4 ms·

The real news is BYD is making blade2, and flash charging (10m - 97%) standard going forward, i.e. all models all price points will recharge as fast as filling

The real news is BYD is making blade2, and flash charging (10m - 97%) standard going forward, i.e. all models all price points will recharge as fast as filling gas tank. The flash charging network supposedly also open to all compatible vehicles (probably some sort of titration on older batteries). Once blade2 proliferate, flash charging throughput = can convert many small lots, i.e. convenience stores into recharge stations like gas stations. I think flash charging infra basically just has a fuckload of old blade batteries drip charging from existing electric infra, so no need for major grid overhaul = the station in box charging infra almost anywhere (i.e. if grid supports heavy industrial AC, it can support flash cabinets) is going to be as big as charging time. E: @10m charge per car, the system basically is scaled to typical gas tank up transaction times, i.e. 10-12m per car. The battery storage sized to survive rush hour throughput then charge off grid or roof solar during lull. Basically parity with gas infra. The plan is also to second life old batteries, i.e. 60-70% capacity blade1s... lots of 1st gen bats retiring, storage is going to be essentially "free" via upcycling. AKA entire battery circular economy PRC mandated recently. The last part is what makes BYD so cracked, IIRC central gov legislated extended producer responsibility recently, i.e. BYD (largest battery producer) legally had to take back and recycle batteries - cradle to grave responsibility, instead of billions in logistics for storing/recycling/shredding they're just slapping them in flash stations to increase deprecation cycle.

close04 6mo ago
Flash charging is 1MW - 1000V, 1000A. That peak power has to be provided by the grid or some energy storage that can be replenished in time for the next cars. We need cheap and smallish batteries (because burying is not always an option) to be able to install that much storage and peak power output in the MW/MWh range all over the place. We'll eventually have this but we might be talking in decades rather than years. Or, why not, maybe we need fewer cars.
- danielsamuels 6mo ago
  1MW/800V architecture = 1250A
  - adrian_b 6mo ago
    The 1 MW chargers installed until now (in China) are 1000 V / 1000 A. So the charging voltage has been increased, to allow a less increase in the charging current. I assume that the car negotiates with the charger the charging voltage and the maximum charging current, and then the charging proceeds at the limits established by the least capable of the two.
- maxglute 6mo ago
  I think flash charging infra math is: 1500 kWh battery storage + 600kW grid draw + 30kW solar canopy (if weather allows). @10m per car across 2 cables = ~36 cars over 3 hour rush hour throughput (comparable to 2 gas pump) = need ~4000 kWh, 1500 kWh from battery 1800 kWh from grid = 3300 kWh. This worst case all cars 0-97% charge, realistically mixture so 3300 kWh should cover typical peak scenarios. Then midday lull for grid to refill batteries for evening rush, after again overnight charge from grid. Basically 1x2cable station can service 100-150 cars/day comparable to high-density gas pump. In terms of physical size, battery storage smaller than gas in terms of physical infra = doesn't need to be underground (assuming long term blade safety ensured). Gas needs storage for multiple days / week so need to scale underground tank accordingly. 1500kWh scaled for rush hour = battery storage (recycled) realistically ~30 recycled blade1s slapped into racks, a couple parking spaces worth, and can be stacked vertically. It's much more space efficient (and cost efficient) than gas.
  - ZeroGravitas 6mo ago
    I don't think it's that big a deal. Even older supercharger sites in crumbling post-collapse USA are 1MW (4x 250KW stalls). I think Tesla has an off grid(!) supercharger site in California with 168x 500kW peak v4 chargers. It seems pretty doable to just spread the chargers around to meet the same throughput without causing any hot spots on the grid. The cars already have internet connected navigation systems that can react to how busy a site is and direct you to the next one.
    - maxglute 6mo ago
