Why Tesla doesn't make a CCS adapter like Chademo?

[Ulmo]

I think you are confusing AC vs DC charging.

An adapter between Tesla and J1772 AC charging is trivial because the pins and protocols are identical except for the physical plug design.

An adapter between Tesla and CCS is much more involved. CCS DC charging is the effectively the same regardless of USA type 1 CCS or European type 2 CCS plug format. The signaling protocols and messages are the same between the two CCS DC plug designs, I'm pretty sure.

Tesla's DC protocols are actually closer to CHAdeMO since they seem to use similar, but perhaps not identical, signaling that is based on the so-called CAN bus design that is typically used to network together internal computers and components even within conventional cars. However, Tesla does not seem to use the same CHAdeMO messaging between the car and the DC charger on top of CAN.


The CCS uses a completely different signaling system for DC charging called PLC and sends messages on that data transmission layer using the same tcp/ip protocols that the Internet is based on. The actual messages sent between the car and the DC charger are unique to CCS.

From a design perspective, the engineering behind the Chademo adapter is probably akin to the engineering behind a similar CCS adapter.

I see two routes in the future, both being implemented by Tesla: (1) Tesla creates a CCS->Tesla adapter with built-in signalling conversion (much like Chademo), for use with CCS fast charging to legacy (all current) Model S & X, sold for around $700 ea. (2) Tesla also introduces additional hardware into all new Model S, Model X and all Model 3 as well as any redesigned Roadster to do the necessary signalling (ethernet modem), and includes in the car charger bundle a CCS->Tesla adapter plug, at a fraction of the cost and complexity as their (1) CCS->Tesla adapter, sold for around $20 ea. Both will exist on the Tesla shop for decades. Software in all Tesla cars (old and new Model S & X, Model 3, etc.) will be upgraded to talk CCS logic to CCS chargers (such as having tcp/ip stacks, CCS protocols, authentication, billing, etc.), whether or not the signalling modem is in-car (2) or in-adapter (1). (Some of the software/firmware for the ethernet & tcp/ip and even the CCS protocol handling can be put into the physical area of the modem, or not. Kind of irrelevant from an outside perspective --- just as long as Tesla is consistent enough about it to make it easy on themselves to engineer reliably.)

For those who have both a new car with new modem hardware and an expensive adapter for old cars, I suppose a compatibility mode from the car would just choose one of the modems to use. That would save someone $20 when they trade in their old $80K car for a new $80K car AND loose the $20 converter plug. Silly footnote. Software-wise it would be trivial, though, so I threw it in here for comprehension's sake.

How far off am I?

 

[Ulmo]

Tesla has limited bandwidth and funds. I would put a CCS adapter somewhere between low priority and no priority over the next year.

I could see it being started in ernest in the next quarter after the raise. Model 3 is coming and so is CCS. They would probably design a CCS->Tesla adapter plug for use on Model S & X mules modified to have Model 3 prototype charger hardware, test it out a bunch, then incorporate it into the logic and hardware of the Model 3, testing it out on the Model 3 mules a bunch, so that when Model 3 hits the market, it has CCS signalling built in and off-the-shelf CCS compatibility with a $20-$30 adapter plug, included in any Model 3 charge plug adapter kit.

I could also see it being done a few years down the road once Model 3 has been in the marketplace for a while. But if they do it that way, they'd probably try to make it easier to upgrade Model 3 for CCS compatibility so that Model 3 owners aren't stuck with a $650 converter adapter like the less unwealthy Model S & X owners were, since that might leave a sour taste in their mouth to spend 5% of the car's worth on a "dumb"* plug to plug into the going standard. If Tesla chooses this route, and I were Tesla, I'd also try to run many modified mules of Model S, X, 3 to develop compatibility with CCS before releasing Model 3, so that at least as much easy hardware upgrade for existing Model 3 can be done to mitigate costs for early Model 3 adopters who upgrade to CCS.

* Dumb from the point of view of a consumer.

 

[MP3Mike]

I see two routes in the future, both being implemented by Tesla: (1) Tesla creates a CCS->Tesla adapter with built-in signalling conversion (much like Chademo), for use with CCS fast charging to legacy (all current) Model S & X, sold for around $700 ea. (2) Tesla also introduces additional hardware into all new Model S, Model X and all Model 3 as well as any redesigned Roadster to do the necessary signalling (ethernet modem), and includes in the car charger bundle a CCS->Tesla adapter plug, at a fraction of the cost and complexity as their (1) CCS->Tesla adapter, sold for around $20 ea.

