Why Charging Sucks — And What to Do About It

Darren Hau
Published in
6 min readMay 1, 2023


“Our culture’s obsession with innovation and hype has led us to neglect maintenance and maintainers.”

This is Lee Vinsel, assistant professor of science and technology studies at the Stevens Institute of Technology, speaking on the Freakonomics podcast episode “In Praise of Maintenance”. And today, as annual EV sales are poised to break the one million mark for the first time in the US, and auto companies compete to hitch more ponies to your wheels and place fancier tech onto your hyperscreens, we’re going to talk about one of the most unsexy things in the industry — charger maintenance.

What’s the problem?

Despite plenty of interest in electric vehicles, 47% of US adults say it’s unlikely they’ll buy an EV for their next car, with 80% of the public citing lack of charging infrastructure as their primary reason. According to a J.D. Power survey, “one out of every five respondents ended up not charging their vehicle after locating a public charger. And of those who didn’t charge, 72 percent indicated that it was due to the station malfunctioning or being out of service.” Even in the San Francisco Bay Area, arguably the birthplace of mainstream EVs, Berkeley and Stanford researchers found that more than a quarter of non-Tesla public DCFC stations were inoperable.

This obviously doesn’t have to be the case. Tesla’s Supercharger network is often held up as the gold standard of reliability and ease-of-use. In the J.D. Power survey, Tesla was the only DCFC operator with a satisfaction score above the average — meaning it’s singlehandedly pulling up everyone else.

Why is the 3rd party ecosystem performing so terribly? Having spent time on both sides of the aisle, allow me to offer some thoughts.

First of all, the non-Tesla ecosystem is extremely fragmented. When you charge at a Combined Charging System (CCS) station (the standard that all other automakers use, except for the vestige of Nissan Leafs on CHAdeMO), you’re effectively relying on a half dozen entities:

  1. The automaker — obviously the EV has to be hardware and software compatible
  2. The charge point operator (CPO), i.e. EVgo, Electrify America, Voltera, TeraWatt
  3. The payments provider that supplies the RFID and/or credit card reader
  4. The charger management system (CMS) provider — this enables the CPO to monitor and diagnose their network
  5. The charger manufacturer, i.e. ABB, Tritium, Zerova, ChargePoint
  6. All the tier 2 suppliers to the charger manufacturers for communication modules, charging cables and connectors, etc.

This fragmentation means problems become much more tedious to diagnose. The CPOs don’t typically staff up all of their technicians, instead relying on a network of service providers like Pearce Renewables, ChargerHelp!, and InCharge Energy. These service providers can use their own or others’ CMS dashboards to perform preliminary diagnosis, but these dashboards often only provide basic OCPP messages and charge session data. If technicians require more charger data, they have to ask the manufacturers for error logs, introducing additional delay.

Unfortunately, the charger manufacturers don’t have particularly strong software competencies. Most of them evolved out of building power electronics, and those modules are pretty robust, but many charging failures occur as a result of communication errors between the charger and vehicle, or buggy firmware in the charger screen (also known as the HMI), or poor integration with the communication or payments modules. As mentioned before, these manufacturers are often not directly sharing data streams with the service providers.

The fact that the manufacturers are integrating other suppliers’ parts also introduces more opportunities for failures to occur in the first place. Your power electronics may be bulletproof, but if your coolant system, controls, HMI screens, or backend communications fail — or if your charging cable/connector breaks because someone dropped it on the ground — all the customer will see is a non-functioning charger.

Finally, there are some inherent disadvantages to the CCS1 standard vs. Tesla’s connector design (which they’ve dubbed the North American Charging Standard, or NACS). The CCS1 connector is incredibly large and bulky compared to the Tesla connector. The larger size and number of pins increases the insertion force required, and makes it harder for a user to lug around. The moving latch on the CCS1 connector is prone to breaking, while the Tesla connector has no moving parts. And while a larger connector and cable might suggest that a CCS1 connector can handle higher currents and voltages, the Tesla connector actually can deliver higher currents while being compatible with both 500V and 1000V vehicle architectures.

How does Tesla do it?

The biggest differentiator with Tesla, of course, is that they are vertically integrated — they design and control the vehicle and the charger. This simplifies payments (just plug it in, no credit card or app required!), eliminates charger-to-vehicle communication hiccups (CharIN hosts the poorly-worded “Testivals” for other OEMs to test compatibility), and streamlines diagnosis. I still remember trying to diagnose a communication failure while working on the Tesla Supercharger team, and an engineer showing me a dashboard indicating that the capacitance of a particular signal line was just above a 6µF threshold.

Furthermore, this vertical integration means Tesla’s highly incentivized to fix issues quickly. There’s no one to blame but them if something goes wrong!

Tesla doesn’t have this problem.

Finally, Tesla simply had the time and resources to staff up a hyper-competent service team. As they scaled up the Supercharger network, they did the same with their human resources, subsidizing both with high-margin vehicles.

What to do for everyone else?

There’s obviously not much others can do about CCS1 vs. Tesla NACS unless all automakers and charger manufacturers decide to do a wholesale pivot. But there are plenty of straightforward, if not easy, steps to take to better position the 3rd party ecosystem to survive and thrive as Tesla opens up their network to non-Tesla vehicles.

First of all, train more technicians! I’m a big fan of ChargerHelp!’s mission to provide both upskilling and service. More broadly, I suspect what we’ve been seeing recently is a renewed appreciation for “blue-collar”, skilled labor over “white-collar” desk jobs. As tech, finance, and consulting firms lay off employees, we still face labor shortages of electricians, building contractors, welders, etc. Ironically, AI seems more likely to replace aspects of knowledge workers’ jobs than those that require interaction with the physical world. (Although I tried to get ChatGPT to write this article, and it’s not quite there yet.)

Second, service providers should build processes to capture institutional knowledge and drive continuous improvement. This one is pretty uncontroversial, and many companies are pursuing this.

Third, the industry needs to expect better designs from charger connector suppliers. The connector is the most abused part of a charger, and the part that is the most fragile, especially with that silly moving latch. Every time a cable/connector needs to be replaced, it can cost upwards of $3000. Why can’t connectors be made more robust, with better shock absorption, tighter tolerances, and lower cost? If accomplishing this requires joint investment in NRE, we should be doing this.

Fourth, charger manufacturers should share charger logs and raw data streams more directly with service providers. It serves no one to keep the root causes under wraps if end customers lose confidence in the overall experience. They won’t care if the problem is due to the CPO, service provider, or the manufacturer. Reducing the amount of time playing telephone will go a long way to resolving charger issues quickly.

The final point represents a common thread discussed above, and that is to establish true partnerships among all players in the ecosystem. While we can’t expect individual drivers to do so, commercial EV fleets, CPOs, service providers, and manufacturers should engage with a spirit of collaboration to improve reliability and utilization. After all, this is still a relatively immature industry, and a rising tide lifts all boats (or EVs?). Everyone has their own economic interests, but this is an area where we can feel good about collaborating because we’ll be pushing forward the overall mission of decarbonizing transportation. Charge on!