An Introduction to AC and DC Coupled Solar Battery Systems

1 Reply 11159 Views

I believe it is important to know the difference between AC and DC coupled solar battery systems, so I will try and describe it in simple terms without complicating diagrams.




I will start by clarifying the difference between power and energy because we very commonly talk about the “power system” and power poles.  Power is a measure of the rate at which energy can be transferred per unit of time, whereas energy is the use or storage of power over time.  Another way to consider it is power can’t be stored but energy can.  The unit of power is a watt and in simple electrical terms, power is measured by multiplying current and voltage.  A 5 kW inverter is capable of power output of 5,000 watts which, at 250 volts, means it can produce a current of 20 amps.  If that inverter outputs 5 kW for 2 hours, that would be 10 kWh of energy.




Solar panels produce direct current (DC), generally in the range of 200 to 500 volts, which then has to be converted to alternating current (AC) at 240 volts to power the household, charge the solar battery, or to export to the grid.  There are many articles on the web, such as How do solar panels work.


In a DC coupled system, the solar panels are connected directly to the solar battery via a DC to DC battery charger.  The voltage that solar panels produce is not constant, so the charger uses some clever electronics called MPPT (maximum power point tracking) to extract the absolute maximum energy from the panels under all conditions.  The battery then powers an inverter, commonly called a hybrid inverter because it can charge the battery from the grid as well.  This inverter generates the 240 volt alternating current (AC) which powers the home and exports any excess to the grid at the feed-in tariff rate.


In the event of a prolonged grid outage, the whole system can be isolated from the grid, and the solar panels will continue to charge the battery.  The inverter will continue to provide as much energy as can be generated each day from the sun to power up the backup circuits, or critical load panel, and to charge the battery.  If, during an outage, the solar panels produce more than the household and the battery require, then solar production will be automatically stopped to protect the battery from overcharging, but will restart when required.


In an AC coupled system, the solar panels connect directly to an MPPT solar (grid-tied) inverter, and the resultant 240 volt AC current is used to charge the battery, power the home or export to the grid.  Another inverter, called a battery inverter, connected directly to the battery, is used to power the household when solar energy is not available, such as at night time.


The first important thing to know about inverters designed to connect to the grid is that they must be able to synchronise their output to both the grid voltage and the grid frequency of 50 cycles per second (50 Hz).  The second vital point is that if the grid is down, they must stop working in order to protect any power workers who might be working on the grid.  To do this, solar inverters look for a 240 volt 50 Hz AC grid voltage and frequency on their output side before switching on.  A DC coupled system can just isolate itself from the grid and keep working because the panels are connected via DC to the battery charger and battery.  However, an AC coupled system will only produce solar energy to charge the battery if the solar inverter can see the required 240 volts and 50 Hz frequency at its output.  This is normally provided by the grid, but if the grid is out, this voltage must come from the battery inverter.  The problem arises when the battery runs down and shuts off to protect itself.  The solar inverter then can’t operate because it can’t see a 50 Hz voltage with which to synchronise from either the grid, or the battery inverter, and so it can’t restart until the grid becomes live again.  Therefore, in an AC coupled system with backup circuitry, all you get is most of what’s in the battery at the time of the outage, and from then on, your solar is useless.  You have all those panels and no way to utilise them.


The obvious question then is why are AC coupled systems so popular?  The reasons are all pretty well related to cost, as you might expect.


One of the most common scenarios is adding batteries to an already installed solar system.  In this case an AC coupled system just requires connecting to the output of the existing solar inverter, and to the grid.  From the installer’s point of view, this is safe and simple because there is no need to disturb the existing solar inverter and the solar panel wiring behind that.  Unfortunately many existing installations could be non-compliant because standards have changed over time, or because of previous non-compliant workmanship.  In a DC coupled installation to an existing solar system, as in our case, the installer is required to remove the existing solar inverter, and re-route the solar panel cables into the cabinet of the DC coupled system.  In our case, during this process, it was discovered that our installation was non-compliant, and considerable work was needed to correct that situation.  I give great recognition to CME who did this work at no cost to us.