How far off am I?

My guess is a long ways off. The J1772 -> Tesla adapter currently costs $95. A CCS version would be bigger/more complex and cost more. There will never be a $20 adapter.

My guess is that the only option will be a large, unwieldy, adapter like the CHAdeMO one.

 

[stopcrazypp]

I could see it being started in ernest in the next quarter after the raise. Model 3 is coming and so is CCS. They would probably design a CCS->Tesla adapter plug for use on Model S & X mules modified to have Model 3 prototype charger hardware, test it out a bunch, then incorporate it into the logic and hardware of the Model 3, testing it out on the Model 3 mules a bunch, so that when Model 3 hits the market, it has CCS signalling built in and off-the-shelf CCS compatibility with a $20-$30 adapter plug, included in any Model 3 charge plug adapter kit.

I could also see it being done a few years down the road once Model 3 has been in the marketplace for a while. But if they do it that way, they'd probably try to make it easier to upgrade Model 3 for CCS compatibility so that Model 3 owners aren't stuck with a $650 converter adapter like the less unwealthy Model S & X owners were, since that might leave a sour taste in their mouth to spend 5% of the car's worth on a "dumb"* plug to plug into the going standard. If Tesla chooses this route, and I were Tesla, I'd also try to run many modified mules of Model S, X, 3 to develop compatibility with CCS before releasing Model 3, so that at least as much easy hardware upgrade for existing Model 3 can be done to mitigate costs for early Model 3 adopters who upgrade to CCS.

* Dumb from the point of view of a consumer.

You beat me to posting this. I would expect Model 3 to have the hardware built-in to the car to support PLC signaling necessary for a "dumb" CCS adapter (if Tesla didn't have it already in Model S/X, just not enabled). On the other hand, this is not possible for CHadeMO, because the Tesla connector is not pin compatible with CHadeMO (Tesla connector has less pins than the minimum needed to do the CHAdeMO handshake, so the adapter must always be a "smart" one that emulates the missing pins).

 

[stopcrazypp]

My guess is a long ways off. The J1772 -> Tesla adapter currently costs $95. A CCS version would be bigger/more complex and cost more. There will never be a $20 adapter.

My guess is that the only option will be a large, unwieldy, adapter like the CHAdeMO one.

It might not be $20, but it should be possible to have a purely physical "dumb adapter" similar to the J1772 adapter today. The pins are compatible and the first analog half of the handshake is exactly the same. Tesla just has to deal with supporting PLC, which they can either do in the adapter or in the car.

I had a thread on this three years ago:
SAE J1772 DC (Combo) Connector Adapter for Model S
There's actually a DC level 1 mode that the current J1772 adapter (and in Europe no adapter necessary) can already support physically, but it seems no CCS chargers have implemented this type of connector and mode.

 

[miimura]

When there is a significant number of 200 amp CCS stations installed it will make sense for Tesla to offer a CCS adapter. I don't expect them to do it before then.

 

[Jonas_man]

Im really waiting for the ccs adaptor. Chademo is huge and expensive.

 

[MP3Mike]

Im really waiting for the ccs adaptor. Chademo is huge and expensive.

What makes you think the CCS adapter will be any smaller/cheaper? Have you seen a CCS connector?

 

[McRat]

It is not to Tesla's advantage to produce a CCS adapter. All that will do is accelerate the expansion of the CCS network and encourage more long range EVs to be developed.

When the CCS grid is complete, and other long range cars are in significant number, I imagine at that point it would make sense.

Today, Tesla has an advantage as being the only pure BEV to be able to go coast to coast in a reasonable amount of time. It's a great selling point, even to those who don't drive interstate.

But once other brands have the same ability to go coast to coast with a CCS network, the cheapest way to expand Tesla Charging Infrastructure for the rapid deploy of the Model 3 will be a CCS adapter.

 

[RDoc]

There was a long thread that became pretty technical about whether a "dumb" CCS adapter was possible, note that the original name for CCS was J1772 Combo.

SAE J1772 DC (Combo) Connector Adapter for Model S
the "dumb" adapter discussion starts around post 145

This is a quote from JB Straubel in an SAE interview about Tesla's CCS compatibility:

SAE: What about the communication protocol of the Combo Connector? It’s considered essential for V2G.

