Storage to the fullest

The word battery storage unleashes a new wave of energy in the solar industry in Germany at the moment. There has been no holding back ever since it became clear that there is now a subsidy for such systems. There is hardly a company that does not want to also design, plan, build or install such systems – depending on where they happen to be in the output chain. Moreover, in the meantime the production costs for solar electricity for roof installations are also clearly lower than the price of electricity for households.
This makes private power consumption of solar electricity in Germany very attractive instead of supplying it to the grid – particularly since the feed-in tariff is constantly declining. If storage systems are inexpensive enough, then they can contribute to clearly increasing the share of private power consumption.
All of that was reason enough for pv magazine to research a new overview of storage systems that can be used in single-family and smaller multi-family homes and commercial enterprises.
These systems are usually available as complete systems “off the rack.” In order to review them we sent a questionnaire to manufacturers and suppliers, including wholesale dealers. It contained more than 100 questions with regard to technical specifications. This number is necessary in order to get within reach of what the market urgently requires: namely, transparency. If one compares the brochures provided by the manufacturers, the differences are hardly discernible for the majority of prospective customers. Admittedly, the systems and their interaction are amazingly complex and can hardly be represented in brief. How then is a buyer supposed to decide? Or how should an installer give advice?
38 suppliers completed our questionnaire and provided the details of 176 systems; among them are all of the relevant – let us say – basic systems (see table starting on p. 78). Wholesale dealers integrate them with different batteries – and in part also with solar inverters – into complete systems so there are a number of different combinations. Depending on the application, various specifications may make sense for the solar inverter output, the charging and discharging capacities, the phase nature, the battery type and the energy management system. And, naturally, the price. This was indicated by the suppliers for approximately 100 systems and we were also able to research it.
When it comes to the phase nature in particular, Germany represents a special case in international terms. The three phases of the three-phase network run up to the service line so in several households devices are attached to different current phases. In many other countries, however, the house mains are operated with only one phase.
The regulation in Germany makes connection more complicated because one also has to consider the phase on which a particular device is to be connected to the solar installation and the storage system. With a connected load of 4.6 kVA (roughly 4.6 kW) a solar plant must feed into the grid in several phases in any case. For this reason there are also several three-phase inverters and storage systems, which under certain circumstances represent an advantage for the electricity mains.
What is interesting is to ask those suppliers who offer several systems about the differences. RWE Effizienz, for example, offers both the Solar Battery from Prosol Invest and the Engion Family from Varta.

Size as an argument

It is not as though there were not already enough technical differences between the various systems. However, what is crucial is sometimes another, even apparently banal parameter: “Several installers gave us the feedback that the system from Varta is too big for some applications,” says Carsten Welge, who is responsible for storage systems at RWE Effizienz. At 1.80 meters in overall height plus the 30 centimeters of required free space, it does not fit into some cellar rooms. The Solar Battery is smaller.
E3DC, one of the pioneer companies in the storage business, also offers two systems. In one it combines its storage technology with a Solutronic solar inverter, and with an inverter from Kaco in the other one. The difference is that the equipment with the Kaco inverter only has one MPP tracker and a higher input voltage range. Thus which equipment is the better fit depends on the solar generator.
The input voltages also play a role for Kostal in defining the solar inverter power output of their storage unit. According to the market overview it is suitable for outputs of 3.6 to 10 kilowatts – for no other device is a whole range indicated. The lower limit is due to the fact that the voltage of a conventional solar plant then just suffices. However, the device works with greater efficiency if it is not operated at the lower limit of the capacity range.
Service is also becoming a very important distinguishing feature. “For three months we have been struggling to get a defective device replaced,” says Jan van der Walle, an installer from Mölln in northern Germany. The device keeps indicating that the storage capacity is full. And another one is acting up because the cable to the current sensor, which measures how much output the battery system must make available, is now approximately 80 meters long instead of the 30 indicated in the manual.
But what should he do when the network connection point in his agricultural enterprise is so far away from the barn with the solar system. The length that can be bridged is different for each type of device.
The subsidy could also result in differentiations (see box on p. 71). Since mid-April manufacturers are busy with getting their systems eligible for subsidies. One of the important points is the current value warranty over seven years. It only has to apply to the batteries. For the manufacturers this represents a simple risk assessment.
Several of them have said on the quiet that the systems will simply become somewhat more expensive as a result. That can mean that customers must then decide whether to purchase the variant eligible for subsidies or equipment that cannot be subsidized. However, other manufacturers assure that such additional costs will be covered otherwise. The subsidy also requires that manufacturers submit a safety concept. Yet this is still a somewhat innocuous demand because these concepts are not evaluated. The point is to check and be able to make manufacturers responsible in cases of damage.