An existing solar inverter will have its own MPPT controller built in, so removing the solar inverter means that the DC system has to have an MPPT charge controller built in, which adds to the cost.  Also, a simple one way DC to AC inverter is cheaper to make than a hybrid inverter that is designed to generate AC from the battery, and also convert AC to DC to charge the battery from the grid.


The net result of this is that AC coupled systems are cheaper to make and cheaper to install.  They are also slightly more efficient if most of your electricity is used during the day when solar energy is available. 


In an AC coupled system, during the daytime, the solar panel DC current has to pass once through the solar inverter to be converted to AC, and then is used in-house, or to charge the battery, or is exported to the grid, with no battery charge/discharge losses involved.   At night, however, the energy used is retrieved from the battery via the battery inverter, so this solar generated electricity has passed through two inverters, an AC-DC battery charger, and includes battery charge/discharge losses.   


In a DC coupled system, day or night, the solar panel DC current has passed through a relatively efficient DC-DC MPPT charge controller, once through the hybrid inverter, and also includes battery charge/discharge losses.


The efficiency of an inverter is a measure of the initial input power that is available at the output, after losses inherent in the inverter’s DC to AC conversion process.  The losses are the difference between the input power and the output power, normally expressed as a percentage of the input.  If a process is 90% efficient, that means the losses are 10%.


Battery round trip efficiency is sometimes simply defined based on the losses associated with charging and discharging a battery.  I am using a slightly more comprehensive version of round trip efficiency to include the storage losses associated with charging and discharging the battery, but also including the inverter losses associated with converting the stored battery energy into usable 240 volt AC energy for use in-house and for exporting to the grid.


In the case of either solar panels only with no battery, or an AC coupled battery system, round trip losses include the solar inverter when it converts DC solar panel current into 240 volts AC.  It is not normally possible to measure the input DC energy to a solar inverter, so solar production is normally measured as the output of the solar inverter.  This means that the losses associated with the inverter are not included in any calculations based on solar production.  If we want to include solar inverter losses in any comparisons, we are forced to refer to the manufacturer’s specification sheet, bearing in mind that it will likely be an optimum figure from their perspective, and not able to be achieved in the real world.  In the case of solar panels without a battery, the round trip losses are the solar inverter losses only.


In the case of a DC coupled battery system, where there is no solar inverter, solar production is measured normally at the battery, and therefore does not include the MPPT charge controller losses.  Again, if we want to include all losses, we must refer to the manufacturer’s specifications for that part of the losses.


For the AGL Virtual Power Plant (VPP), the total solar energy produced is obtained from Solar Command, as defined by AGL below, and when we subtract from it the energy exported to the grid for feed-in, we are left with the total solar energy available to the household.


The total solar energy consumed is also obtained from Solar Command, as defined by AGL below, and when we subtract the energy imported from the grid, we are left with the total solar energy consumed by the household.


The difference between these two numbers is the solar energy lost through charging and discharging the battery, and losses in the battery or hybrid inverter, where DC current from the battery is converted to 240 volt AC for household consumption.  I have called the total solar production minus the losses net solar production.


Produced energy is the amount of energy that is generated from the solar power system as measured by the monitoring device.


Total consumption is the gross amount of electricity consumed from both the solar power system and from the grid as measured by the monitoring device.


Another source of loss is the delay involved in an inverter responding to a change in household load demand.  As soon as the load is switched on, assuming the battery can power the load, the inverter will start.  However, there will likely be a finite period before the inverter is up to full load, and during that delay, power will be drawn from the grid.  To this end, the modern electronic inverters will probably respond faster than a transformer based inverter.  I am not able to provide any engineering analysis to quantify this effect, nor can I find anything on the web.  Rapidly switching loads, however, are likely to be more susceptible to this effect.


Please see my profile if you’d like to learn more about other articles I have written on solar energy and my participation in AGL’s Virtual Power Plant solar battery storage program.


0 Replies 4367 Views

Thank you Richard,


Another very informative and helpful article.