Straubel: That’s fine. We’re definitely commonizing with all of that. The only thing that’s up for debate in all of these standards is the physical geometries of the pins and sockets. Everything else is pretty easy to adapt to. The communication standards are pretty universal. We’re 100% compliant with all the J1772 communication levels, signaling, voltage, everything.​

Tesla Motors CTO talks future batteries and charging protocols - SAE International​

Here's a link to a video of a Tesla engineer who was on the SAE committee that designed CCS saying that Tesla is software and electrically compatible with CCS. The video is pretty interesting in its own right.

The SAE Combo discussion starts at about 1:45.

From my reading I believe Tesla could make a dumb CCS adapter using the cars' built in DSP to implement the PLC communications protocol that CCS uses. It would be a bigger version of the current J1772 adapter, and like it, but unlike the CHAdeMO adapter, wouldn't require any electronics.

 

[McRat]

There was a long thread that became pretty technical about whether a "dumb" CCS adapter was possible, note that the original name for CCS was J1772 Combo.

SAE J1772 DC (Combo) Connector Adapter for Model S
the "dumb" adapter discussion starts around post 145

This is a quote from JB Straubel in an SAE interview about Tesla's CCS compatibility:

SAE: What about the communication protocol of the Combo Connector? It’s considered essential for V2G.

Straubel: That’s fine. We’re definitely commonizing with all of that. The only thing that’s up for debate in all of these standards is the physical geometries of the pins and sockets. Everything else is pretty easy to adapt to. The communication standards are pretty universal. We’re 100% compliant with all the J1772 communication levels, signaling, voltage, everything.​

Tesla Motors CTO talks future batteries and charging protocols - SAE International​

Here's a link to a video of a Tesla engineer who was on the SAE committee that designed CCS saying that Tesla is software and electrically compatible with CCS. The video is pretty interesting in its own right.

The SAE Combo discussion starts at about 1:45.

From my reading I believe Tesla could make a dumb CCS adapter using the cars' built in DSP to implement the PLC communications protocol that CCS uses. It would be a bigger version of the current J1772 adapter, and like it, but unlike the CHAdeMO adapter, wouldn't require any electronics.

It seems you are just restating my point. Yes, it is possible and would take little resources or time.

So the reason you don't see one yet is financial strategy, not technical limitations.

 

[Saghost]

It is not to Tesla's advantage to produce a CCS adapter. All that will do is accelerate the expansion of the CCS network and encourage more long range EVs to be developed.

When the CCS grid is complete, and other long range cars are in significant number, I imagine at that point it would make sense.

Today, Tesla has an advantage as being the only pure BEV to be able to go coast to coast in a reasonable amount of time. It's a great selling point, even to those who don't drive interstate.

But once other brands have the same ability to go coast to coast with a CCS network, the cheapest way to expand Tesla Charging Infrastructure for the rapid deploy of the Model 3 will be a CCS adapter.

Remind me again, what is Tesla's Mission Statement?

Do you doubt that they mean it?

The fact is that even with faster charging rates, until CCS stations are installed in clusters and along interstates outside of cities, they won't be a suitable replacement for Superchargers as a whole, and Tesla's advantage isn't under any risk.

Being able to charge at CCS stations will be of more value to Tesla as a marketing benefit to their cars than the encouragement it might give to building a network would be detrimental - even if they weren't interested in fostering EVs in general.

 

[RDoc]

It is not to Tesla's advantage to produce a CCS adapter. All that will do is accelerate the expansion of the CCS network and encourage more long range EVs to be developed.
<snip>

Why would that accelerate the CCS network in any significant way? I'm pretty doubtful that Tesla selling a CHAdeMO adapter has accelerated that network.

Long range EVs are going to need a nationwide network of >100kW chargers to be practical, which means CCS 2 and significant build out expense. I see no one one ready to take on a project like that for a long time.

IMHO this lack is going to seriously limit the number of Bolts GM is going to sell. It won't take too many stories about long distance travel in a Bolt taking twice or three times as long as in an ICE to turn off a lot of potential customers.

 

[McRat]

Pay per charge sites are expanding. They rely on customers. The more customers, the more sites.

 

[RDoc]

Pay per charge sites are expanding. They rely on customers. The more customers, the more sites.

Sure, but how many times is a Tesla going to use a CCS site? I've only used the J1772 adapter twice and I doubt that many people with CHAdeMo adapters use them more than an few times per year. The Superchargers are so superior that CHAdeMo and CCS are only a distance second choice. I'm very doubtful that small increased use would enhance CCS networks enough to have any impact on other BEV manufacturers.