Throttling for the subsidy

Anyone who aims to install a subsidized system should obtain information on how manufacturers can provide for the required 60% throttling. The respective systems may only supply up to 60% of the output of the solar generator to the grid so that grids are relieved. “For very large battery capacities that is trivial,”says Martin Rothert, who is responsible for product management in the Off-Grid Solutions Unit at SMA, where storage systems are managed. “But with small systems that only works with a forecast of the expected irradiation and the load course in the respective household.” The reason being that a system may not charge the storage unit too early as it would then not be able to accommodate any more energy once it has to be throttled.
And it may not charge too late because otherwise the storage will be empty when the electricity is needed for private power consumption. Thus on cloudy days the battery must be charged earlier, and on sunny days later. One third of the devices in our overview are able to take solar forecasts into consideration.
There are also significant differences with regard to dimensioning the individual values: the rated output of the solar inverter, the battery storage capacity and the maximum discharging capacity. Tritec, for example, offers a 15 kilowatt inverter with 15 kilowatts of discharging power at only 5.5 to 11 kilowatt hours of storage capacity. The German power supply offers the reverse case: high battery capacity at a relatively low discharging power. The discharging power is important when a system goes into standalone operation – this then limits the maximum possible consumption – and if one wants to calculate how large the share of private power consumption will be.
The graphic “Individually created household load profiles” (see p. 74) shows the output of conventional consumers and it will probably surprise many that the household dishwasher accounts for nearly three kilowatts.

DC and AC coupling

However, the principle distinction between the systems first depends on where the battery and the battery charge controller are coupled. This has a great influence on both the flexibility of the systems and their overall efficiency.
For overall efficiency the entire chain must be taken into consideration: the efficiency with which the produced solar electricity is used in order to charge the battery, and the efficiency with which the energy stored in the battery can be supplied to the house mains. In DC-coupled systems (DC: direct current) the battery is connected to the DC intermediate electric circuit of the inverter (see graphic “DC and AC storage systems”, p. 70). Thus in these systems two less transformation stages are required compared to AC systems (AC: alternating current) for the solar generator-battery-consumer chain.
In these the battery is positioned directly on the alternating current house mains. It runs independently of the solar installation, except that the control unit has to use the yield values.
###MARGINALIE_BEGIN###

