Just add batteries: considerations for hybrid systems

There’s more to consider than just the brand or size when adding storage to a solar system. Damien Moyse and Nick Carrazzo highlight some of the issues to consider in a field with ever-evolving technology.

There are multiple ways that batteries can be added to an existing or new solar PV system. These different configurations will influence the system’s capabilities so it’s important to carefully consider the approach you take. This article covers the most common approaches currently available in Australia, but note that technology and options are developing rapidly so we will be updating this advice regularly.

The majority of solar PV systems currently installed in Australia are unlikely to be ‘battery-ready’—an existing solar customer cannot simply purchase a lead-acid, lithium ion, flow or sodium battery and have it retrofitted to their existing system.

The solar panels can be retained, of course, but an additional or replacement inverter and charging components will likely be needed to charge and use the batteries.

One approach (DC coupling) involves replacing the existing grid-interactive inverter with a new hybrid inverter; such inverters can both control charging of the battery and conversion of electricity from DC to the AC required for household use. As a cheaper alternative, in a fairly recent development, the replacement of the grid-interactive inverter can be avoided through fitting a DC to DC converter between the solar array and the battery bank—thereby negating the need to replace the existing grid-interactive inverter.

A second approach (AC coupling) requires installation of a second battery-dedicated hybrid inverter (with integral charger controller), with the existing grid-interactive inverter retained.

As such, almost all the new battery products currently on the Australian market are either sold with a new inverter (some as part of an integrated ‘all-in-one’ storage unit and some with the inverter separate from the battery) or require an inverter to be purchased separately.

Thus, most existing solar customers will need to replace their existing grid-interactive solar inverter, add a second inverter or add a DC to DC converter to their system. Which approach is taken depends on whether the system uses AC or DC coupling and the capabilities required of the system. Coupling refers to where within the system the batteries are connected.

DC coupling

DC coupling involves siting the battery on the DC side of (or indeed plugging it directly into) the solar inverter (see Figure 1).

The wires connecting to the battery on this side of the solar inverter carry DC electricity, so the solar-generated electricity can be used to charge the battery prior to it being converted to AC for household consumption or export to the grid.

This approach means that the solar inverter will need to be a ‘hybrid’ inverter, which has a higher level of functionality than a traditional grid-interactive ‘string’ inverter. A hybrid inverter converts the DC power from the panels and battery to AC and also takes care of the required battery control and switching functions. Given this increased functionality, hybrid inverters are more expensive than traditional string inverters.

Another key component of the system is the isolation switch. This enables the system to operate in the event of a power outage on the electricity grid. Standard solar and solar-battery systems do not have this capability.

As an alternative to replacing the grid-interactive inverter, a DC to DC converter, such as the GoodWe BP series, can be fitted between the solar array and the battery bank (note that the GoodWe requires a 48 V battery). This enables the battery to charge directly from the solar array. When the battery discharges the DC to DC converter steps the battery voltage back up to a nominal voltage that meets the requirements of the grid-interactive inverter. The main disadvantage of using this DC to DC converter approach is that the system cannot run without the grid, so battery backup isn’t available.

Figure 1: DC coupling configuration. DC solar is fed through the hybrid inverter for charging of the battery or conversion to AC for use by the household. The hybrid inverter also controls use of the DC electricity from the battery. In a fairly recent development, an alternative approach to DC coupling involves using a standard grid-interactive solar inverter with a DC to DC converter.

AC coupling

AC coupling involves siting the battery on the AC side (the grid or household side) of the solar inverter. This means the wires connecting the battery to the solar system are 240 VAC (see Figure 2).

Given all batteries operate in DC, this AC coupling requires a second battery-dedicated inverter which includes the ability to charge the batteries, adding to the cost of the overall system. This inverter needs to:

• convert the battery’s DC to a household/grid compatible AC
• convert the solar inverter’s AC output to DC in order to charge the battery
• control the charging so that the battery is not damaged
• charge the battery with solar electricity or electricity from the grid, e.g. during cheaper off-peak times
• only discharge the battery when the household/site requires it—and not back to the grid, unless you have a system enabled to sell electricity from your batteries to the wholesale energy market via a retailer, such as offered by Reposit Power.

Once again, it is important to note the need for an isolation switch to enable operation in the event of a power outage on the grid. This is not included in a standard AC coupled system.

A situation where AC coupling is unavoidable is where the solar array is fitted with microinverters, which are tiny grid-interactive inverters attached to each panel. The panels themselves are effectively AC output devices, so DC coupling isn’t possible.

