By Paul Homewood
This is a follow up to yesterday’s post on Steve Broderick’s work on the impact on the grid of EV charging.
Perhaps the first thing to note is that he is not the young student doing his PhD, which I think we all imagined. To put it kindly (!) he looks nearly as old as me!
In fact he has oodles of experience in engineering, and I gather he undertook the research for a PhD largely because he wanted to devote more of his time on the topic, not just for the qualification.
Since 2017 he has continued researching the issue, and had this paper published in 2020:
Image may be NSFW.
Clik here to view.
Image may be NSFW.
Clik here to view.
His latest work backs up his original conclusions, that the LV network simply will not be able to handle large scale rollout of EVs.
His comment about heavier cars is relevant. I cannot comment on the harmonics issue, but I am sure plenty of readers will!
His latest figures reckon that EV charging could add 1820 kW for a 200 house system designed to supply 300 kW:
Image may be NSFW.
Clik here to view.
And he quotes 2012 costings of £62 bn for network reinforcement. You can double that for current prices, totting up to over £4000 per household.
Image may be NSFW.
Clik here to view.
One interesting conclusion is that Vehicle to Grid (V2G), that is draining power from EV batteries to power the grid, could make matters worse, as it would increase the charging time needed at night.
Heat Pumps
Finally, let’s take a broad look at the load requirements for heat pumps.
A few very simple assumptions:
- Average household gas consumption – 15000 KWh
- Average household electricity consumption replacing gas for heat pumps, hot water and cooking – 6000 KWh – most of this, say 5000 KWh will be used between October and March
- 5000 KWh over 6 months equals 28 KWh a day
- But during periods of extreme cold, this could peak at maybe 40 KWh, which = 1.66 kW
The above assumes that heat pumps are running on a continual basis through the day. It is quite likely, however, that DSR will lead to heat pumps being switched off at certain times of the day, thus increasing their consumption during other times, notably at night when EVs will also be charged.
As Broderick notes, most typical networks are rated at around 1.5 kW per house, so heat pumps will immediately lead to overload, even if there are no other appliances running. This obviously won’t be the case during day time and evening, and matters will be even worse at night when EVs are charging, even if there is a control system to share the load around.
It is important to point out of course that heat pumps are designed to work for long periods, as they only provide low heat. To expect them to adequately heat homes throughout the day on the basis of a few hours operation is pie in the sky.
For most houses therefore, estimated at 80% by Broderick, the LV network will not be able to support either EVs or heat pumps, never mind both together.
Timescales
Finally let’s deal with the problem of timescales. Broderick thinks that smart control could delay major spend for long periods.
I think this is not feasible. Since 2020, when the paper was published, the government’s plan to roll out EVs has become clear, with not only all new car sales to be electric by 2030, but also mandates for the vast majority to be so in years leading up to 2030.
Back in 2020, the plan was to ban ICEs only after 2040. What Broderick rightly saw as a problem for the distant future will soon be upon us.
Maybe smart control will be able to manage the problem for the next few years, but it certainly won’t when the majority of cars are electric, which won’t be long coming.
And therein lies the rub. To rip up roads, replace cables and upgrade substations will take years, probably decades. The job will have to start now, smart controls or not.