7 ms·

That's over ~$750 for the power stored in a single Tesla battery pack. Not that the cars are capable of back-feeding the grid (V2G), unfortunately, nor is the i

by zaroth 7y ago

That's over ~$750 for the power stored in a single Tesla battery pack. Not that the cars are capable of back-feeding the grid (V2G), unfortunately, nor is the infrastructure in place to pull from them in times like this... It sure seems like if we're going to have millions of these 100kWh batteries plugged in, all nicely geographically distributed and internet connected, it would be swell if the owners could set a price at which they would be willing to sell their energy. It would just add another way that EVs are better than ICE cars.

taneq 7y ago
AC Propulsion was touting "vehicle-to-grid" as a feature way back in the 2000. At the time it seemed crazy with car batteries having cycle lives in the low hundreds, but it's looking better and better now...
- zaroth 7y ago
  I think if you could set your own price limit, and the software can account for it somehow in terms of de-rating the battery warranty, it would work. The big factor I'm not sure about is the added COGS to the car. This [1] seems to say it's only $200-$300 in added cost of parts in the car. It's not clear if that's in addition to the added cost in the charge port, or instead of. One important factor is even though the battery is capable of feeding perhaps up to 100 KW, if your charger is on a 40-amp 240v circuit, you are limited there to just 9.6kW. This drastically limits your ability to profit in the case of a spike, but then again, if there are several hundred thousand batteries competing with you, it's less likely to need to pull more than 10kW from any one battery. [1] - https://www.nrel.gov/docs/fy17osti/69017.pdf https://www.nrel.gov/docs/fy17osti/69017.pdf
nradov 7y ago
Over time it might be more practical for EV manufacturers to take back decommissioned vehicles and install the battery packs in centralized facilities for grid storage and stabilization. That would avoid the need for expensive and complex grid connection equipment in individual EVs. Even if the reused batteries are down to 60% capacity, stacking a bunch of them could deliver enough power for demand spikes.
- 0xB31B1B 7y ago
  Are you a battery engineer? My understanding is the EoL for Li Ion batteries is when they are at 80%. Below that, their internal resistance is too great to do anything useful with.
  - adammunich 7y ago
    Too great for cars but plenty fine for solar demands.
  - EmpirePhoenix 7y ago
    As someone recycling old laptop cells that are decades old (even got a few sony fukushima ones in last batch). Having a absolute large amount of them can cancel this out. Sure one fresh cell might have the small top current than 10 parallel old ones due to resistance, but if space is no concern the older ones are really great economically.
  - dsfyu404ed 7y ago
    Quantity has a quality all its own. They just have to have enough capacity to be cost positive after overhead.
- jillesvangurp 7y ago
  No need to wait until they are decommissioned with vehicle to grid. Charge at night at cheap rates, use during peak demand.
  - michaelt 7y ago
    You know how cell phone batteries wear out after a few years, because many li-ion batteries will lose 20% of their capacity after ~500† charge-discharge cycles - and users often have to throw away their phone in response? Vehicle-to-grid would have to pay me a lot if it means I have to scrap my EV after 5 years instead of 10. And if the power network is going to use half of my EV's battery's cycle-life and pay half my battery's cost, why wouldn't they just buy their own battery? And if they're only going to activate V2G for a few hours a year, it'll have to pay high hourly rate to cover the cost of the extra car components. On the other hand, just limiting charge rates in times of high demand doesn't degrade cars' batteries at all. Makes much more sense IMHO. †Yes, it depends on the battery model, the temperature, the current - but we all know from phones that battery degradation is a real thing.
reitzensteinm 7y ago
The new Leaf is V2G capable. Devote your battery to shaving peak load a few times a year and you're going a long way towards closing the price gap with an ICE. IIRC, Tesla decided against V2G years ago out of concern for battery life, but given that they're now claiming the batteries will outlive the cars that decision may change.
- goodcanadian 7y ago
  The old Leaf is, too. Pretty much any Leaf from 2012, onwards. If you are in the UK and have a Leaf with a 30kWh battery or larger, you can participate in a V2G trial: https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char... https://www.indra.co.uk/v2g https://www.indra.co.uk/v2g Technically, Leafs with 24kWh batteries work, too, but they aren't normally accepted in the trial as there is not enough capacity to be worthwhile.
Thev00d00 7y ago
Ovo Energy in the UK is trialling V2G with the Nissan Leaf [0] 0. https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char...
monkeynotes 7y ago
This sparks an interesting discussion around 'what if we all had EVs'. Texas would not be the only place struggling to meet demand. If everyone is fueling their cars with electricity, demand is going to be high in all cities. Without fossil fuels we need to find new sources for power. It's not good enough to say 'electricity', how do you generate it at a capacity that fills in for the 20.5 million barrels of oil consumed a day in the USA? The numbers are interesting. There is 17500 kw/h or 1.7 mw/h in a barrel of oil. Using 20.5m barrels a day, that's 34.85 million mw/h a day. The US currently generates around 10 million mw/h of electricity a day[1]. Granted, a fraction of a barrel of oil is used for gasoline, but you can see that the grid will need a sizeable upgrade if it is to fill in for our dependency on oil. Please forgive me if I have some of this wrong! [1] https://en.wikipedia.org/wiki/List_of_countries_by_electricity_consumption https://en.wikipedia.org/wiki/List_of_countries_by_electrici...
- zaroth 7y ago
  The biggest factor is how much of the EV charging can be done at algorithmically controlled times. If the load can be computer controlled to exactly fit the supply curve, the electricity becomes extremely cheap to provide. If that same load can basically instantly switch from charging, to paused, to discharging we end up with an extremely smart and efficient grid that can very effectively utilize high variance generation sources.
- Klathmon 7y ago
  EVs don't need to be charged at peak usage times. We have ours scheduled to charge starting at 11pm every day (which is just before the lowest demand part of the day in my area), and most EVs include this functionality. And as EVs become more common, I could absolutely see them being a useful tool in power generation. I'd absolutely allow the power company to control the home-charging of my car for a discount (and some agreement that i'd be able to charge to full at least once a day). The power company could enable more EVs to charge when there is a valley in demand, and turn some off when there are spikes. And all of that can happen without vehicle-to-grid systems which come with their own bag of tradeoffs. Also, keep in mind that gas engines aren't 100% efficient (IIRC they are under 50% efficient). So even though there is theoretically 1.7 mw/h of power in a barrel of oil (I'm taking your number as true here), I assume that cars are only using a fraction of that (especially after all the refinement as well, which I'm assuming cuts into that efficiency even more).
etimberg 7y ago
The big challenge with V2G is planning for it. Consider the case of a single distribution feeder with 10 vehicle chargers that each support V2G. There are infinite combinations of real power (P) drawn by the chargers. For simplicity though, let's limit to one of three states (0, full charging, full discharging). There are still 3^10 combinations for that one feeder. The utility needs to know that the system will work in all of those cases. Doing the analysis for that is quite complicated, and a lot of the tooling isn't there yet.

