Supercharger V3 Power - 1000kW?

[cizUK]

Elon tweeted in December that 350 kW would be a 'a children's toy' for the Supercharger V3
So what do you expect the power for this to be?

350-400 would give 100 miles in about 6 mins & 1000 would make that about 2 mins

Tesla raised the bar by 240% with the Supercharger, so could they be increasing it from the 400 kW by the same? - this would make 1000 kW

It's quite a jump, but what do you expect from Tesla?

Obviously, Model S & X would not support it, but could the model 3 with the new cells?

 

[arg]

Your options in the poll don't go low enough.

I don't expect any kind of Supercharger to deliver more than 200kW to any Tesla vehicle we currently know about. Delivering that sort of power level into Model 3 will already be a significant feat - with a smaller battery, and expected lower consumption so more miles-per-hour of charging at a given kW charging rate.

If Elon's tweet is any more than a throwaway line, I suspect it refers to the per-cabinet power or the per-station power rather than the per-stall power.

 

[cizUK]

I wouldn't think so
The major automakers are coming up with 350 kW chargers, EVgo has launched their 350 one and Charge point have launched their 400 kW one.
Surely Tesla will be setting their sights above this for their V3 as it did before.
After all, a power of 1700 kW will be required to be on the same ground as refilling a petrol car (400 miles in 5 minutes)
This would just be another step on the way, albeit a big one.

I must admit I doubt it will be 1000 kW, but I would consider it to be close. (I voted 850-1000)

 

[Craig-Tx]

1000 kW or 1 MW DC is absurd.
Even if they completely change the voltage of the system to be 1000 V (compared to the 400V they use now) that would still be 1000 A.
The current connector likely can't cope with 1000 V or even close to 1000 A without melting.
Even liquid cooling a much thicker cable would help to dissipate some heat, but I can't imagine (much less even find) what size wire would be needed at the supercharger as well as inside the vehicle to support 1000 A. But I can assure you it would not be manageable for the average driver to be able to plug into a vehicle.

My guess is that IF (BIG HUGE IF) every connector / cable / etc. can be proven to support 1000V, they will do that and leave the amperage close to the current 400 A. This would equate to 400 kW without the need to make the cables any larger than they currently are.

With smart switching inside the battery pack, they could possibly reconfigure battery modules between series and parallel to switch the module to 1000 V for incredibly fast DC charging and back to 400 V for normal driving.

 

[cizUK]

1000 kW or 1 MW DC is absurd.
Even if they completely change the voltage of the system to be 1000 V (compared to the 400V they use now) that would still be 1000 A.
The current connector likely can't cope with 1000 V or even close to 1000 A without melting.
Even liquid cooling a much thicker cable would help to dissipate some heat, but I can't imagine (much less even find) what size wire would be needed at the supercharger as well as inside the vehicle to support 1000 A. But I can assure you it would not be manageable for the average driver to be able to plug into a vehicle.

My guess is that IF (BIG HUGE IF) every connector / cable / etc. can be proven to support 1000V, they will do that and leave the amperage close to the current 400 A. This would equate to 400 kW without the need to make the cables any larger than they currently are.

With smart switching inside the battery pack, they could possibly reconfigure battery modules between series and parallel to switch the module to 1000 V for incredibly fast DC charging and back to 400 V for normal driving.

So is the cable the limiting factor?
The 1000V/400A you've suggested is exactly what ChargePoint are doing with their ExpressPlus and there is already a 350kW charger in California.
I would have thought the limiting factor would be that cars can't currently take that power, but I was thinking that the Model 3 could change that.

 

[scaesare]

Well, if you are going to include that last option:

 

[jsmay311]

I wouldn't think so
The major automakers are coming up with 350 kW chargers, EVgo has launched their 350 one and Charge point have launched their 400 kW one.
Surely Tesla will be setting their sights above this for their V3 as it did before.

All of those "350kW" chargers attain such a high (maximum/theoretical) power mostly by doubling the maximum operating voltage from 500V to 1000V. That doesn't help increase the charge rate of any existing passenger EV on the road, all of which have ~400V battery packs.

And given that Teslas are already having occasional issues with having to limit power at Superchargers due to handles/connectors overheating, I don't see how it will be possible to significantly increase the amount of current being pumped through existing charge ports in existing Teslas. (I believe heat generation through the connector increases with the square of the current. P=I^2*R. Anyone know if that's right?) I know you can liquid cool cables, but I'm not sure how much that would help cool the handle/connector. I guess you could probably design a liquid-cooled handle too(?).

