CHAdeMO Announces High Power (150KW) Version of the Protocol

[techmaven]

http://www.chademo.com/wp/wp-content/uploads/2016/06/2016-06-01_High_power_CHAdeMO.pdf

They expect first deployments in 2017 and has a 350 amp limit. It's only slightly slower than what a Supercharger can do with a Model 70 pack (370 amps) and has more than enough current to cover the rest of the Model S packs.

The revision is still being worked on so the standard isn't finished. They are going to use the same plug and remain backwards compatible.

 

[Xenoilphobe]

Now if they could just make the CHAdeMO stations reliable and maintain them it would awesome.

 

[Jeff N]

The 150 kW in the press release is a little confusing. Apparently they mean 500V @ 350A which one might think would be 175 kW. Note that the slide also refers to today's 500V @ 125A as 50 kW so what it may really mean is that it can go up to 500V and up to 350A but not at the same time -- that they total power must be capped at 50 kW. I think ChargePoint's "50 kW" CHAdeMO charger is specified that way and it's probably because the existing CHAdeMO specification makes that limitation a requirement for some reason. Does the same kind of limitation apply to the future "350 kW" specification such that it will really be limited to 300 kW or so?

Notice how the slide also says something about CHAdeMO - CCS? Apparently they are working jointly with the CCS folks in some way on future specs possibly because the charging station vendors want to build stations with the same power levels with cables and plugs for both connectors.

I found a presentation slide online:

 

[stopcrazypp]

The 150 kW in the press release is a little confusing. Apparently they mean 500V @ 350A which one might think would be 175 kW. Note that the slide also refers to today's 500V @ 125A as 50 kW so what it may really mean is that it can go up to 500V and up to 350A but not at the same time -- that they total power must be capped at 50 kW. I think ChargePoint's "50 kW" CHAdeMO charger is specified that way and it's probably because the existing CHAdeMO specification makes that limitation a requirement for some reason. Does the same kind of limitation apply to the future "350 kW" specification such that it will really be limited to 300 kW or so?

I found a presentation slide online:

The issue is based on how battery charging works (constant current then constant voltage), when you hit the peak current (125A or 350A in your examples), you are guaranteed to be at a lower charging voltage. By the time you hit the max charging voltage (500V), you would necessarily have to ramp down current. Thus peak current and peak voltage will never occur at the same time.

150kW seems like a move by CHAdeMO to harmonize with the other IEC standard (CCS). They are probably seeing the writing on the wall if they don't match the power of CCS, and have taken a similar approach as Tesla in pushing the pin size as far as it go (like Tesla and CCS, I wonder if this means liquid cooling of the cable also).

The 350kW part is very interesting to me. They are specifying 1000V @350A for 350kW, which by the way they talk about it seems to be what CCS might be doing. If so, that is quite disappointing (would be a spec that can't be reached in the real world).
"In terms of higher power, for example 350kW (1 000V x 350A), technical studies are ongoing and the
Association will determine its further development around 2018, should there be market demand."

I was hoping 350kW meant a higher charging current, so that it can be reached in the real world (probably like 700V@500A).

 

[David99]

Now if they could just make the CHAdeMO stations reliable and maintain them it would awesome.

All eVGo NRG stations I visited worked and were 100% reliable even when charging for an hour straight. This discussion is about the CHADeMO standard. There is obviously no reliability problem with the standard itself.

 

[Jeff N]

The issue is based on how battery charging works (constant current then constant voltage), when you hit the peak current (125A or 350A in your examples), you are guaranteed to be at a lower charging voltage. By the time you hit the max charging voltage (500V), you would necessarily have to ramp down current. Thus peak current and peak voltage will never occur at the same time.

Yes, I understand that but I'm not sure that is really related to what I was pointing out. With 125A today the peak realistic power kW with real battery packs is still less than 50 kW. And, I was trying to point out the inconsistency of talking about 500V @ 125A as 50 kW and 500V @ 350A as 150 kW while at the same time referring to 1000V @ 350A as being 350 kW.

The 350kW part is very interesting to me. They are specifying 1000V @350A for 350kW, which by the way they talk about it seems to be what CCS might be doing. If so, that is quite disappointing (would be a spec that can't be reached in the real world).
"In terms of higher power, for example 350kW (1 000V x 350A), technical studies are ongoing and the
Association will determine its further development around 2018, should there be market demand."

Right, as I was just saying....

I was hoping 350kW meant a higher charging current, so that it can be reached in the real world (probably like 700V@500A).

I doubt they will specify anything above 350A continuous with the existing connector although they might allow higher rates during the first few minutes of charging before temperatures get too high.

 

[Blu Zap]

On top of all this the eVGo CHADemo chargers only allow 30 minutes of charging at a time. And that's at a couple of bucks a hit. Yikes!