      Big deal for ubiquitous charge and moving away from charge hub model. Stop by any local convenience store and getting topped up in 10 min, no need to even consider detour time, i.e. BYD parterning up with KFC china to add flash charge - KFCs everywhere in PRC. The bigger deal is infra play, battery buffer mitigates grid hotspots (supercharger sites), so drop station in box in more places with less regulatory drama / capex in grid rework. Combining short charge time + ubiquitous/dropin charger combo that makes it work. Removes friction for drivers and much more economical for builders, can replace 60 spot supercharger site with 10 flash chargers. Instead of building out a charging hub, slap a few chargers in some existing retail lots, i.e. charge where you fastfood with minimal of permitting and construction - the hardware are cabinets on concrete ground pad- it's closer to modular appliance like industrial hvac than infra. Potential for proliferation very fast. IMO it's intermodal container moment for charging.
- adrian_b 6mo ago
  The already existing 1 MW chargers have internal batteries, so they do not need from the grid the peak power, but only the average power. The average power is lower, as there are idle times between cars. Moreover, in the beginning only a few cars would be able to charge at 1 MW, so the average power will be even lower, allowing a later upgrade of the connection to the electrical grid, when fast-charging cars would become more frequent and when there would also be more EV owners, so that more cars would have to be charged per day.
- mekdoonggi 6mo ago
  My understanding from watching a video on this last night, the same Blade 2.0 cells going into the cars are going into the charging stations. And it just so happens that BYD is massively scaling the manufacturing of these.
pjc50 6mo ago
I think this is underestimating just how much 1MW is, even allowing for the fractional duty cycle of time spent changing over who's using the charging station. 1MW of 500W solar panels is 2000 panels, for example. It's probably going to be reserved for highway stations that are also conveniently located near substations. Although you're now making me wonder at what point it becomes more economical to ship electricity in batteries rather than do lengthy, expensive, and annoyingly controversial grid upgrades.
- adrian_b 6mo ago
  The 1-MW chargers have internal batteries, so they can pull a much lower average power from the electrical grid. The connection to the electrical grid of a charging station is not dimensioned based on the charging times. It is dimensioned based on the number of cars that must be charged during a given time interval at that location (assuming a certain average charging energy). So regardless if fast chargers or slow chargers are used, what matters is how many electric cars are used in a region and how much they travel each day. Fast chargers can matter only indirectly, if their presence will convince more of the car users to switch to EVs, requiring the electrical power suppliers to take into account this increased consumption.
  - close04 6mo ago
    > The 1-MW chargers have internal batteries, so they can pull a much lower average power from the electrical grid. We go back to what I was saying earlier [0], you'll need a lot of local storage to keep that charger running at full speed without breaking the back of the distribution grid. That comes on top of the solid grid capacity at every one of those many small parking lots. One stall alone needs 1MW maybe equally split between the grid and the local battery, for probably 50-70kWh per car. Just 3-4 of these cars charging in a parking lot would mean a constant multi-MW pull from the grid on top of the battery that also needs to be recharged from somewhere. To guarantee that you lower that average you need to have enough local storage. The technology exists, it works really well for large dedicated charging spots. The problem is the cost of scaling when you have to deal with lots of small plots (what OP proposed). A 1kW installation with 2-3kWh battery storage is the size of a small container and we've had them for years. Why do you think every small parking lot hasn't been equipped with one? > requiring the electrical power suppliers to take into account this increased consumption This doesn't scale like software. Adding production is comparatively easy, adding distribution capacity isn't. It needs a lot of equipment that's not really available [1], and a lot of construction work to expand the grid capacity. [0] https://news.ycombinator.com/item?id=47932115 https://news.ycombinator.com/item?id=47932115 [1] https://news.ycombinator.com/item?id=47604887 https://news.ycombinator.com/item?id=47604887
Rekindle8090 6mo ago
[dead]

