"Why America's EV chargers keep breaking" - Article

[or1]

agree 100% - minor tweak for emphasis - but yeah - CCS1 means you can NOT have an effective EV charging infrastructure with high availability.

a standards change is required - the euro CCS2 plug would've been another choice. It didn't have to be NACS - but it had to be something.

But the incentive for Ford and GM to reach an agreement with Tesla has been enormous. Going for anything other than NACS would have been to return to square 1.

Sponsored

 

[tigerbalm]

short answer - your CCS1 adapter being broken has no impact on my 16 stall charging site's uptime…yeah it sucks to be you if you drop the adatper and the latch breaks, but my charging site and all it's stalls remain 100% functional for everyone else. And it's your expense (not mine) to fix/replace replace the adapter…

if reliability is what you're going for it makes perfect sense.

Could an adaptor having an optional (at extra cost to buyer) cable attached to it – solve the "short cable" problem at Tesla stations in the short term – until v4 stations are the norm ?

Or would cooling requirements prevent it?

 

[or1]

Repair costs could be brought down, but not breakdown frequency: the plug should have been modular. Even a «CCS1-to-CCS1» adapter is imaginable, secured to the present plug in a way that any electrician could exchange it for a new one.

I don’t mean this as a serious solution, just as an illustration of how little real design thinking has gone into the CCS1 design. CCS1 is already dead.

 

[daveo4EV]

Could an adaptor having an optional (at extra cost to buyer) cable attached to it – solve the "short cable" problem at Tesla stations in the short term – until v4 stations are the norm ?

Or would cooling requirements prevent it?

yeah now I don't know - at that point your patching a patch and doing thing "on your own" while across the street at the 64 stall Supercharger things are going smoothly.

 

[rich_r]

Thanks Dave. You confirmed a lot of things I had suspected about this whole "NACS" announcement. It boils down this:
-Tesla does a much better job of keeping their charging stations operational than other North American providers.
Likely this is due to 2 main reasons:
1) they dont have wonky POS systems
2) They dont have CCS1 connectors which are flawed and prone to breaking so much so that its not economically viable to constantly fix them. So, in essence, the CCS networks in North America don't have a business case for achieving better relaibility.

Also, thanks for confirming that this change doesn't really change the underlying protocol that the cars will communicate to the charger with--- it'll still be CCS. I see many people online conflating the connector type with the underlying protocol.

 

[WasserGKuehlt]

Having managed to catch up (Dave's been busy! ?), it seems a couple of logical jumps/skips have been made in this thread:
- we understand that the CCS1 plug design is prone to failure, but on what basis is it being assumed to be _the_ failure? IOW, are we really saying that all/most of EA failures are broken latches? Why don't EVGo's break?
- regarding the conflation of "CSS the plug" and "CSS the protocol" - forgive me for being naive, but a _standard_ is the formal specification of the entire system - that is, both plug _and_ protocol. We can speak "CSS handshake" over "NACS connection", but that is not _a standard_ (I think the word is "a hack".) I've no doubt it works, but for certification purposes it sounds like the CSS standard would have to be amended?

No disagreement @daveo4EV - you're definitely onto something, and that apparently the proprietary NACS standard is being enhanced/extended to support the CSS protocol.

 

[daveo4EV]

Thanks Dave. You confirmed a lot of things I had suspected about this whole "NACS" announcement. It boils down this:
-Tesla does a much better job of keeping their charging stations operational than other North American providers.
Likely this is due to 2 main reasons:
1) they dont have wonky POS systems
2) They dont have CCS1 connectors which are flawed and prone to breaking so much so that its not economically viable to constantly fix them. So, in essence, the CCS networks in North America don't have a business case for achieving better relaibility.

Also, thanks for confirming that this change doesn't really change the underlying protocol that the cars will communicate to the charger with--- it'll still be CCS. I see many people online conflating the connector type with the underlying protocol.

confirmation is an "overstatement" I would suggest "informed speculation" and also given that my unmodified Taycan already works (and other unmodified CCS vehicles) - it's pretty clear you can run CCS over an NACS connector/cable - and why would they recode their charging software if it "already works".

also knowing what I've observed with automotive industry lead times for "change" - they do nothing in less than 18 months -unless it's already working - which means since they are claiming 2024 for adapters with existing CCS vehicle's - it must be pretty far along for them to announce something like that...

last bit of "evidience" is that Supercharger's in Europe are already "bilingual" i.e. they can clear "talk" native supercharging and/or CCS over the CCS2 connector/cable - so there is no reason for them not to do the same in North America...

