It's a high-voltage connector; the actual AC voltage input, battery charger resides in the car. The DC fast charger (supercharger, ionity, etc) resides outside the car.
They contain huge AC/DC inverters. The car has a small, usually 10 or 20kw inverter built in that handles level 1 and 2 charging. Usually about the size of two textbooks per 10kw.
When you're at a DC fast charger they still only have access to AC power from the grid, but you want 250kw, 350kw of DC power to charge the car. Not only do you need 20x the inverters, but because they don't have to fit in a car and move around, they're fitted with larger, more efficient cooling.
I've wondered this too. You could set the gear in neutral, connect the AC power directly to the motor terminals, and then basically run the inverter in regenerative mode.
It looks elegant on paper, at least, but I'm sure there are reasons why it's not done. Maybe due to safety considerations, maybe because the inverter can't handle that much continuous power, or maybe it's just cheaper to have a few big supercharger inverters than to slightly increase the cost of connectors on every car.
It seems like thermal dissipation is the issue based on the grandparent post. An EV duty cycle using the built-in inverter won’t drain the battery in the same amount of time as high-speed charging would charge the battery.
Perhaps there are also weight optimizations being made that limit bidirectionality of the inverter.
One of the early Renault Zoé with the Continental power train did. It had up to 43kW charging on AC, which was a bit unique. So unique that very very few chargers offer more than 22kW AC.
I also heard that it was very noisy during charging.
Level 2 connectors are only rated to ~19kw[1] so I imagine that must have had it's own custom connector, and I don't particularly see the 80amp chargers becoming particularly common for home use, newer homes generally only have a 200amp main breaker, so 2 of those would take 100% of your continuous power load [2]. I have a 48amp charger and if I need faster than that I go to one of the local fast chargers and worst case wander around a store for 15/20 minutes.
[1] 80amps at 240v, at least in the US, not sure about the type 2 connector used in EU
[2] code in the US says you should only draw 80% of the rated current if the load is continuous
There are a handful of them at businesses and it's wonderful if your car can support it. We get about 60mph on those chargers, so even a quick food stop gets you 20-30 miles of range. 50% charge in about 2 hours!
Most public AC chargers around me are 6.6kw and it's truly a useless amount of power. It's about 20mph, so a 20 minute food stop is giving you ~7 miles of range. To get a 50% charge takes about 6 hours. Even in a 9-5 full work day you're not able to get a full charge.
Yeah, I generally assume the AC chargers at business's are for employees cause it's just not worth it to plug in if you're running in to shop for less than an hour.
A little bit but not much. They're a contactor, a GFCI, some current sensors and limiters, a small section to determine if the EVSE is operating at L1 or L2, and a small microcontroller to control it all. They're very simple devices.
The signaling is surprisingly simple. There is no active negotiation, the EVSE just outputs a PWM signal to specify how much current the car is allowed to draw, and the car uses a variable resistor to request the power be turned on or off.
Yeah, most of this circuit is fluff for comms and networking, Basic EVSE such as J1772 can be implemented with a relay, comparator and square wave generator.
Neat tech, but reading it made it clearer why I'm not interested in using these. (They even managed to get Microsoft involved with Azure.)
Are there any 'privacy condoms' for public chargers that can be used, along with cash? Is that simply impossible, or just financially discouraged for the provider?
> Are there any 'privacy condoms' for public chargers that can be used, along with cash? Is that simply impossible, or just financially discouraged for the provider?
Public chargers aren't going to have a coin slot or a bill acceptor, that would necessitate frequent site visits to take the money, and there goes your profit margin.
DCFC do have car-id communication between the car and charger, and they can be configured to connect to specific accounts from the car so you can"plug and charge" with nothing else if your car is registered with the charger network. I've never heard of a system that took cash, they used to take credit cards at some chargers, now they often require an app install on your phone (and you hope your phone didn't die).
Level 2 chargers in the us use j1772. I'm not aware of any vehicle id communication but there could be something? This article was about European chargers, I suspect (without direct knowledge) they use the same "unauthenticated" level 2 protocol with j1772?
I was thinking similar thoughts the other day. I believe both J1772 & CHAdeMO essentially invite the external equipment onto your vehicle's CAN bus, based on my (fallible) memory that my car's VIN, SoC, etc., have shown up on various external equipment's displays during past charging sessions.