by zaroth 7y ago

taneq 7y ago

AC Propulsion was touting "vehicle-to-grid" as a feature way back in the 2000. At the time it seemed crazy with car batteries having cycle lives in the low hundreds, but it's looking better and better now...

zaroth 7y ago
I think if you could set your own price limit, and the software can account for it somehow in terms of de-rating the battery warranty, it would work. The big factor I'm not sure about is the added COGS to the car. This [1] seems to say it's only $200-$300 in added cost of parts in the car. It's not clear if that's in addition to the added cost in the charge port, or instead of. One important factor is even though the battery is capable of feeding perhaps up to 100 KW, if your charger is on a 40-amp 240v circuit, you are limited there to just 9.6kW. This drastically limits your ability to profit in the case of a spike, but then again, if there are several hundred thousand batteries competing with you, it's less likely to need to pull more than 10kW from any one battery. [1] - https://www.nrel.gov/docs/fy17osti/69017.pdf https://www.nrel.gov/docs/fy17osti/69017.pdf

nradov 7y ago

Over time it might be more practical for EV manufacturers to take back decommissioned vehicles and install the battery packs in centralized facilities for grid storage and stabilization. That would avoid the need for expensive and complex grid connection equipment in individual EVs. Even if the reused batteries are down to 60% capacity, stacking a bunch of them could deliver enough power for demand spikes.