close04 6mo ago

Flash charging is 1MW - 1000V, 1000A. That peak power has to be provided by the grid or some energy storage that can be replenished in time for the next cars. We need cheap and smallish batteries (because burying is not always an option) to be able to install that much storage and peak power output in the MW/MWh range all over the place. We'll eventually have this but we might be talking in decades rather than years. Or, why not, maybe we need fewer cars.

danielsamuels 6mo ago
1MW/800V architecture = 1250A
- adrian_b 6mo ago
  The 1 MW chargers installed until now (in China) are 1000 V / 1000 A. So the charging voltage has been increased, to allow a less increase in the charging current. I assume that the car negotiates with the charger the charging voltage and the maximum charging current, and then the charging proceeds at the limits established by the least capable of the two.
maxglute 6mo ago
I think flash charging infra math is: 1500 kWh battery storage + 600kW grid draw + 30kW solar canopy (if weather allows). @10m per car across 2 cables = ~36 cars over 3 hour rush hour throughput (comparable to 2 gas pump) = need ~4000 kWh, 1500 kWh from battery 1800 kWh from grid = 3300 kWh. This worst case all cars 0-97% charge, realistically mixture so 3300 kWh should cover typical peak scenarios. Then midday lull for grid to refill batteries for evening rush, after again overnight charge from grid. Basically 1x2cable station can service 100-150 cars/day comparable to high-density gas pump. In terms of physical size, battery storage smaller than gas in terms of physical infra = doesn't need to be underground (assuming long term blade safety ensured). Gas needs storage for multiple days / week so need to scale underground tank accordingly. 1500kWh scaled for rush hour = battery storage (recycled) realistically ~30 recycled blade1s slapped into racks, a couple parking spaces worth, and can be stacked vertically. It's much more space efficient (and cost efficient) than gas.
- ZeroGravitas 6mo ago
  I don't think it's that big a deal. Even older supercharger sites in crumbling post-collapse USA are 1MW (4x 250KW stalls). I think Tesla has an off grid(!) supercharger site in California with 168x 500kW peak v4 chargers. It seems pretty doable to just spread the chargers around to meet the same throughput without causing any hot spots on the grid. The cars already have internet connected navigation systems that can react to how busy a site is and direct you to the next one.
  - maxglute 6mo ago
    Big deal for ubiquitous charge and moving away from charge hub model. Stop by any local convenience store and getting topped up in 10 min, no need to even consider detour time, i.e. BYD parterning up with KFC china to add flash charge - KFCs everywhere in PRC. The bigger deal is infra play, battery buffer mitigates grid hotspots (supercharger sites), so drop station in box in more places with less regulatory drama / capex in grid rework. Combining short charge time + ubiquitous/dropin charger combo that makes it work. Removes friction for drivers and much more economical for builders, can replace 60 spot supercharger site with 10 flash chargers. Instead of building out a charging hub, slap a few chargers in some existing retail lots, i.e. charge where you fastfood with minimal of permitting and construction - the hardware are cabinets on concrete ground pad- it's closer to modular appliance like industrial hvac than infra. Potential for proliferation very fast. IMO it's intermodal container moment for charging.
adrian_b 6mo ago
The already existing 1 MW chargers have internal batteries, so they do not need from the grid the peak power, but only the average power. The average power is lower, as there are idle times between cars. Moreover, in the beginning only a few cars would be able to charge at 1 MW, so the average power will be even lower, allowing a later upgrade of the connection to the electrical grid, when fast-charging cars would become more frequent and when there would also be more EV owners, so that more cars would have to be charged per day.
mekdoonggi 6mo ago
My understanding from watching a video on this last night, the same Blade 2.0 cells going into the cars are going into the charging stations. And it just so happens that BYD is massively scaling the manufacturing of these.

pjc50 6mo ago

I think this is underestimating just how much 1MW is, even allowing for the fractional duty cycle of time spent changing over who's using the charging station. 1MW of 500W solar panels is 2000 panels, for example. It's probably going to be reserved for highway stations that are also conveniently located near substations. Although you're now making me wonder at what point it becomes more economical to ship electricity in batteries rather than do lengthy, expensive, and annoyingly controversial grid upgrades.

adrian_b 6mo ago
The 1-MW chargers have internal batteries, so they can pull a much lower average power from the electrical grid. The connection to the electrical grid of a charging station is not dimensioned based on the charging times. It is dimensioned based on the number of cars that must be charged during a given time interval at that location (assuming a certain average charging energy). So regardless if fast chargers or slow chargers are used, what matters is how many electric cars are used in a region and how much they travel each day. Fast chargers can matter only indirectly, if their presence will convince more of the car users to switch to EVs, requiring the electrical power suppliers to take into account this increased consumption.
- close04 6mo ago
  > The 1-MW chargers have internal batteries, so they can pull a much lower average power from the electrical grid. We go back to what I was saying earlier [0], you'll need a lot of local storage to keep that charger running at full speed without breaking the back of the distribution grid. That comes on top of the solid grid capacity at every one of those many small parking lots. One stall alone needs 1MW maybe equally split between the grid and the local battery, for probably 50-70kWh per car. Just 3-4 of these cars charging in a parking lot would mean a constant multi-MW pull from the grid on top of the battery that also needs to be recharged from somewhere. To guarantee that you lower that average you need to have enough local storage. The technology exists, it works really well for large dedicated charging spots. The problem is the cost of scaling when you have to deal with lots of small plots (what OP proposed). A 1kW installation with 2-3kWh battery storage is the size of a small container and we've had them for years. Why do you think every small parking lot hasn't been equipped with one? > requiring the electrical power suppliers to take into account this increased consumption This doesn't scale like software. Adding production is comparatively easy, adding distribution capacity isn't. It needs a lot of equipment that's not really available [1], and a lot of construction work to expand the grid capacity. [0] https://news.ycombinator.com/item?id=47932115 https://news.ycombinator.com/item?id=47932115 [1] https://news.ycombinator.com/item?id=47604887 https://news.ycombinator.com/item?id=47604887

Rekindle8090 6mo ago

[dead]