SparkE said:
Infinion said:
Did you know that the J1772 connector is designed to DCFC on the same contacts it uses for AC charging? It was designed for 600V isolation and currents up to 80A. {...}
I find that hard to believe. I had always understood that J1772 was AC only - never DC. Yes, max current is 80 amps - of AC power. The max power (in the US) is around 19 kW - of AC power.
And when using the European plug and 3-phase 240V power, it CAN provide around 40 kW - of AC power. (OK, technically, it's the international version of the J1772 standard : IEC 62196, using a different {type 2} connector.) The Renault Clio offered an option for this in Europe. I had never heard that the original J1772 ever defined a DC interface/connection. Do you have a pointer to the standard where this is called out?
(Well, obviously, the newer "CSS/SAE Combo" plug - an extension to J1772 - provides DC power, but that is the "fast charge" standard, and a later addition with two new connectors.)
All the EVSE and level 2 chargers over North America have primarily used AC power on those contacts, as a matter of choice, and automakers like Chevy have put fuses in their receptacles to protect their wiring to the on-board charger (20A spark ev fuses, 40A bolt ev fuses), but it wouldn't be reasonable to say that DC has never been used on the type 1 connector.
This graphic was produced by prof John Kelly at Weber State University based on the SAE J1772-2017 document.
https://weberstate.app.box.com/s/wcksjm6j80ubrqdqkqr8sy8o6vhlgr0t
John also has a comprehensive breakdown of the J1772 CCS type 1 charge receptacle that you can verify below:
He will mention DC level 1 charging on Contacts #1 and #2 rated at 80A.
I find that hard to believe. I had always understood that J1772 was AC only - never DC. Yes, max current is 80 amps - of AC power. The max power (in the US) is around 19 kW - of AC power.
The current will be an RMS rating, so it doesn't matter if you're flowing DC or AC, they're both equivalent.
https://youtu.be/jZBsOud4O9Q
I had never heard that the original J1772 ever defined a DC interface/connection. Do you have a pointer to the standard where this is called out?
I would, but it's behind a paywall, SAE J1772-201710. You can probably find pdfs of the older documents floating online but the numbers will be different and there's apparently much less content (before 2017 SAE J1772 was just a recommended practice). So it's reasonable to find secondary sources such as institutions like WSU that openly discuss some of the paywalled contents of the standard.
So like I was saying, the standard isn't stopping automakers from accepting/transmitting HVDC to these pins besides upgrading their cabling, plugs, receptacles, and the on-board charger to 600V if it wasn't already. Then they could adopt the same PLC communication and contactor interlocking procedure as regular CCS combo charging.
A much simpler implementation is a buck-boost topology for an on-board charger, eliminating a lot of cost and footprint area to reach the 48kW max cont. power rating. We already see this in use in the 650-800V European EVs where they will step up a 400V DC Fast charger with their DC-DC converter, or use the motor inverter.
If we're lucky, eventually the type 1 connector and its upgraded CCS combo version will both be phased out in the future for the Tesla connector. Aptera did a petition to do just this, and I really agree with them. The free market made a superior connector over the type 1 connector. It essentially got rid of pin #1 & #2, and moved #6 and #7 to the same general area, with larger spacing for the larger pins and higher voltage rating for DC Level 2 charging. I think the only argument in favor of a larger connector would be if it allowed better liquid cooling or higher power density, but I don't think the J1772 CCS combo connector even uses #1/#2 for AC Level 2 or DC level 1 charging while using CCS in a DC Fast charge session. For that matter, I don't think anything is outlined to negotiate two separate current limits in PLC communication, so the connector really is pointlessly redundant and unoptimized.