Chapter 3:

A way to cut bills?

It is important to note that it is not universally accepted that a drop in voltage will reduce energy consumption and therefore bills.

Keith Bell, Scottish Power chair for future energy systems at the University of Strathclyde and director of the UK Energy Research Centre, says: “The point about lowering voltages in order to save energy, I think it’s quite tenuous.”

At the heart of this disagreement is the mix of devices being used in homes and businesses at any particular moment. There are three main types of electrical load that all respond to changes in voltage in different ways: constant power, constant current, and constant impedance.

Constant current and constant impedance devices both use power at a lower rate if voltage is reduced.

For constant current devices, this drop is directly proportional to the reduction in voltage. However, for constant impedance devices, voltage is directly proportional to current, meaning the reduction in power is even larger.

As the name suggests, there is no reduction in power consumption from constant power devices, which instead maintain this level by drawing more current when voltage is lowered.

Bell points out that, in the case of constant current and constant impedance devices, a reduction in instantaneous power usage, while potentially helpful, does not necessarily mean a reduction in energy usage over time.

Taking a kettle as an example: “It’s heating up the water to a certain temperature. It needs a certain amount of energy to do that... If you reduce the power, it's just going to take longer to do that. No savings in terms of energy.”

For electrical home heating, Bell says there may be “a bit of a benefit” in that “you’re coming up more slowly, it doesn’t overshoot and you’re not getting that wastage. But my guess is it’s pretty minimal.” In some cases, he says, a reduction in instantaneous power may also prevent undersized heating systems from overcoming thermal losses from homes.

And when it comes to constant power devices, Bell says a reduction in voltage could actually be detrimental given that resistive losses are proportional to the square of current: “If you reduce the voltage, the current’s going up and what you should be doing to reduce network losses is to make the voltage as high as you can.”

“This is why I'm sceptical about the broad claim that it's going to make any significant difference to energy use,” he remarks.

Bell additionally has concerns that a general reduction in voltage levels could leave less footroom for the National Energy System Operator (NESO) to work with in case of an emergency: “Reduction in voltage is a kind of emergency measure that the system operator has in its toolbox..."

If a large generator trips and NESO’s normal reserves are unable to fill the gap, “one of their emergency measures would be to call the DNOs and get them to reduce the voltages to reduce the power.

“But if you've already done the reduction of voltage, there's no more footroom to offer it as an emergency service.”

Andy Howard, innovation programme manager at ENW, says Bell’s scepticism towards the claimed energy savings is not borne out by the results of their trials. He says their data shows that although there is a degree of time-shifting when voltage is reduced, there is also a genuine reduction in energy usage over time.

Responding to the suggestion that voltage reduction could increase resistive losses on the network, Howard says any increase in current draw from constant power devices, which include consumers electronics such as laptops and TVs, is “masked and overridden” by the reduction in current from constant impedance devices.

Roger Hey, now a member of the Energy Geeks thinktank, says a reduction in network voltage can also make devices such as laptop chargers, which use voltage choppers to drop the voltage to lower levels anyway, operate more efficiently when this step down is smaller: “The less it chops it down, the less hot it gets and therefore the more efficient it is.”

'The response is measurable'

Howard says they now have a pretty clear idea of what kind of response they are going to get in aggregate: “Different connected equipment operates in different ways and gives a different level of response, or no response, but as a whole, we can see that response and we can predict that response. And we know it varies by time of day, and it varies site by site because the nature of that mix is different site by site.”

This relationship between voltage and power can be calculated using a mathematical formula and expressed as a Kp value, ranging between 0 and 2.

A value of 2 denotes a strong relationship between voltage and power as seen with constant impedance devices, whilst a value of 0 indicates no link between voltage and power as seen with constant power devices. A value of 1 represents a 1:1 relationship between voltage and power, as seen with constant current devices.

Howard says the Kp value measured across its network is typically at, or slightly above, 1, with a 1% reduction in voltage leading to a 1% to 1.2% reduction in instantaneous power. He is hesitant to give an equivalent figure for the reduction in energy usage over time, but when pushed says the responses to prolonged reductions in voltage applied over weeks or months as part of its Smart Street trial tended towards a Kp value of 1.

This chart was published by ENW in 2022 and shows the Kp values measured by the company at just one of its substations, averaged for each of the 48 half-hour settlement periods per day over the summer and winter seasons. The Kp values ranged from about 1.18 to 1.29 over the summer, and from 1.24 to 1.3 over the winter.

Howard says the mix of devices customers are using is changing over time but “the core response is not going to change significantly”.

“When we look down through the network through millions, possibly hundreds of millions, of devices across that spectrum, we do get a response, and that response is measurable and repeatable.”

He expresses frustration that “despite all of the academic work, and all of the real-world experience that we've got; and monitoring and measurements that we've got; and the fact that NESO is paying us multi-million pounds a year for the services that we're providing... we do continue to get this question to a degree".

Hey remarks: “Now, every time you talk about turning the voltage down, it seems to have this almost like irrational reaction from some people in the industry." He says there is a feeling that by lowering voltage, consumers are somehow getting short-changed: “It's like a Mars bar getting smaller.”

Howard says very few customers will notice any difference from voltage reduction: “I’m old. I can remember filament lamp bulbs. I can remember cathode ray tube TV screens. Voltage reduction was visibly quite noticeable on those devices. Your lights would dip. Your picture would reduce in size if you got a low voltage on the network.

“Modern LCD screens, modern LED lighting, just doesn’t give you that response. They work well to a much lower voltage and then they stop working. It’s a much more digital response.”

He says most appliances work at a wider range than the current statutory limits, but “the problem is identifying those bits of equipment that don’t.”

Despite affecting 2 million of their 2.4 million customers, Howard says their manipulation of voltage through their CLASS system has only resulted in a few hundred queries. He says they have generally concerned instances in which the customer was at the end of a power line where voltages were already a bit low, and they have then installed a new power-hungry appliance: “A customer who might have seen a dip once every six months, or regularly but only in the winter period, might start to see those the rest of the time.”

In these cases, he says the network was probably always going to need reinforcement eventually and “it’s just slightly moving the boundary of when reinforcement is going to be required.”

Howard says there have also been issues with some commercial customers, for example, those with standby generators that are designed to kick in when there is a loss of supply. He says customers may have skimped on the equipment used to detect the loss of supply, or the equipment may have been incorrectly configured to trigger whenever voltages move outside of their historical range: “For any particular customer, they may have only experienced plus or minus 1 or 2%... And then we come along and utilise the full 6%.”

He recalls several customers whose standby generators were turning on and off and who thought there was a problem with the network, even though there was no loss of supply and voltages hadn’t gone outside of the regulatory limits. Their view was that the generator was working perfectly fine previously and so it must be the network’s fault.

He says avoiding these issues can be difficult as customers aren’t generally interested in engaging with their local network until they have a problem: “But as soon as their equipment stops working, they become very, very interested because that basically means their production line has gone down and that's costing them.”