###MARGINALIE_END###
###MARGINALIE_BEGIN###

###MARGINALIE_END###
###MARGINALIE_BEGIN###
Two scenarios for what storage systems should cost. Based on the graphic showing the “Degree of Autarky”, a 50% degree of autarky is achieved with the shown here. Any income and expenditures are converted to prices for 2013 in accordance with the required 2% interest charge. The calculation on the left is conservative as it assumes a very moderate increase in household electricity costs because the maintenance costs were calculated at €225 per year, because a replacement investment for the battery in the amount of €400 per kilowatt-hour was assumed for the year 2023. If these parameters are modified, then more can be invested in order to operate a storage system with a return of 2% (right column).
###MARGINALIE_END###
Transparency on the market still leaves a lot to be desired. A number of manufacturers are unfamiliar with their efficiencies or choose not to disclose them. In the case of those who have indicated them the maximum efficiencies of the solar generator-battery-consumer chain are between 86 and 95% in the case of DC-coupled devices; the latter is the value for the Sunny Boy 5000 Smart Energy, which SMA aims to launch on the market this year. The values for the AC-coupled devices lie between 83 and 89%, whereby it was assumed that they are combined with a solar generator with 97% efficiency in order to make the efficiencies comparable with those of the DC-coupled devices. In this case the solar battery from Prosol Invest takes the lead. As one of the first systems on the market it has already been sold 800 times, according to information provided by the company. The relevance of efficiency, however, is not as great as in the case of solar plants where a percentage point has a direct effect on revenue. In the case of battery storage systems it is at least just as important how well the control unit functions if the system is to limit its maximum feed-in of 60%. In addition, most manufacturers indicate maximum efficiencies. Precisely in the case of storage systems what is important is how efficient they are when they operate at only low capacity. One indication is provided by the difference with regard to the European efficiency rating which, by way of analogy with inverters, is the weighted average of efficiencies with different capacities.
In the case of one manufacturer who provides this information, the European efficiency rating is 10% below the maximum, which suggests poor performance apart from the input power where the efficiency is at maximum. In the case of another the European efficiency rating is only one percentage point below the maximum; which demonstrates that even with storage systems it is possible to work efficiently over a broad capacity range.
It is important to note that these values do not take into account that losses also occur even in the batteries themselves. For lead batteries the available information indicates an efficiency of between 80 and 90%, and between 90 and 98% for lithium batteries. For a household that consumes 4,000 kilowatt hours of electricity per year and is able to cover approximately half of its annual consumption through the storage system (50% degree of autarky), the relevance can be estimated.
In this case ten percentage points in efficiency account for approximately 120 kilowatt hours of additional losses for that share of the energy that is intermediately stored, thus around €40 per year. Since financing is difficult as a whole, this does play a role and reduces the share of the budget that is available for the storage system by approximately one fifth (see table “Profitability calculation” on the left).

Lead versus lithium

A battle of opinions continues to be waged with regard to this question. Apart from the different degree of efficiency, there are several other differences between these battery types. For example, there are questions with regard to safety (see article, pp. 84 – 85). With lithium batteries, more precautions have to be taken in order to prevent them from burning under unfortunate circumstances. Although lead batteries do not burn that easily, small quantities of hydrogen may develop during use, which in turn makes demands on the room ventilation.
An essential difference is in the number of cycles. Lead batteries fulfill fewer cycles before their capacity drops to 80% of the nominal value. They get empty faster, but they cost less. For lead the manufacturers indicate values of between 1,600 and 3,000 cycles, for lithium between 2,600 and 15,000.
However, these cycle lives should be taken with a pinch of salt. “Lead batteries are sensitive,” notes SMA expert Martin Rothert. The company manufactures systems for both battery types. The weakness of lead batteries is revealed already by taking a look at the data sheets. If they are operated at 35 instead of 25 degrees, then the number of cycles already drops by 30%. “It also becomes problematic if Two scenarios for what storage systems should cost. Based on the graphic showing the “Autarky degree”, a 50% autarky degree is achieved with the example shown here. Any income and expenditures are converted to prices for 2013 in accordance with the required 2% interest charge. The calculation on the left is conservative as it assumes a very moderate increase in household electricity costs because the maintenance costs are calculated at €225 per year, as a replacement investment for the battery in the amount of €400 per kilowatt-hour was assumed for the year 2023. If these parameters are modified, then more can be invested in order to operate a storage system with a return of 2% (right column).they are totally discharged or overcharged,” points out Rothert. Thus, management of the charging process is very important.
A household in Germany requires approximately 250 cycles per year because the solar plant generates only a little electricity from November to February and therefore the storage system can be put to sleep in the winter. Due to the terms of the Renewable Energy Resources Act [EEG] most people still calculate using a required lifespan of 20 years. Accordingly, approximately 5,000 cycles are necessary in this case. If the battery does not last that long, then it has to be replaced or one simply accepts the lesser capacity. In addition, batteries also fail by merely standing around. In the case of lead batteries the manufacturers indicate a calendar life span of between 8 and 15 years, for lithium batteries between 10 and 25 years.
There are also big differences within both battery types. In the case of lead, most companies use lead gel batteries. The company Deutsche Energieversorgung uses lead-acid storage batteries. They are more inexpensive because production is more automated. The price is also at the lower end in the overview as well. On the other hand, the maintenance costs are higher because water must be refilled about once a year; a task that must be performed by the installer. In addition, a pump is installed in the switchgear cabinet that stirs the acid every six weeks.
According to plant manager Stephan Riss, this takes place automatically without any action being required by the operator. Such batteries are used otherwise in forklift trucks or for the emergency power supply in hospitals. In order to assuage the fears of customers that the battery may not last 20 years, the company offers the possibility of purchasing a spare battery for €1,000, along with the initial purchase of the system, which is then supplied after 10 or 11 years if required.