I could be wrong on all this - but actually migrating to Tesla Supercharger protocol "native" is a lot of work, and it's extra work because they will still need to support Native CCS for non-Tesla fast charging - if I were the automakers I'd have just one "stack" of fast charging software - and just use CCS everywhere - and no care about the physical connector - it makes very little difference from the onboard software point of view.

also we know you can run J-1772 across NACS - Tesla's been doing it for over a decade - so there is precedent for running SAE 5-wire protocols over NACS and doing clever things to "sniff" the handshake so you can distinguish between standard usage and non-standard usage…

again this is all reasonable speculation on my part based on observed evidence of what works today, how these guys do things, and a little knowledge about software and hardware.

friendly lunch on me in my local SF bay area if I'm wrong and anyone wants to claim "the prize"...

NOTE: even our favorite vehicle vendor is doing something "hacky" in that they run an ISO-xxxx protocl over J-1772 when talking to a PMC+/PMCC - that's how they get meta data like SOC% and charging duration reported from teh car displayed on the PMCC screen - that's not standard J-1772 - so it appears not only tesla is doing "clever" things when they can detect a mutual client/EVSE…it's hacking to run an ISO-xxxx protocol over J-1772 - that's not what it's standard calls for.

 

[daveo4EV]

Having managed to catch up (Dave's been busy! ?), it seems a couple of logical jumps/skips have been made in this thread:
- we understand that the CCS1 plug design is prone to failure, but on what basis is it being assumed to be _the_ failure? IOW, are we really saying that all/most of EA failures are broken latches? Why don't EVGo's break?
- regarding the conflation of "CSS the plug" and "CSS the protocol" - forgive me for being naive, but a _standard_ is the formal specification of the entire system - that is, both plug _and_ protocol. We can speak "CSS handshake" over "NACS connection", but that is not _a standard_ (I think the word is "a hack".) I've no doubt it works, but for certification purposes it sounds like the CSS standard would have to be amended?

No disagreement @daveo4EV - you're definitely onto something, and that apparently the proprietary NACS standard is being enhanced/extended to support the CSS protocol.

it's not "the reason" it is _A_ reason - I would also suggest it's a major contributor and Ford/GM know it and is a driving force in their decision to switch to NACS - I believe the dramatic abandoment of CCS1 physical plug is implied confirmation that it has problems - it was also noted in both Ford and GM announcements noted reliability issues with CCS1 because it has "moving" parts…again I don't know for a fact I'm reading tea leaves and likely informed speculation - it's pretty clear NACS and CCS2 in europe have better track records for low incident of causing a charging stall to be offline.

 

[daveo4EV]

Having managed to catch up (Dave's been busy! ?), it seems a couple of logical jumps/skips have been made in this thread:
- we understand that the CCS1 plug design is prone to failure, but on what basis is it being assumed to be _the_ failure? IOW, are we really saying that all/most of EA failures are broken latches? Why don't EVGo's break?
- regarding the conflation of "CSS the plug" and "CSS the protocol" - forgive me for being naive, but a _standard_ is the formal specification of the entire system - that is, both plug _and_ protocol. We can speak "CSS handshake" over "NACS connection", but that is not _a standard_ (I think the word is "a hack".) I've no doubt it works, but for certification purposes it sounds like the CSS standard would have to be amended?

No disagreement @daveo4EV - you're definitely onto something, and that apparently the proprietary NACS standard is being enhanced/extended to support the CSS protocol.

EVGo in my experience is just as prone to breakage in my local area - chargers are often unavailable due to broken connectors.

 

[daveo4EV]

Having managed to catch up (Dave's been busy! ?), it seems a couple of logical jumps/skips have been made in this thread:
- we understand that the CCS1 plug design is prone to failure, but on what basis is it being assumed to be _the_ failure? IOW, are we really saying that all/most of EA failures are broken latches? Why don't EVGo's break?
- regarding the conflation of "CSS the plug" and "CSS the protocol" - forgive me for being naive, but a _standard_ is the formal specification of the entire system - that is, both plug _and_ protocol. We can speak "CSS handshake" over "NACS connection", but that is not _a standard_ (I think the word is "a hack".) I've no doubt it works, but for certification purposes it sounds like the CSS standard would have to be amended?

No disagreement @daveo4EV - you're definitely onto something, and that apparently the proprietary NACS standard is being enhanced/extended to support the CSS protocol.

I think this will come - or it will remain a "hack" I agree a standard like this normally includes physical specifications and protocols - but it's pretty obvious with 5 wires you can run CCS over a physical NACS cable/connection.

I do hope they amend the "standard" to offer NACS physical spec as an "alternative" connector - they can support CCS1 and NACS as alternate/complementary physical standards and note the handshaking and electrical standards remain unmodified.