 

[stopcrazypp]

Yes, I understand that but I'm not sure that is really related to what I was pointing out. With 125A today the peak realistic power kW with real battery packs is still less than 50 kW. And, I was trying to point out the inconsistency of talking about 500V @ 125A as 50 kW and 500V @ 350A as 150 kW while at the same time referring to 1000V @ 350A as being 350 kW.

Okay, fair point.

Right, as I was just saying....

Actually my point was a bit different. I don't really care about the new CHAdeMO standard (since the way I see it CHAdeMO is on the way out except in Japan). What I mean is that the way the document is worded, it implies CCS 2.0/3.0 (which also promised "350kW") might be using the 1000V@350A specification and CHAdeMO got their number from there.

I doubt they will specify anything above 350A continuous with the existing connector although they might allow higher rates during the first few minutes of charging before temperatures get too high.

For CCS, they have extra power pins available (ones used for AC charging) that can allow for more current carrying capacity. We know Tesla already uses the Type 2 socket to enable 370A peak. If the new CCS standard will be pushing 350A through the extra power pins on the bottom, that means it can carry up to 720A with those extra pins.

The Type 1 socket is a bit more limited (as it doesn't have 4 power pins like Type 2). It is currently rated for 80A, but if they "pushed it" like they did in these cases, they can probably bump it to the 150A need to make up the deficit to 500A.

 

[renim]

quick relevance snapshot

2016
EV Sales: World Top 10 - April 2016
for the top 10 pluginvehicle vehicles ytd
CCS has 1 vehicle, and 6% marketshare
Chademo has 2.5 vehicles and 39% marketshare

2015
EV Sales: World Top 10 April 2015
for the top 10 pluginvehicle vehicles ytd
CCS has 2 vehicle, and 15% marketshare
Chademo has 2.5 vehicles and 46% marketshare

so globally CCS appears to be collapsing, and Chademo is gently receding. Some would say lies, damn lies and statistics, but with the data at hand, and YTD figures, can we provide better information?

China's choice of a CAN bus based, separate port for high power DC architecture for their standard effectively the global standard as far as manufacturing scale is concerned.

 

[Jeff N]

For the next 3 years or so it looks like Nissan may be the only CHAdeMO car maker selling significant numbers of cars in North America or Europe. Maybe Honda will sell a few but that's far from certain.

Meanwhile, many of the major European, Korean, and US makers will be selling CCS BEVs.

The market is still tiny and immature. What matters is future sales, not past sales.

 

[stopcrazypp]

quick relevance snapshot

2016
EV Sales: World Top 10 - April 2016
for the top 10 pluginvehicle vehicles ytd
CCS has 1 vehicle, and 6% marketshare
Chademo has 2.5 vehicles and 39% marketshare

2015
EV Sales: World Top 10 April 2015
for the top 10 pluginvehicle vehicles ytd
CCS has 2 vehicle, and 15% marketshare
Chademo has 2.5 vehicles and 46% marketshare

so globally CCS appears to be collapsing, and Chademo is gently receding. Some would say lies, damn lies and statistics, but with the data at hand, and YTD figures, can we provide better information?

China's choice of a CAN bus based, separate port for high power DC architecture for their standard effectively the global standard as far as manufacturing scale is concerned.

Japan (which will be CHAdeMO) and China (which will be the GBT standard) is irrelevant. What matters is the USA and Europe (where CCS and CHAdeMO are competing).

Given the implosion of Mitsubishi and subsequent purchase by Nissan, they are really one company now.

PSA has switched to the CCS camp:
PSA Drops CHAdeMO And Joins CCS Combo for Future EVs

Hyundai has switched to CCS for their main EV, the Ioniq (from using CHAdeMO in their Kia Soul compliance car)
Hyundai Confirms SAE Combo 100 kW Fast Charging For IONIQ Electric

While Tesla has a CHAdeMO adapter, they recently joined the CCS group as a core member (which means they are either developing an adapter or might adopt the next version of the standard):
Tesla Joins CCS-Based CharIn Association

 

[widodh]

The current CHAdeMO adapter of Tesla is limited to 125A. If they would release a new version which can handle up to 250A or maybe 300A we could get >100kW charges at CHAdeMO stations.

In the Netherlands we have Fastned[URL="http:// and they provide reliable stations with CCS and CHAdeMO along the major highways.

Would love to get >100kW charges there. Suddenly we have 50 additional 'SuperChargers"] and they provide reliable stations with CCS and CHAdeMO along the major highways.

Would love to get >100kW charges there. Suddenly we have 50 additional 'SuperChargers'![/url]

 

[Jonas_man]

Is there any new info about a CHAdeMO adaptor for 100kW , like rumours or so?

I will need to buy an adaptor for the summer, so im still waiting for it to come

 

[miimura]

Okay, fair point.