How easy it is to add storage will depend on the microinverters used. For example, Enphase microinverter-based systems can have the Enphase AC battery added to the system without a separate hybrid inverter, as each battery module has its own Enphase inverter and battery charger built in, and all of the Enphase devices can communicate with each other for simple system setup.

Other batteries could be used with Enphase or other microinverter brands, but the specifics of how the added components communicate with the existing ones will need to be addressed on a system-by-system basis.

Figure 2: AC coupling configuration. DC solar is converted to AC by the solar inverter for use in the household or fed through the battery-dedicated inverter-charger for charging of the batteries. A communications link is needed between the solar inverter and the inverter-charger to allow the PV production to be controlled when the system is running in backup mode.

Considerations for existing PV owners

For consumers with existing grid-connected solar PV considering retrofitting batteries to their system, and for whom the economics of the project are of primary importance, ATA would suggest waiting for storage prices to drop and the market in Australia to consolidate around the most optimal technologies and systems.

For consumers with existing solar PV wanting to retrofit batteries now, irrespective of the economics, they will need to consider the following.

First, they will need to consider whether they want their retrofit battery system to provide power in a blackout. If so, then an isolation switch (costing $250 to $450 + installation) will be required along with a new hybrid inverter and communications system that can handle system operation in the event of a power outage. Note that not all hybrid inverters can operate in a grid-islanded situation.

Should back-up power not be desired, a DC to DC converter may be an option, allowing the retention of the existing grid-interactive inverter.

Even for a system involving a DC to DC converter, it will still be difficult to establish a retrofit project for less than $10,000 installed in 2016, even for a system involving a relatively small battery.

Next, existing PV owners need to consider whether they want the system to be able to charge from the grid. Some new inverters (for example the SolarEdge inverter) have an AC to DC charger and can facilitate battery charging from the grid. Grid charging facilitates the use of cheaper electricity (e.g. overnight off-peak tariffs) as a supplement to charging from solar on low solar resource days. However, optimising this approach requires the use of weather and electricity consumption forecasting, functionalities that aren’t available without additional software to control battery charging (e.g. through the Reposit Power software, costing about $800).

Considerations for homes without existing PV

For consumers considering investment in a new solar and storage system, the above choices regarding backup and grid charging equally apply. The main challenge for these consumers will be choosing an inverter for their new system.

Given 2016 battery prices are still relatively high, many prospective solar+battery system owners are waiting for prices to come down. In the meantime, however, many will be keen to purchase and install a standard grid-connected solar PV system to start realising the benefits of such a system, prior to installing storage.

If the solar customer later decides to add storage and this requires replacement of the originally purchased grid-interactive inverter, this replacement would likely occur part way through the usable life of the solar inverter. The lost value is offset to an extent by the initial saving from installing a standard solar inverter as compared to a hybrid inverter (a saving in the order of $1500).

Alternatively, should this customer decide to purchase a more expensive hybrid inverter initially as part of their grid-connected solar PV system, there is a risk that this hybrid inverter may be obsolete or unnecessary (due to an increasing prevalence of DC to DC converters) within three to four years. It may not be able to support new and improved battery technologies available at the time of adding storage. In addition, the functionality of battery systems is continuously improving and legacy hybrid inverters may not provide functionality required by the market at a future point in time.

Although both decisions involve pros and cons, ATA would advise that the risk of technology obsolescence is too great at this point in time to recommend the latter option. In theory this could lead to a freezing of the solar market over the next few years, but in reality consumers will continue to make decisions on the basis of other reasons besides economic rationality and system optimisation (e.g. early adopter, peer competition) as experienced in the solar PV market in the late 2000s/early 2010s.

Daytime load

Another important factor in the economics of storage is the consumer’s daytime load, with the most beneficial load profile contrasting that of the most beneficial load shape of solar-only customers.

For consumers not on premium feed-in tariffs, solar PV without storage offers the greatest potential benefit where significant electricity consumption occurs during the daytime, during solar generation hours.

By contrast, a load shape with lower daytime and higher night-time consumption will realise greater benefits from a solar+battery system. A consistently high daytime load leads to solar generation being used directly on-site, leaving insufficient excess solar left to charge a battery for use in the evening and overnight.

Charge/discharge rate

The discharge rate is the time, usually expressed in hours or parts of an hour, it takes to discharge a battery before it is fully discharged. Expressed as the ‘C’ rate, this is the theoretical capacity of the battery when charged or discharged at the consistent rate over time.

