With our recent change of EV we now have access to DC chargers, and even better free charging for up to 30 minutes, and preferential per-kWh pricing after that. So far, we have used them twice, but the experiences were very different.
Up until now, the only public charging network we had access to was the slower level 2 chargers. We also had experience with Tesla’s supercharger network following the rental of Tesla vehicles both here in the US and in the UK.
The first test of Electrify America was at our local Target store, where, although the battery was low, I still had the backup option of plugging in at home overnight. This was very much a test run for me to work out the steps needed to get the charger working.
How does the Electrify America experience compare to our previous experiences? Well, compared to the level 2 chargers, and I will restrict this to the ones that have a fee, the experience was similar. Unlike our experiences at ChargePoint and Blink however, we used the app to start charging instead of tapping an RFID card. The Tesla experience, however, is much simpler: all you need to do is plug in and the car does the rest.
The charging itself was also slower than I had expected. In the almost 30 minutes the car was connected to the charger, and with a battery level at just 17% initially, I was expecting to see close to the advertised 150 kW charging rate, at least initially. Unfortunately, the charging started at only 84 kW (and this was at the end of a roughly 80 mile drive, mainly at freeway speeds, so the battery should have been warm too), and rose to 90 kW.
Final statistics on the charge were 45 kWh added in just under 30 minutes, confirming that 90 kW charge rate. For somebody used to using 7 kW public AC chargers, this was a massive improvement.
The next experience was a more remote charging session, some 70 miles from home and with a battery level that would not get us home. Unfortunately, this was not as smooth sailing. There were four units at the location, three with dual 150 kW CCS connectors and one with 50kW CHaDEMO and CCS connectors. I tried the first unit, and was presented with a charging start error, after waiting 30 seconds or so.
I tried a few times, then gave up and moved to another of the 150 kW units. Same thing. The charger was failing to communicate with the car (and note this is a car from the VW group of companies, so there should be no problem with VW Group chargers talking to it).
After failing on this one, I called the toll free number printed on the charger to see whether they could help. Very quickly I was connected to an agent who verified some information, and then proceeded to reset the charger for me. After it rebooted, we tried again, but the same happened.
Finally, she suggested that maybe I should try one of the other two units. Once again I moved the car, and reconnected it. This time, albeit after a relatively long delay, the vehicle and charger did manage to communicate, and I was able to initiate the charging process from the app.
After thanking the support person, who insisted on staying on the line as I moved the car and tried charger number three, I took a look at the information on the charger screen, hoping to see what speed I was getting from the charger. Instead, I was presented with the statement that I was 1 minute away from 80%.
Clearly, when you start at 17%, it will take longer than 1 minute to reach 80% state of charge, especially on a 150 kW charger. Also, I noticed that the line drawn on the bar, which looks like it was intended to represent 80% state of charge, appears to be labeled 28%.
While it charged, we went to a nearby Chinese restaurant to order dinner (to eat in a nearby park). I was monitoring the charge from the car’s charging app, and noticed there that the rate was being reported as over 150 kW initially, and dropped down to 120 kW as the car approached 80%, just over 20 minutes after starting. At 78%, I walked back to the car.
When I arrived there, I tapped the charger screen to see what it was reporting for the charging rate:
Oddly although the total energy delivered and time charging matched the app, the charging rate was being reported as half the amount (59 kW compared to 120 kW). Simple math here will confirm that 61 kWh in 25 minutes is not possible at a rate of 59 kW (in fact, it requires an average rate of 146 kW).
Poor Platform Choice
When the very helpful support agent rebooted the system, I got to watch the system reboot. Behind the charger is apparently a regular PC running a version of Microsoft Windows. That is, in my professional opinion as somebody who has been working with embedded software systems for almost 3 decades, a poor choice.
Firstly, it was never designed for this kind of application, and, as photos on the internet will confirm, Windows based systems that are used in embedded applications frequently fail, and Windows really isn’t designed for remote administration. Worse still, with a touch screen sitting right there, I may well have been able to interrupt the startup sequence and stop the Electrify America UI from loading.
Secondly, Windows suffers from a particularly poor security record. Whether you blame that on poor software, poor processes or just being the mot common platform and therefore the one hackers concentrate on, the last thing I want as the designer of an embedded system, like a rapid charger for EVs, is an operating system that is commonly hacked.
Finally, Windows is really not the best environment to be writing code that has to manage hardware and interact with other systems.
I was slightly shocked that in an application as critical as EV chargers, they had picked Windows as the OS.
When they work they perform well. I am disappointed that I can’t tell where the 350 kW ones are located from the Electrify America app or website (mainly because I would like to try one to see how fast that really is). I will certainly be using them again, but, given the issues on our second visit, I am also going to make sure I understand where there are alternatives that are perhaps a little more reliable as a backup plan.
It is also clear, that like the in-vehicle software in the VW group cars, the charger software needs some work to make sure it is showing correct information. Given the app knew the correct rate, the data must have been available to get that displayed correctly on the charger screen too.