We get annoyed waiting for web pages to load, for the light to turn green and for the person who just can’t decide what to order. So, why do EV manufactures think half-hour “fast” charging times are acceptable? Granted, if you have a home charger, you only need to use public charging intermittently. But many people do not have off-road parking and must rely on public charging points. I doubt most of us would want to hang around waiting for our cars to charge after a grueling day’s work.

Charging is last great hurdle EVs must clear to become mainstream. Charging times should be comparable to filling up a gas tank. But what wattage of charger could do that? Is it even possible?

The Gas Pump

To start, we need a baseline. How fast do gas pumps transfer energy? Fuel pumps in the USA deliver fuel at 10 gallons per minute (0.631 liters per second). This rate set by the government, so you can blame all those wasted minutes filling up your car on ‘The Man’. If only ‘The Man’ let fuel flow faster…

Each liter of gasoline contains 34.7 Megajoules. If we multiply the flow rate of fuel (as stipulated by The Man) by the energy in each liter of gas, we find the power transferred by fuel pumps each second.

Gas pumps deliver energy at 21.92 megawatts (MW).  To put that into perspective, if you had solar panels producing 200W/m² you would need a solar farm 1km² to generate that much power! It is a 50^th of the capacity of a typical 1 GW nuclear reactor. These are not precisely fair comparisons because most of the energy released by the fuel in our engines is lost as heat, and fuel pumps are not in continuous use. However, they still give perspective on the magnitude of energy that’s contained in fossil fuel.

The Charger

For each unit of energy, EVs will travel further than gasoline vehicles. Therefore, to make an accurate comparison of gas pumps and chargers, we need to find out how much energy electric and gasoline vehicles use to travel a certain distance.

Firstly, let’s look at the premium end of the market. A Tesla Model S with a 100kWh battery uses 33kWh/100 miles. A 2017 Mercedes-Benz S550, uses 4.8 gallons/100 miles (18.2l/100 miles).

To convert 18.2l/100 miles into kWh, we multiply it by the energy density of gasoline, then divide by the number of seconds in an hour. This gives 175.4kWh/100 miles. So, we can see the Tesla uses 5.3 times less energy than the Mercedes to travel the same distance (based on EPA figures).

Filling the Mercedes at a fuel pump adds a certain amount of fuel, and thus range, each second; as does charging the Tesla. We know gas pumps deliver energy at 21.92MW. The Tesla goes 5.3 times the distance per kWh than the Mercedes. Therefore we need a charger 5.3 times less powerful to add range at the same rate, which would be 4.1MW.

The fastest car chargers available theoretically charge at a peak of 0.5MW, just over one-tenth of our required charging rate.

How Long Would it Take to Charge a 100kWh EV with 4.1MW?

To find that out, we convert the capacity of the battery pack from kWh to joules, by multiplying the capacity in kWh, by 3,600—the number seconds in an hour. Then we divide this number by 4.1MW, to give us the charge time, which is 87 seconds. Very reasonable; I wouldn’t wait a second longer.

The issue with this calculation is it changes with each vehicle you compare. A Toyota Yaris, which uses 2.9 gal/100mi, is more efficient than a Mercedes. Therefore, it can go further on a given amount of fuel and a Tesla Model S would need a 5.7MW charger to add range at an equivalent rate.

These calculations show that a direct comparison between chargers and gas pumps is not so straightforward, since it depends on the efficiency of both vehicles. Fortunately, there’s another way to approach this problem.

How Long Are we Willing to Wait for EVs to Charge?

In my not-too-scientific experiment, I found people typically spend 3-8 minutes at the gas station.

Methodology: Sit outside a gas station with a stopwatch and try to look as inconspicuous as possible.

It was clear that people spent most of the time inside the store paying and waiting, rather than standing the nozzle pumping away, which only took a couple of minutes. If people spend up to 8 minutes at the station, then merely accounting for filling time in our analysis is unfair, we should consider for time spent at the gas station.

Let us suppose for a minute that a new network of charging stations will be built with payment systems on the machines, so we will not have to go inside to pay. Thus, the 87 second charge time of an EV could be extended to total time spent at the station.

What’s So Interesting About 750kW?

I’m glad you asked.

A 100kWh car will have a range exceeding 300 miles, and at 750kW would charge in 8 minutes, which is the upper bound of the time people spent at gas stations. If you were in an emergency, rushing to the airport, for instance, you could add 150 miles of charge in 4 minutes.

To me, that sounds more than acceptable, but then again I’m rarely in a rush. Others might find waiting a whole 8 minutes for a car to charge egregious.

Is it even possible to charge at that rate?

I’ve been chatting with people at ABB, who say that charging technology is currently limited by EV motor voltages and contacts between the charger and the vehicle.

Motors over 1,000V in cars are impractical because of problems with electricity arcing, though I’m unsure if is this an unsolvable issue. However, someone reading this might have a solution. If you do, please share your thoughts in the comments.

This technology will no doubt continue to develop, and we will surely see charging rates become faster over time, but we also need EVs to be capable of accepting that power.

No announced car can charge this fast yet.

Can Battery Packs Charge at 500kW?

Yes, if you have enough cells in the vehicles.

Typically, vehicle battery packs are made up of many similar cells, with the charging rate of the pack dependent on their number and arrangement. Simply put, when building a battery pack, you need to optimize for voltage and capacity. Voltage is dependent on the number of batteries wired in series, while capacity is dependent on the number of in parallel. You can wire blocks of batteries in series in parallel and vice versa to produce the characteristics of the battery pack out want.

This is a very simplistic explanation, but further analysis extends beyond the scope of this article.

If you want to learn more about batteries I recommend reading this.

How Thick Will EV Charging Cables Be?

Two equations can enlighten us on this question:

Power = Current x Voltage

and

Resistance = Voltage / Current

More power with the same voltage requires a higher current. A higher current increases the resistance, and hence we must compensate by increasing the cable diameter.

There is a workaround for this: you split the battery pack and charge each module separately. So you could attach, for example, two 400kW chargers instead of a single 800kW charger, thereby allowing for multiple thinner cables which would be easier to handle. But, to do this, the vehicle would require multiple charging ports.

How Thick are We Talking?

This is dependent on a host of factors, including ambient temperature, permitted voltage drop across the wire, current and resistance. If the cables have good cooling systems, you can use a thinner wire.

Using this online calculator we can calculate cable thicknesses, though I must admit I’m unsure of the accuracy of this calculator in relation to this application.

800kW (800V, 1000A) - 500mm² (25 mm diameter)

400kW (800V 500A) -185mm² (14mm diameter)

If we wanted to deliver energy at 4MW (to fill an EV at the same rate as a gas pump), with current motor technology limited to 1000V we would need a current of 4,000A. I do not believe it’s possible to provide this much current through a cable suitable for public changers, but if anyone reading this has a counter-example, please share it in the comments!

Nevertheless, it is apparent that very high currents will require thick cables, which are not entirely unworkable, but they are unwieldy. As previously mentioned, one way around this is to use multiple thinner cables to charge various battery modules at lower currents, but cumulatively the same power. i.e. 400kW+400kW to charge at 800kW.

Why Does EV Charging Matter?

In cities, if you do not have a personal parking space, an EV will be a pain to own. Maybe you won’t care about waiting half an hour to fill up your car with a ‘fast’ charger, but I’d bet a lot of people will.

If we can charge an EV in 8 minutes or less, then the problem is solved and there will be no need for a network of small city chargers. We can have charging stations like gas stations which will deliver energy in a matter of minutes. If so, all these slow chargers that have been popping up around cities could be a huge waste of money.

Whether or not we can build charging stations which suck up megawatts of power is another question, one which I hope to answer in the coming weeks.

EV vs ICE – Summing Up

To add range to an electric car at a rate equivalent to a gas pump would require a charger with more than 4MW of power.

However, if we ask what is acceptable charging time for an EV, that changes things. I say about 8 minutes for a 100kWh battery pack, which would require a 750kW charger. You may consider this to be too long, but if we accept that much of the time spent in filling stations are paying for fuel and waiting, not just refilling, a case can be made that 4MW required to charge an 100kWh EV in 87 seconds is excessive.