The capacity of some batteries (specifically lead-acid technologies) is reduced if the battery is discharged over a shorter period. In the case of lead-acid, C10 to C20—discharging over 10 to 20 hours—tends to be the highest level of discharge without significantly reducing the capacity of the battery.

At higher discharge rates, the energy output capacity is reduced as well as the asset life (which is expressed as an absolute number of charge cycles before the battery fails or suffers significant depletion of capacity). This is an important consideration for households or businesses who may wish to access the energy stored in a battery relatively quickly (e.g. during a daytime or evening peak).

Newer technologies (e.g. lithium ion batteries) do not suffer from these charge/discharge constraints in the same way, improving their effective operation. Hence the discharge rate of lithium batteries can comfortably exceed C1, with the capacity of other components potentially being the limiting factor, not the battery.

Battery utilisation

Battery utilisation is another important factor in the overall economics of storage. Battery utilisation is the average daily discharge of the battery (on an annual basis) as a percentage of its usable capacity.

Although a battery may have a certain usable capacity (e.g. 10 kWh), this does not necessarily mean that the entire storage capacity will be used on any given day, or indeed consistently across many days of the year.

Battery utilisation is a function of the ability of a certain-sized solar system (or the grid in AC coupling) to fully charge the battery to its rated capacity, as well as the consumption profile of the individual customer. Both of these involve significant variability as weather patterns change and the behaviour of the household changes over days, weeks and seasons.

Standards

There are several existing Australian standards relating to battery design and installation within buildings. Most relate to traditional lead-acid technology and do not address recent product innovations including the increasing prevalence of lithium-based chemistries and hybrid (grid-connected) solar+battery installations.

Given a lack of standards specific to emerging battery technologies and configurations, in April 2016, the Clean Energy Council (CEC) and the Australian Energy Storage Council (AESC) separately released interim guidelines for battery installation and safety. Both note these guidelines should be considered stop-gap measures until formal standards are finalised.

Both the CEC and AESC note that the Australian standards do not address the following innovations in the residential storage market:

• the increased energy density of lithium-based cells (in particular) which significantly exceeds that of lead-acid and nickel cells
• improvements in grid connection equipment (e.g. inverters)
• packaged or ‘all-in-one’ systems, which combine the battery, inverter and other control equipment into a single unit with pre-engineered connections.

In response, Standards Australia commenced public consultation on the development of Australia’s first comprehensive set of industry standards for battery storage in May 2016. Standards Australia is working with the COAG Energy Council to develop new standards and support the safe and efficient uptake of new storage technology in Australia.

Storage alternatives

The current capital cost of energy storage in batteries is likely to remain too high in the short to medium term (i.e. prior to 2020 and potentially after) for many existing and new solar customers.

With many higher feed-in tariffs nearing the end of their legislated lives around the country, some level of renewable energy storage will be beneficial for solar homes to maximise the value of their solar-generated electricity. There are other ways to store renewable energy, which are cheaper and just as effective as using a battery.

Thermal energy storage (for example, as heat in water) is a concept that has been around for a long time. Electric heating and storage of water for domestic use is broadly done in two ways :

• by using a traditional electric storage hot water (ESHW) system, which uses single or multiple resistive electric elements in a tank to heat and store water
• using a heat pump, which involves the compression and expansion of a refrigerant through a heat exchanger to extract heat from ambient air (which creates multiple units of heat output for each unit of electricity used), heating water stored in a tank.

Both of these systems use electricity as an input to the system and can be powered directly from solar PV, provided the home or business is configured for net metering, the ESHW or heat pump is connected to the main electrical circuit (i.e. not a separate, dedicated circuit established for off-peak hot water), and the ESHW or heat pump operates during the day (i.e. when the solar system is generating electricity).

ESHW and heat pump systems offer the potential for existing (and new) solar customers to maximise the usage of their solar-generated electricity, without the need to invest in as yet expensive chemical energy storage in batteries.

Summary

Energy storage systems are more complex both technically and economically than solar PV systems.

System configuration and battery management both have a significant impact on the overall consumer experience. A consumer’s location and consumption pattern also makes a significant difference to their personal economic outcome.

Products in the current Australian storage market do not as yet provide the full range of potential options for system configuration and battery management. This means the most optimal strategy for any individual household may not yet be available for them to take up.

All of this complexity will make decision-making difficult for consumers, particularly during the early phase of evolution in this market. Consumers will need assistance with simple, accurate and independent information and advice to help them navigate the Australian storage market as it evolves.

This article was first published in Issue 137 (Oct-Dec 2016) of Renew magazine. Issue 137 is an off-grid and hybrid special.