Interestingly, while stopped at a CHAdeMO charger operated by a network I hadn't used before I was allowed to charge for free based on it being my first time. (Presumably because my VIN had never been observed there.) It made me wonder that it ought to be fairly trivial to spoof one's VIN through the use of some kind of CAN bus proxy.
--
edit: looks like I might be wrong about the CAN bus bit, but I do recall seeing my VIN & SoC show up on some EVSE displays. Perhaps it's a trick with the Control Pilot Signal?
> but I do recall seeing my VIN & SoC show up on some EVSE displays. Perhaps it's a trick with the Control Pilot Signal?
You were probably using a CCS charger. Level 3 CCS chargers have a communication capability for stuff like vehicle id and state of charge, which enables plug-and-charge functionality. Level 2 chargers do not.
It was definitely not a CCS charger; my car will only play with CHAdeMO & J-1772. Aren't the only differences of CCS over J-1772 the addition of DC power pins at pack voltage, and the lack of populated AC line voltage pins? Or is there an alternate signaling protocol in use as well? I'll have to pay closer attention to the display the next time I'm at a public charger.
> Level 2 chargers in the us use j1772. I'm not aware of any vehicle id communication but there could be something?
I don't think J1772 provides vehicle ID communication. I know because I wanted to see if I could "lock down" my relatively smart EVSE (a Juicebox) to particular vehicles, and the support engineers at the company told me that it's not possible because no vehicle ID is provided over J1772.
28 comments
[ 3.1 ms ] story [ 71.8 ms ] threadWhen you're at a DC fast charger they still only have access to AC power from the grid, but you want 250kw, 350kw of DC power to charge the car. Not only do you need 20x the inverters, but because they don't have to fit in a car and move around, they're fitted with larger, more efficient cooling.
It looks elegant on paper, at least, but I'm sure there are reasons why it's not done. Maybe due to safety considerations, maybe because the inverter can't handle that much continuous power, or maybe it's just cheaper to have a few big supercharger inverters than to slightly increase the cost of connectors on every car.
Perhaps there are also weight optimizations being made that limit bidirectionality of the inverter.
I also heard that it was very noisy during charging.
[1] 80amps at 240v, at least in the US, not sure about the type 2 connector used in EU
[2] code in the US says you should only draw 80% of the rated current if the load is continuous
Most public AC chargers around me are 6.6kw and it's truly a useless amount of power. It's about 20mph, so a 20 minute food stop is giving you ~7 miles of range. To get a 50% charge takes about 6 hours. Even in a 9-5 full work day you're not able to get a full charge.
https://youtu.be/RMxB7zA-e4Y
https://youtu.be/TcoLCTkM0ys
The DC fast charger has smarts, with circuits that measure resistance, temperature, etc, to keep things from melting.
Edit: info from here: https://www.barbouri.com/2017/03/10/diy-open-evse-v4-23/
https://en.wikipedia.org/wiki/SAE_J1772#Signaling
Are there any 'privacy condoms' for public chargers that can be used, along with cash? Is that simply impossible, or just financially discouraged for the provider?
Public chargers aren't going to have a coin slot or a bill acceptor, that would necessitate frequent site visits to take the money, and there goes your profit margin.
https://www.gridserve.com/electric-vehicle-charging/electric...
Level 2 chargers in the us use j1772. I'm not aware of any vehicle id communication but there could be something? This article was about European chargers, I suspect (without direct knowledge) they use the same "unauthenticated" level 2 protocol with j1772?
Interestingly, while stopped at a CHAdeMO charger operated by a network I hadn't used before I was allowed to charge for free based on it being my first time. (Presumably because my VIN had never been observed there.) It made me wonder that it ought to be fairly trivial to spoof one's VIN through the use of some kind of CAN bus proxy.
--
edit: looks like I might be wrong about the CAN bus bit, but I do recall seeing my VIN & SoC show up on some EVSE displays. Perhaps it's a trick with the Control Pilot Signal?
You were probably using a CCS charger. Level 3 CCS chargers have a communication capability for stuff like vehicle id and state of charge, which enables plug-and-charge functionality. Level 2 chargers do not.
I don't think J1772 provides vehicle ID communication. I know because I wanted to see if I could "lock down" my relatively smart EVSE (a Juicebox) to particular vehicles, and the support engineers at the company told me that it's not possible because no vehicle ID is provided over J1772.