Actually my point was a bit different. I don't really care about the new CHAdeMO standard (since the way I see it CHAdeMO is on the way out except in Japan). What I mean is that the way the document is worded, it implies CCS 2.0/3.0 (which also promised "350kW") might be using the 1000V@350A specification and CHAdeMO got their number from there.


For CCS, they have extra power pins available (ones used for AC charging) that can allow for more current carrying capacity. We know Tesla already uses the Type 2 socket to enable 370A peak. If the new CCS standard will be pushing 350A through the extra power pins on the bottom, that means it can carry up to 720A with those extra pins.

The Type 1 socket is a bit more limited (as it doesn't have 4 power pins like Type 2). It is currently rated for 80A, but if they "pushed it" like they did in these cases, they can probably bump it to the 150A need to make up the deficit to 500A.

The problem is that you cannot use parallel pins to divide the current to a single battery pack. There is no way to ensure that the current through each path is what you want. In order to take advantage of the DC-Mid Type-2 pins the way Tesla currently does as well as the DC-High pins the way CCS is specified, you would have to divide the pack in two parts and charge them separately. This is possible, but not at all likely. Similarly unlikely as taking a 400VDC vehicle and reconfiguring the pack to 800VDC just for higher speed charging, then putting it back to 400VDC for driving.

 

[stopcrazypp]

The problem is that you cannot use parallel pins to divide the current to a single battery pack. There is no way to ensure that the current through each path is what you want. In order to take advantage of the DC-Mid Type-2 pins the way Tesla currently does as well as the DC-High pins the way CCS is specified, you would have to divide the pack in two parts and charge them separately. This is possible, but not at all likely. Similarly unlikely as taking a 400VDC vehicle and reconfiguring the pack to 800VDC just for higher speed charging, then putting it back to 400VDC for driving.

Type 2 DC-Mid as Tesla uses already uses parallel conductors when in DC mode. I'm talking about doing a triple parallel (using the bottom conductors also as the third pair). This is more complex than the DC-mid case, as the different conductors are no longer going to be the same characteristics so the current flowing through will not be the same. But it's not something impossible to implement and doesn't require dividing the packs (only involves the conductors and the DC chargers). The standard itself can specify relative conductor/ connector resistance for each of the pairs in order to keep things consistent.

 

[miimura]

Type 2 DC-Mid as Tesla uses already uses parallel conductors when in DC mode. I'm talking about doing a triple parallel (using the bottom conductors also as the third pair). This is more complex than the DC-mid case, as the different conductors are no longer going to be the same characteristics so the current flowing through will not be the same. But it's not something impossible to implement and doesn't require dividing the packs (only involves the conductors and the DC chargers). The standard itself can specify relative conductor/ connector resistance for each of the pairs in order to keep things consistent.

You can specify it all you want, but the real world has a way of getting things dirty so they're out of spec and that would be very dangerous when there are multiple current paths between one sink and one source. There has been a lot of discussion about how the Tesla connector on California Superchargers has been giving very uneven results, frequently leading to low power delivery. I would expect that the theoretical benefit to your triple-parallel scheme would quickly deteriorate the same way because the system would have to drag everything down to the least common denominator when there is any increase in contact resistance at any point in any of the chains, just to maintain safe operation.

 

[stopcrazypp]

You can specify it all you want, but the real world has a way of getting things dirty so they're out of spec and that would be very dangerous when there are multiple current paths between one sink and one source. There has been a lot of discussion about how the Tesla connector on California Superchargers has been giving very uneven results, frequently leading to low power delivery. I would expect that the theoretical benefit to your triple-parallel scheme would quickly deteriorate the same way because the system would have to drag everything down to the least common denominator when there is any increase in contact resistance at any point in any of the chains, just to maintain safe operation.

Well the DC-mid on the Type 2 plug used by Tesla in Europe already uses parallel conductors that has multiple current paths, so apparently they figured that out already. And there seems to have actually be less problems reported on those superchargers (although this may be a reporting difference as this forum is mostly English).

The unequal characteristics in the triple parallel scheme will be the biggest challenge, but not something insurmountable. You just allow enough headroom in the spec so that you won't run into problems in the real world.

 

[TonyWilliams]

For the next 3 years or so it looks like Nissan may be the only CHAdeMO car maker selling significant numbers of cars in North America or Europe. Maybe Honda will sell a few but that's far from certain.

Honda will use SAE-CCS-Combo 1 on their "80 mile" EV that will be sold in CARB-ZEV states in the USA.

I can assure you that Nissan might sell a few more LEAF v2.0 with 200+ miles of range using a CHAdeMO port.

 

[Jonas_man]

In europe at least Tesla might just use CCS ports. They are compatible with Type2, so that would be better than using adaptors.
Maybe the model 3 will be this way?