Co-location: what are the costs and benefits of DC-coupling?

Last week, we explored the operation of Great Britain’s largest co-located battery - at Whitelee Wind Farm. But what about batteries co-located with solar? Currently, the majority are AC-coupled - but could DC-coupling actually be more beneficial?

Wendel explores the costs and benefits of DC-coupling.

What is DC-coupling?

DC-coupling refers to a co-located battery and solar farm being connected behind a shared inverter - where power is in direct current (DC).

Both assets operate in DC - and require an inverter to convert electricity to/from the alternating current (AC) in which the electricity grid operates.

Pro: DC-coupling a battery and solar farm eliminates the need for more than one inverter. This means less duplication of equipment - and should reduce capital costs.

Con: However, it's not that simple. Different equipment is required in the form of a DC-DC converter. This ensures that the battery and the solar are operating at the same voltage levels. This, alongside reduced cost efficiencies elsewhere, can erode any cost benefits from DC-coupling.

Therefore, the preference for AC-coupling versus DC-coupling may come down to the operational benefits of each. For DC-coupling, the key to this benefit is in oversizing the solar farm - and the value of the subsequent “clipped” energy.

What is oversizing?

Oversizing means installing solar panels whose capacity is greater than the capacity of the inverter. The inverter limits the power that the solar farm can export to the grid. So, why do it?

Reductions in the costs of solar panels mean they are now one of the cheaper elements of a solar project.
Oversizing means the solar farm will have a wider generation profile - allowing it to export more power when it’s most valuable.
It’s quite rare for solar panels to generate at full power - this will only occur during the middle of the day, on the sunniest days.

The solar farm will sometimes generate more power, in DC, than it can convert to AC and export to the grid. We say that this power lost at the inverter has been “clipped”.

DC-coupling a battery with the solar farm allows you to charge the battery - for free - from the clipped energy (which would otherwise be lost), and store it to be discharged later.

What is the value of DC-coupling for a battery?

The capture and shifting of clipped energy can provide a lot of value for a battery.

For example, on 26th April, prices in the wholesale market meant batteries were able to make good money.

First off, what value could a normal battery have achieved on this day?

A battery trading two cycles could’ve charged at £86/MWh overnight, and sold that power for £136/MWh in the morning.
It could’ve then charged again at £98/MWh in the middle of the day, and sold that power for £150/MWh in the evening.
Taking efficiency losses into account, this would’ve delivered £76/MW.

But what about a DC-coupled battery?

With a solar farm generating excess power during the middle of the day, the second cycle would’ve been achieved with a charging cost of £0/MWh.
This means the battery could’ve made £188/MW for the day - a 147% increase on what the standalone battery could’ve earned.

How much clipped energy is actually available for the battery?

So, how do different oversizing ratios affect the volume of potential clipped energy? And how much of that energy can a battery then capture?

Clipped volumes remain low until oversizing approaches 150%.
A one-hour battery is able to capture all of the would-be clipped energy up to an oversizing of 140%.
Beyond this point, the system starts to run out of capacity on some days - but it’s still able to capture over half of all clipped energy at very high oversizing ratios.
For a two-hour battery, the longer duration enables it to capture even more clipped energy.

But what do these numbers mean in practice?

A solar farm oversized by 150% will see around 5% of its generation lost to clipping each year.
By installing a one-hour battery, 90% of this energy can be captured - and discharged to the grid once the sun has started setting.

These volumes are low compared to the normal operation of a battery - capturing this spilled energy would be enough to deliver 50 cycles per year.

What is the value of this energy?

The clipped energy captured by a DC-coupled battery is free. Therefore, the value depends entirely on the price it can be sold for.

This can be good for building a business case, as it reduces the exposure of the battery to intraday price volatility. The increase in gas prices over the last two years also benefits DC-coupled systems - the upwards shift in power prices gives batteries a higher price to sell at.

Are there any other benefits?

Charging from solar achieves a higher efficiency than charging from the grid.
Low-voltage harvesting also allows the battery to charge from solar energy that would otherwise be lost.

What are the downsides of DC-coupling?

This may seem like great news - but there is currently one major downside of DC-coupling:

It is extremely difficult for a DC-coupled battery to deliver frequency response - especially the ultra-accurate response required for Dynamic Containment.

Why?

The power from the battery being exported to the grid is combined with the power from the solar. Even when the solar is not generating, this creates a lot of noise.
This means a DC-coupled battery might even be unable to deliver frequency response overnight.
This could be avoided if the output of the battery were able to be measured using a DC meter - but this is currently not allowed (according to the rules for providing frequency response).

What would this cost in revenues?

In 2022, Dynamic Containment was responsible for 63% of battery energy storage revenues - in real terms, this meant that Dynamic Containment was worth around £100k/MW last year to the average battery energy storage system. A DC-coupled battery, unable to provide frequency response, would have lost out significantly.

However, since saturation, prices have fallen significantly. In fact, in 2023 prices for Dynamic Containment have averaged just £6.80/MW/hr - equivalent to earnings of £20k/MW.

Therefore, the premium that frequency response earns over a pure trading strategy has narrowed significantly - and the uplift from selling otherwise clipped energy can help make up for those lost frequency response earnings.

With additional value from trading clipped solar energy, a two-hour DC-coupled system can completely offset the lost frequency response revenue.
A one-hour DC-coupled battery can expect around 75% of the revenues that an AC-coupled battery would receive from Dynamic Containment.

What does the future hold?

Currently in Great Britain, the majority of co-located battery energy storage systems are AC-coupled. This is because they can be metered and operated independently - which often makes financing these projects easier. And, to date, the available revenue from frequency response has been extremely valuable - but, as we’ve seen, that value is decreasing.

Cirencester Hybrid Solar Farm (a 24 MWp site with a 10 MW battery attached, owned by Warrington Borough Council) is Great Britain’s first DC-coupled grid-scale battery - and there are others in development:

For example, in late 2022, JBM Solar sold two projects - totaling 105 MWp of solar, with 65 MW of battery energy storage - to Vantage RE. These projects are expected to be operational this year. (JBM Solar was later acquired by RWE.)

Until now, decisions on whether to develop an AC-coupled or DC-coupled battery have been driven by preference.

However, as DC-coupled batteries become operational (i.e. start to enter markets), we will get a better view of the actual financial costs and benefits of each.

Want more?

Interested in reading more about co-location? Check out our other Phase articles: