Sunrise over the sea

EV adventuring with resilience

Road trips are the most demanding EV use case currently in Australia, especially to remote destinations. However, my planning showed that they were still quite doable. Did those plans survive contact with reality?


In short, it was a pleasure to drive an EV long distances and the only inconvenience was faulty public charging infrastructure, in which I include vehicle-to-infrastructure interaction. Despite this, there was enough redundancy in the network and our vehicle’s capacity that faulty chargers were only an inconvenience, rather than a showstopper.

I’m still confident heading off on road trips, but I’ve shifted thinking of charger faults from the exception to the norm, and shifted planning accordingly. Range anxiety isn’t a thing; but charger anxiety may be!

It’s something I’m willing to deal with to access destinations like this.

Plan vs actual

The plan was to stick closely to the ideal–in terms of quickest arrival time–charging profile proposed by ABRP. The outbound trip from home to a remote destination is illustrated in the chart below, which shows the cumulative distance travelled over time, and the estimated range of the vehicle at each point (proportional to battery state of charge). The ideal range estimate is compared to the minimum actual range estimate we had at any point in the actual road trip.

Chart showing ideal EV road trip charging profile. Two short high speed charges and running down the battery/range to nearly zero

The trip starts just after 08.30 and ends just after 16.00. Ideally, we start with a full charge and full range, make a small charge about 90 mins into the journey and then deplete range/charge to nearly zero over a longer stretch, before making a large charge about 90 mins from the destination, which leaves us with nearly 2/3 range/charge at the remote destination.

With only an hour for stops in this plan, I’d already noted that we would probably stop longer and more often on the journey, for comfort and recreation, so there’s no compromise compared to an ICE vehicle in this plan.

The actual profile is illustrated in the chart below, with the same actual range estimate minimum line for comparison to the ideal plan.

Chart showing actual EV road trip charging profile. A long wait and four medium speed charges keeping the the battery/range above two thirds of maximum

It’s apparent that we charged more often and maintained a higher minimum range/charge than the ideal case. We charged four times in total and maintained range/charge above two thirds of maximum at all times. Our average speed was also a little below plan. These two reasons are partly why it took us nearly three hours longer to reach the destination, just after 19.00 in the actual case, but another reason is that we spent an hour waiting for a functioning charger at our first planned stop. This is where reality started deviating from the plan!

State of chargers

Let’s first look at the functional chargers and then come back to the non-functional chargers. At each stop, we charged at 50kW chargers, rather than 250kW+ chargers in the ideal plan. We were also charging at a higher SOC (typically 70-90%) than the ideal profile, which meant lower power (as low as 5kW) towards the end of each charge. This accounts for actual charging taking over 2 hours, as compared to 1 hour in the ideal plan.

50kW is fine by me, as an average rate, at a vehicle efficiency of about 5km/kWh. That represents a range replenishment rate of 250km/h. If chargers are near Points of Interest (POIs) such as food, parks, etc, and wait time is minimal, 50kW chargers make for more leisurely road tripping, with 1 hour break for every 2.5 – 3 hours driving, but they don’t make it infeasible. As above, we may have stopped for 2 hours total on the 7.5 hour journey anyway, even if we had infinite range.

Photograph of electrical wires in Tokyo against a blue sky and sun

The bigger issue was the non-functional chargers. Our vehicle completely failed to charge at one of the major charging networks, which is the one we’d planned to use at both stops in the ideal case. Although we’d tested this network successfully before starting the trip, and we tried all three plugs multiple times at both of the two locations at which we’d planned to charge, we were only able to obtain about 2kWh total from this network, with about 90% of sessions dropping when the power hit about 18kW a few seconds after initiation of the session.

This meant we had to fall back to other networks. After our nearly total charge failure at our first stop, I wanted to charge as much and as soon as possible so that we were as resilient as possible to further charging issues. As you can see from the charts, we could have easily gone on to the second on third stop before we dropped anywhere near or below 10% of range/charge, but we would have had really limited options if we’d reached a non-functional charger at a remote township in that state.

Another charging network had a location just 3km down the road from the first stop. As it turned out, two of the four chargers at this location were non-functional, so we had to wait longer to access a functional charger at this location too. This reinforced my revised plan to maintain as much charge as we could, and after waiting an hour to that point, we spent another 40 minutes charging over 95% again before hitting the road once more.

From that point on, the alternative network’s chargers worked well and we had no further issues of functionality, but as the various EV apps we were using sometimes reported different statuses to what we observing in reality, we remained wary nonetheless!

Note that on the return trip we only needed to charge twice. We picked functional chargers from the outbound trip. We also had progressively more charging options the more dense the population, and could sacrifice resilience (reserve range) for efficiency (shorter charges). We arrived home with 16% SOC.

A real positive of more and longer stops was meeting other EV drivers. It was fun to connect and learn about their travels and their vehicles, and get accurate intel about chargers up and down the route.

Vehicle efficiency

Where reality beat the planning was in the vehicle efficiency. Estimating about 4.3km/kWh as the base case, we were able to get closer to 6km/kWh in city driving, about 4.5-5.0km/kWh on highway and gently undulating country roads and about 4.4km/kWh on windy and hilly country roads.

Switchbacks up a mountainside

Reducing speed from 100km/h to 95km/h also seemed to have a positive benefit of up to 0.5km/kWh but we didn’t test this under controlled conditions.

This was at the base case described in EV adventuring, as we weren’t dealing in a major way with range-reducing external gear, net elevation gain, or problematic weather or road conditions.

Resilient planning

What I’ve taken from this experience is that an ideal charging plan for an EV road trip may need multiple back-up options, as of now, in Australia.

Our rule of thumb was to always maintain enough charge to go on to the next charger, in case a targeted charger was non-functional (regardless of what status may be reported in any given app). This requirement becomes more demanding the more sparse the charging network, such as in rural and remote areas.

It becomes more demanding again in the case of an out-and-back trip to a remote destination, which is why we aimed to reach our campsite with 67% SOC, in case the final charger was non-functional on the return trip (which would have required us to travel 330km between charges). I think this is also good practice in remote areas where there might be a risk of natural disasters, as the Australian summer of 2019/2020 demonstrated.

Heatmap of charger site value based on road network analysis

ABRP provides a number of options to improve resilience of a charging plan, such as exploring alternative chargers, and specifying a minimum SOC before charging is triggered, which I’ll explore more fully on my next road trip. If these don’t meet my needs, I’ll think further about how to fill the gaps. The most resilient plan may look a lot different (and indeed have no overlap) with the ideal charging plan. It’s also possible the charging network could be intentionally designed to improve the resilience of more road trips.

For now, it’s back to less demanding EV use cases for a while (and more cycling as an alternative to driving too!), and dreaming of the next road trip.



, , ,