Upgradeability

With its lithium storage system, which is also sold by RWE and Baywa r.e., Varta advertises that the entire battery does not have to be exchanged when the capacity diminishes. Instead it is possible to reequip individual battery modules that have a capacity of approximately 0.46 kilowatt hours.
Since each of these modules contains its own electronics, the technology does not even have to be the same as that of the originally purchased modules – probably an important feature in ten years. However, if one believes the cycle lives indicated by most manufacturers, then the batteries will last in any case. But modularity is not only an advantage if the batteries fail, but also if the operator wants to extend the capacity. In the case of a number of manufacturers the batteries cannot be exchanged in part because they are connected in parallel or in series, and because then the weakest element would determine the total output. Among several manufacturers whose battery modules can be individually exchanged, there are some that are larger than those from Varta.
Returns and recycling are also in their infancy. Although it is clear that battery manufacturers and importers are obliged to participate in the GRS common collection scheme, most companies that do not come from the battery business neither know who collects them nor where.

Connection: Handling the phases

When it came to handling the current phases there was still great confusion even nine months ago. Nonetheless, the discussion appears to be nearing its end. In Germany the forum Mains Technology/Mains Operation in the Association for Electrical, Electronic and Information Technologies (FNN) is responsible for directives regarding mains connections. A network of experts formed last November published a technical reference in June on connecting storage systems. Anyone who plans to take advantage of the storage system subsidy will have to take this directive into consideration if the grid operator does not already insist on it anyway. The manufacturers in the market overview were also much better informed about the topic than just nine months ago.
Handling the phases can certainly be quite relevant when it comes to accounting for private power consumption. The electricity mains have the three phases L1, L2 and L3, in which the alternating voltage is respectively shifted by a third of a period. If a single-phase battery storage system is attached to current phase L1 and a stove plate to L2, then the single-phase battery system does not supply the electricity on the right phase. In principle, however, on phase L1 it is able to feed into the mains precisely the capacity which the stove draws from the mains on L2. If so-called phase balancing meters are installed, then the electricity meter stops in this case.
The battery system thus serves its purpose: The electricity consumption of the stove is nominally covered by the storage system. But in order to do this, the capacity of the storage system must be regulated in such a way that the net current over all three phases is zero or as small as possible. In the meantime this is possible with nearly all of the specified storage systems, of which 104 operate in single-phase mode.
This question equally applies to the 71 systems that feed in symmetrically on three phases. All of these systems thus produce a so-called unbalanced load when in operation. This is the designation used when different amounts of energy are fed in or consumed on the phases. Only asymmetrical three-phase feeding storage systems can provide precisely the capacity on each phase that consumers require. Nevertheless, this is possible with 32 systems. It can also be done, for example, with the Engion Family from Varta. “We currently regulate symmetrically on the net current because this is not yet demanded by the EVU,” writes the company in this regard.
Judging by what is being discussed in the FNN experts network, all variants will be presumably permitted. It looks as though the battery storage system and the solar plant in combination will have to comply with the customary unbalanced load condition of 4.6 kVA between phases. However, the way things stand at the moment, a transition period of one year will apply. Battery storage systems that are installed in this period may comply with the unbalanced load condition independently of the solar plant; thus the unbalanced load may then amount to 9.2 kVA altogether. The transition period makes sense because the actual requirement is only possible with additional measuring technology because the electricity meters installed by the grid operators do not allow for perfect phase evaluation. But with the technical reference from FNN this discussion will continue – although at a slower pace. In the future it may make sense to combine directives for production plants and consumers as storage systems go beyond these boundaries.

Network separation: Standalone

There are still systems that go offline in storage mode. This includes, among others, the Solar Battery from Prosol Invest, which is one of the most widely used systems.
If they operate in single-phase mode, then they connect the three household mains phases in standalone storage mode.