Elon's "children's toy" comment might be in reference to a future Tesla semi that would have a much bigger batter, and maybe double the voltage, and probably a unique larger charging connector.

 

[arnis]

One charger supplies power to 2 SC stalls (mostly).
It's financially reasonable to switch to 1-charger 4-stall design.
And it does make sense to use 1-charger 8-stall design.

And this is what Musk meant under "charger" - he didn't mean the stall.
350kW charger can manage 6 vehicles arriving at different times.
I'm sure the idea is to have "best-bang-for-a-buck" system. Therefore
smaller SC locations might have one charger, bigger ones two, extremely
big ones might have 4. Today's solution definitely costs more per stall and
doesn't distribute available power well enough.

Charger itself as a unit is already redundant enough. If one converter fails it
doesn't kill the whole stack, therefore one charger per smaller SC location is way to go.

 

[Joer293]

The current battery packs and wiring could handle higher energy transfer if the chargers were capable of pulsed DC patterns. Similar to how modern military laser systems are kept cool, but delivery higher total energy per time period. That's still only an incremental gain. New battery modules thermal dissipation designed specifically for a pulsed DC charger could quickly shatter current achievements. Existing charging stations would need power packs, new chargers and ultra capacitor banks to be allowed to operate that way on a power grid. It's expensive, but possible, Tesla has some of the brightest engineers on the planet. The question, is it cost effective today? It's obviously easier to make a 4 or 8 stall distributed stepped constant energy flow DC 400kw-600kw charger for almost no increase in total location cost. Driving down costs while increasing value seems to be Tesla's goal. Follow the money.

 

[phigment]

The current battery packs and wiring could handle higher energy transfer if the chargers were capable of pulsed DC patterns. Similar to how modern military laser systems are kept cool, but delivery higher total energy per time period. That's still only an incremental gain. New battery modules thermal dissipation designed specifically for a pulsed DC charger could quickly shatter current achievements. Existing charging stations would need power packs, new chargers and ultra capacitor banks to be allowed to operate that way on a power grid. It's expensive, but possible, Tesla has some of the brightest engineers on the planet. The question, is it cost effective today? It's obviously easier to make a 4 or 8 stall distributed stepped constant energy flow DC 400kw-600kw charger for almost no increase in total location cost.

Can you exaplain a bit more? I'm not familiar with how modulating the DC current would give higher total energy integrated over time.

By pulsing the DC current, you will have periods of high current, followed by zero current. With lasters you want to pack as much power in to a small duration as possible. With battery charging it is the total energy delivered over a period of time.

For example, getting say 1000kW for 1/10th of a second followed by zero for 9/10 of a second would still only deliver the same energy as 100kW for 1 full second.

Unless you are saying for example that the current system could handle say 100kW continuous, but you could push 1000kW for half a second followed by zero for half a second, thus leading to to a 5X increase over the duration of that second.

 

[Jeff N]

I see no evidence that the next generation of cars from any maker will exceed a peak charging rate of 220 kW or so.

Based on Elon's comment I do expect to see the next generation Superchargers support theoretical rates of more than 400 kW but I think the real target for the highest peak charging rates will be the forthcoming Tesla Semi rather than the Model 3 or Model S.

 

[Joer293]

Unless you are saying for example that the current system could handle say 100kW continuous, but you could push 1000kW for half a second followed by zero for half a second, thus leading to to a 5X increase over the duration of that second.

The second part. Electronic circuits have a few variables that can be taken advantage of. I don't know the cell, module and pack variables. Those would determine how much extra energy could be delivered. When a circuit is designed to optimize the pulse width and shape characteristics, you can squeeze more energy through it. LED lights are great examples. In the beginning constant DC power LEDs were used. Then they invented driver circuits to increase energy throughput and light output, while reducing specific heat build up, and maintain life expectancy. Over the years LED's were redesigned to take advantage and optimize those variables. Now what was once a 10% gain is 500% gain. Lithium battery chargers have gone through the same progress on the small scale. Cell Phones do not charge with constant DC, especially rapid chargers use pulse patterns. Cost effective in a high mark up cell phone battery. The question is, will the economics make sense in a lower profit bigger scale battery module?