0xB31B1B 7y ago
Are you a battery engineer? My understanding is the EoL for Li Ion batteries is when they are at 80%. Below that, their internal resistance is too great to do anything useful with.
- adammunich 7y ago
  Too great for cars but plenty fine for solar demands.
- EmpirePhoenix 7y ago
  As someone recycling old laptop cells that are decades old (even got a few sony fukushima ones in last batch). Having a absolute large amount of them can cancel this out. Sure one fresh cell might have the small top current than 10 parallel old ones due to resistance, but if space is no concern the older ones are really great economically.
- dsfyu404ed 7y ago
  Quantity has a quality all its own. They just have to have enough capacity to be cost positive after overhead.
jillesvangurp 7y ago
No need to wait until they are decommissioned with vehicle to grid. Charge at night at cheap rates, use during peak demand.
- michaelt 7y ago
  You know how cell phone batteries wear out after a few years, because many li-ion batteries will lose 20% of their capacity after ~500† charge-discharge cycles - and users often have to throw away their phone in response? Vehicle-to-grid would have to pay me a lot if it means I have to scrap my EV after 5 years instead of 10. And if the power network is going to use half of my EV's battery's cycle-life and pay half my battery's cost, why wouldn't they just buy their own battery? And if they're only going to activate V2G for a few hours a year, it'll have to pay high hourly rate to cover the cost of the extra car components. On the other hand, just limiting charge rates in times of high demand doesn't degrade cars' batteries at all. Makes much more sense IMHO. †Yes, it depends on the battery model, the temperature, the current - but we all know from phones that battery degradation is a real thing.

reitzensteinm 7y ago

The new Leaf is V2G capable. Devote your battery to shaving peak load a few times a year and you're going a long way towards closing the price gap with an ICE. IIRC, Tesla decided against V2G years ago out of concern for battery life, but given that they're now claiming the batteries will outlive the cars that decision may change.

goodcanadian 7y ago
The old Leaf is, too. Pretty much any Leaf from 2012, onwards. If you are in the UK and have a Leaf with a 30kWh battery or larger, you can participate in a V2G trial: https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char... https://www.indra.co.uk/v2g https://www.indra.co.uk/v2g Technically, Leafs with 24kWh batteries work, too, but they aren't normally accepted in the trial as there is not enough capacity to be worthwhile.

Thev00d00 7y ago

Ovo Energy in the UK is trialling V2G with the Nissan Leaf [0] 0. https://www.ovoenergy.com/electric-cars/vehicle-to-grid-charger https://www.ovoenergy.com/electric-cars/vehicle-to-grid-char...

monkeynotes 7y ago

This sparks an interesting discussion around 'what if we all had EVs'. Texas would not be the only place struggling to meet demand. If everyone is fueling their cars with electricity, demand is going to be high in all cities. Without fossil fuels we need to find new sources for power. It's not good enough to say 'electricity', how do you generate it at a capacity that fills in for the 20.5 million barrels of oil consumed a day in the USA? The numbers are interesting. There is 17500 kw/h or 1.7 mw/h in a barrel of oil. Using 20.5m barrels a day, that's 34.85 million mw/h a day. The US currently generates around 10 million mw/h of electricity a day[1]. Granted, a fraction of a barrel of oil is used for gasoline, but you can see that the grid will need a sizeable upgrade if it is to fill in for our dependency on oil. Please forgive me if I have some of this wrong! [1] https://en.wikipedia.org/wiki/List_of_countries_by_electricity_consumption https://en.wikipedia.org/wiki/List_of_countries_by_electrici...

zaroth 7y ago
The biggest factor is how much of the EV charging can be done at algorithmically controlled times. If the load can be computer controlled to exactly fit the supply curve, the electricity becomes extremely cheap to provide. If that same load can basically instantly switch from charging, to paused, to discharging we end up with an extremely smart and efficient grid that can very effectively utilize high variance generation sources.
Klathmon 7y ago
EVs don't need to be charged at peak usage times. We have ours scheduled to charge starting at 11pm every day (which is just before the lowest demand part of the day in my area), and most EVs include this functionality. And as EVs become more common, I could absolutely see them being a useful tool in power generation. I'd absolutely allow the power company to control the home-charging of my car for a discount (and some agreement that i'd be able to charge to full at least once a day). The power company could enable more EVs to charge when there is a valley in demand, and turn some off when there are spikes. And all of that can happen without vehicle-to-grid systems which come with their own bag of tradeoffs. Also, keep in mind that gas engines aren't 100% efficient (IIRC they are under 50% efficient). So even though there is theoretically 1.7 mw/h of power in a barrel of oil (I'm taking your number as true here), I assume that cars are only using a fraction of that (especially after all the refinement as well, which I'm assuming cuts into that efficiency even more).

etimberg 7y ago

The big challenge with V2G is planning for it. Consider the case of a single distribution feeder with 10 vehicle chargers that each support V2G. There are infinite combinations of real power (P) drawn by the chargers. For simplicity though, let's limit to one of three states (0, full charging, full discharging). There are still 3^10 combinations for that one feeder. The utility needs to know that the system will work in all of those cases. Doing the analysis for that is quite complicated, and a lot of the tooling isn't there yet.