of the array so that loads
receive voltages that are within their own range of acceptability.
Battery Storage Capacity
• Energy storage in a battery is typically given in units of
amp-hours (Ah) at some nominal voltage and at some
specified discharge rate.
• A lead-acid battery, for example, has a nominal voltage of
2 V per cell (e.g., 6 cells for a 12-V battery), and
manufacturers typically specify the amp-hour capacity at a
discharge rate that would drain the battery down to 1.75
V over a specified period of time at a temperature of 25◦C.
• everything having to do with battery storage capacity is
specified in amp-hours rather than watt-hours.
• the amp-hour capacity depends on the rate at which
current is withdrawn. Rapid draw-down of a battery
results in lower Ah capacity, while long discharge times
result in higher Ah capacity.
• The amp-hour capacity of a battery is not only rate-dependent but also depends on
temperature.
• As shown in Fig. 9.42, battery capacity decreases dramatically in colder conditions. At −30◦C
(−22◦F), for example, a battery that is discharged at the C /20 rate will have only half of its
rated capacity. where the C refers to Ah of capacity and the 10 hours it would take to deplete
(C/10 = 200 Ah/10 h = 20 A).
• when discharged — mean that lead-acid batteries need to be well protected in cold climates.
• In fact, a rule-of-thumb estimate is that battery life is shortened by 50% for every 10◦C above
the optimum 25◦C operating temperature.
Coulomb Efficiency Instead of Energy Efficiency
• Batteries are described in terms of currents rather than voltage or energy.
• Battery efficiency is more easily expressed in terms of current efficiency than in terms of energy efficiency.
• The reason, of course, is that battery voltage is so ambiguous without specifying whether it is a “rest” voltage
measured some time after charging or discharging, a voltage during charging, or a voltage during discharge as
well as the state of charge of the battery, its temperature, age, and general condition.
• When a battery approaches full charge, its cell voltage gets high enough to electrolyze water, creating
hydrogen and oxygen gases that may be released.
• Gassing and charging losses can be minimized, by using a charge controller that has been designed to slow the
charging rate as the battery approaches its fully charged condition.
• Charge controllers also protect batteries from over or undercharging.
To help understand where that energy loss occurs, consider the simple Thevenin equivalent for a battery.
 Battery Sizing
• If good weather could be counted on, battery sizing might mean simply providing enough storage to carry the
load through the night and into the next day until the sun picks up the load once again.
• there are no set rules about how best to size battery storage. The key trade-off will be cost.
• the question arises as to which is better. The key difference between the two is the amount of current that
flows to deliver a given amount of power.
• Batteries in series have higher voltage and lower current, which means more manageable wire sizes without
excessive voltage and power losses, along with smaller fuses and switches, and slightly easier connections
between batteries.
• On the other hand, a storage system consisting of batteries in parallel is easy to expand, one battery at a
time. In series, a whole new string of batteries must be added to increase storage.
• The relationship between usable storage and nominal, rated storage (at C/20, 25◦C) is given by

• where MDOD stands for maximum depth of discharge (default: 0.8 for lead-acid, deep-discharge batteries, 0.25 for
auto SLI)
• T, DR stands for temperature and discharge-rate factor
From Fig. 9.46 at 95% availability and 3.1 peak sun hours in December, it looks like we need about 4.6 days
of storage.
 Blocking Diodes
• The system shown in Fig. 9.48a allows the battery to leak current back through the PV module at night, which raises the
question of whether it might be worthwhile to add a blocking diode as shown in Fig. 9.48b to prevent that nightly discharge.
Sizing the PV Array
• a good starting-point estimate for Ah delivered to the batteries: an area with 6 kWh/m2-day of
insolation is treated as if it has 6 h/day of 1-sun, 1-kW/m2 radiation. Then using the rated
current IR at 1-sun, times peak hours of sun, gives us amp-hours of current provided to the
batteries.
• It is common practice to apply a de-rating of about 10% to account for dirt and gradual aging of
the modules.
• The current delivered to the batteries needs to multiplied by the Coulomb efficiency
(Ahout/Ahin) to give Ah delivered from the batteries to the load:

• The only other thing to keep track of is the system voltage. For a 12-V system voltage and 12-V
modules, modules are added in parallel until sufficient Ah are provided for the load. For a 24-V
system voltage and 12-V modules, two modules in series are needed to provide the 24 V, and
then parallel strings of two series modules each are added to deliver the Ah needed by the
load. Or, if 24 V is the system voltage, it may make more sense to simply choose 24-V PV
modules rather than using two 12-V versions.