 

[Joer293]

I see no evidence that the next generation of cars from any maker will exceed a peak charging rate of 220 kW or so.

Based on Elon's comment I do expect to see the next generation Superchargers support theoretical rates of more than 400 kW but I think the real target for the highest peak charging rates will be the forthcoming Tesla Semi rather than the Model 3 or Model S.

I see no evidence the next generation production car will either. The cost/benefit ratio doesn't make sense yet. While possible in a prototype, the production costs have to be engineered way down before we see them everywhere.

 

[Candleflame]

the problem is not the charge voltage but the tapering which unfortuantley does not get adressed, even with a 1000000kw charger.

If there was no taper you could already charge Model S70 from 10% to 80% in 20min....

 

[Joer293]

the problem is not the charge voltage but the tapering which unfortuantley does not get adressed, even with a 1000000kw charger.

If there was no taper you could already charge Model S70 from 10% to 80% in 20min....

right, they need a chemistry change, and pack thermal management design improvements to see major changes. Tesla engineers have actually been improving that taper via software updates to existing charge variables. Take a look around online at the tests, and you'll see the incremental improvements being pushed out. Just remember the taper is a risk limiting charge curve. Tesla uses real world data analysis to constantly reevaluate that risk curve.

 

[Lbkmxp100d]

What charge rate can the 2170 (I believe that's the number of new configured battery) handle? I have noticed an decrease in charge rate at all the supercharging locations over past 2 months and after reading post on forum I'm of the belief that Tesla is putting the slower charge rate in new software releases to prevent overheating and damaging the batteries. If this is true, how would any increase in charging be possible w/o a serious upgrade of a number of things within our vehicles.

 

[arnis]

how would any increase in charging be possible w/o a serious upgrade of a number of things within our vehicles.

There is no serious increase possible right now.
Biggest limitations are charge acceptance and cooling. Without changing chemistry (not form factor to 2170) acceptance will not get better. It is possible to have higher maximum voltage limit, but that will degrade battery faster. Also cooling is a limitation. Tesla is currently limiting battery to 45C or 113F while charging. That limit could be raised but that again will result in faster degradation.

One thing Tesla could do is to precool the pack and be more aggressive with the cooling (like they were before). It appears people hardly notice cooling system maxing out. It is not happening any more. It is also not possible to precool a lot as cold pack will lose acceptance rate.

More cooling power is also not possible as there is no room for more radiator surface area. Model 3 will have even less than S.

I see no evidence that the next generation of cars from any maker will exceed a peak charging rate of 220 kW or so.

I can't even see 220 kW. I don't expect anything above 150kW this decade.

 

[ohmman]

Biggest limitations are charge acceptance and cooling.

Tesla applied for a patent which appears to offload the thermal conditioning portion of charging to the Supercharging station. The idea is that the station could have a much more robust heating/cooling infrastructure in place. The patent has the port on the bottom of the vehicle, which would imply that this will be for future packs only. Some summary images below.

 

[arnis]

Nice. But it costs money. For regular vehicles it is overkill.
Two regular SC's are better than one Ultra SC with ability to cool externally for 150-200kW rate.
I think 50-70kWh within 30 minutes is possible with beefy on-board cooling system.
I think that should be enough.* And it only gets better with more efficient chemistry.
It will definitely not be enough for semis and other big industrial vehicles.

*The only exception might be autobahns and towing a bigger trailer.
But I believe regular car manufacturers will not invest into these specific nuances early on.

 

[KarenRei]

Tesla applied for a patent which appears to offload the thermal conditioning portion of charging to the Supercharging station. The idea is that the station could have a much more robust heating/cooling infrastructure in place. The patent has the port on the bottom of the vehicle, which would imply that this will be for future packs only. Some summary images below.

Huh, funny, I could invalidate this patent with prior art. I've been promoting this concept for ages. Seriously, I think I first started talking about this concept in the late 2000s.

And given that Teslas are already having occasional issues with having to limit power at Superchargers due to handles/connectors overheating, I don't see how it will be possible to significantly increase the amount of current being pumped through existing charge ports in existing Teslas.

The answer is right above. You feed coolant to the vehicle through the same cable as you feed power, so you're simultaneously active cooling (rather than passive cooling) the cable. Lower cable masses, higher powers, less limited internal cooling inside the vehicle. It's always seemed the obvious solution to me.