 

[daveo4EV]

But the incentive for Ford and GM to reach an agreement with Tesla has been enormous. Going for anything other than NACS would have been to return to square 1.

agreed - CCS2 could've been an option - clearly supply chain issues would be minimal and it's an existing standard and has simliar characteristics to NACS for reliability.

but I think that was never on the table…

the point is _ANY_ change (even a revised CCS1) is painful and expensive and disruptive - again the fact that they are making this change means they saw some long term benefit and CCS1 was untenable for reasons undisclosed but potentially easy to observe/speculate - I would love to hear more directly from sources - rather than speculate.

but NACS probably sweetened the "deal" for Tesla for supercharger network access…where CCS2 would've be "adapter" always for North American Tesla supercharger...

NACS is a more political choice if you don't want to piss off Elon and bring him to the table.

 

[daveo4EV]

Could an adaptor having an optional (at extra cost to buyer) cable attached to it – solve the "short cable" problem at Tesla stations in the short term – until v4 stations are the norm ?

Or would cooling requirements prevent it?

I think cooling rules this out - but who knows - I think the short cord issue will get addressed faster than people realize

this sort of transition falls into my of my favorite quotes…I think this will all behind us in 3-5 years and everyone will wonder what all the fuss was about - there will be some CCS1 holdouts (VW/Audi/Porsche, but they'll come around 2028 or shortly after is my bingo card prediction) and adatpers will provide plausible access for the laggards…

Most people overestimate what they can do in one year and underestimate what they can do in ten years. - Bill Gates

 

[daveo4EV]

Thanks Dave. You confirmed a lot of things I had suspected about this whole "NACS" announcement. It boils down this:
-Tesla does a much better job of keeping their charging stations operational than other North American providers.
Likely this is due to 2 main reasons:
1) they dont have wonky POS systems
2) They dont have CCS1 connectors which are flawed and prone to breaking so much so that its not economically viable to constantly fix them. So, in essence, the CCS networks in North America don't have a business case for achieving better relaibility.

Also, thanks for confirming that this change doesn't really change the underlying protocol that the cars will communicate to the charger with--- it'll still be CCS. I see many people online conflating the connector type with the underlying protocol.

again confirmation is an overstatement - but as noted in other posts - I think we can deduce what's going on here from observable "facts" in evidence with MagicDock in North America working with unmodified CCS vehicles, and the fact that Tesla operates pretty well in Europe with CCS…

I'm willing to associate my name/reputation with this informed speculation and will eat my words if Ford/GM end up actually doing Supercharger native handshakes, but I think it makes no sense, is extra work, and they still have to support non-Supercharging CCS for non-Tesla charging sites...

so it's all "fits" if you think it through - but I have no data other than what i can observe, speculate, and infer about how all this works, can work, and shoudl work given Ford/GM both think they can have this "live" by 2024 ;-)

basically CCS running on NACS is about the only thing that can happen this quickly - given that it clearly already works…

 

[daveo4EV]

but a _standard_ is the formal specification of the entire system - that is, both plug _and_ protocol. We can speak "CSS handshake" over "NACS connection", but that is not _a standard_ (I think the word is "a hack".) I've no doubt it works, but for certification purposes it sounds like the CSS standard would have to be amended?

and right on cue - https://apple.news/A07JLOFgQTZKvSfSHvLs7SA

EV charging body says Tesla's charger connector is not standardized yet
…
CharIN said it will convene a task force with the goal of submitting NACS, which was formerly Tesla's propriety, to the standardization process.

It added that an open standardization process will go through a peer review process and interested parties will be able to contribute to development of the standard.
…

if they are smart they'll "rush" this approval - it should be simple…

I think this is over - but everyone wants to get the ink dry and all the comma's in the right place - running CCS over NACS shouldn't be that big of an amendment

unless of course I'm wrong and everyone _IS_ planning to do Tesla Supercharger protocols…I still find that mostly unbelieveable - but stranger things have happened.

 

[Scandinavian]

and right on cue - https://apple.news/A07JLOFgQTZKvSfSHvLs7SA

if they are smart they'll "rush" this approval - it should be simple…

I think this is over - but everyone wants to get the ink dry and all the comma's in the right place - running CCS over NACS shouldn't be that big of an amendment

unless of course I'm wrong and everyone _IS_ planning to do Tesla Supercharger protocols…I still find that mostly unbelieveable - but stranger things have happened.

Well they would say that wouldn’t they. They designed the CCS1 standard. No role for them any longer if the Tesla standard is implemented. Maybe even a lot of license fees saved?

Sponsored