However, having a CCS adapter would significantly expand the travel possibilities for Teslas which would directly help them sell cars. In my case, I'd like to drive up to Nova Scotia but will need either a CCS or CHAdeMo adapter once there to have any reasonable charging options.

 

[Naonak]

Sure, but how many times is a Tesla going to use a CCS site? I've only used the J1772 adapter twice and I doubt that many people with CHAdeMo adapters use them more than an few times per year. The Superchargers are so superior that CHAdeMo and CCS are only a distance second choice. I'm very doubtful that small increased use would enhance CCS networks enough to have any impact on other BEV manufacturers.

However, having a CCS adapter would significantly expand the travel possibilities for Teslas which would directly help them sell cars. In my case, I'd like to drive up to Nova Scotia but will need either a CCS or CHAdeMo adapter once there to have any reasonable charging options.

I use my CHADeMO quite often, probably 15 - 20 times a month. CHAdeMo is just as a fast as a SuperCharger @ > 50% SOC, so there's not really any speed advantage if you charge @ or above 50%. There's not a Supercharger anywhere near me, so that plays a factor as well, of course. Yes, the adapter is a pain and I'd use a CCS adapter if it were smaller and more easily managed, but honestly, I have never seen a CCS station that wasn't also a CHADeMo station - do they exist?

 

[DarkMatter]

I use my CHADeMO quite often, probably 15 - 20 times a month. CHAdeMo is just as a fast as a SuperCharger @ > 50% SOC, so there's not really any speed advantage if you charge @ or above 50%. There's not a Supercharger anywhere near me, so that plays a factor as well, of course. Yes, the adapter is a pain and I'd use a CCS adapter if it were smaller and more easily managed, but honestly, I have never seen a CCS station that wasn't also a CHADeMo station - do they exist?

The nearest BMW dealer has a CCS station that is not a CHADeMo station.

BMW Launches New Low Cost DC Fast Chargers From $6,458

It's only 24kW. Dual chargers can go up to 19.2kW. So it's not exactly a speed demon.

ABB also makes single-standard CCS chargers. They are upgradeable to multi-standard so they are probably not a great example.

 

[TonyWilliams]

There was a long thread that became pretty technical about whether a "dumb" CCS adapter was possible, note that the original name for CCS was J1772 Combo....

From my reading I believe Tesla could make a dumb CCS adapter using the cars' built in DSP to implement the PLC communications protocol that CCS uses. It would be a bigger version of the current J1772 adapter, and like it, but unlike the CHAdeMO adapter, wouldn't require any electronics.

I hate to burst your bubble, but regardless of what you want to believe, Tesla Supercharger is CAN controlled, just like every car and every DC charge protocol around the world EXCEPT for CCS, which is PLC:

Where ----------- Protocol --- Communicarion

Worldwide ----- CHAdeMO ------ CAN
Worldwide ---- Supercharger---- CAN
China ------------ GB/T ------------- CAN
North America - CCS-Combo1 - PLC
Europe ---------- CCS-Combo2 - PLC

The closest that Supercharger comes to mimicking anything from "CCS" is when the a 5% duty cycle PWM (J1772 pilot signal) is sent to the car to trigger digital communication via CAN. The car respsonds with CAN data, starting with the VIN.

There is not going to be a "dumb" CCS adaptor.

 

[TonyWilliams]

Im really waiting for the ccs adaptor. Chademo is huge and expensive.

The CHAdeMO adaptor is $450. I seriously doubt that any other high powered DC adaptor will be cheaper. I also doubt that it will be less clunky.

 

[Saghost]

I use my CHADeMO quite often, probably 15 - 20 times a month. CHAdeMo is just as a fast as a SuperCharger @ > 50% SOC, so there's not really any speed advantage if you charge @ or above 50%. There's not a Supercharger anywhere near me, so that plays a factor as well, of course. Yes, the adapter is a pain and I'd use a CCS adapter if it were smaller and more easily managed, but honestly, I have never seen a CCS station that wasn't also a CHADeMo station - do they exist?

The Royal Farms gas station near my work has a CCS only station. It's true that most CCS locations I see are combo stations with a cable for each, but CCS only do exist and could be useful to charge at.

350kW capable CCS locations are coming, which at 400V can deliver Supercharger power levels; I'm not aware of anything similar on the CHAdeMO side. Having a CCS adapter might be more beneficial than having a CHAdeMO in the future because of the higher rates (having both or one that can do either is of course the best case.)