8 ms·

The electric grid is a big bottleneck that I don't think has a solution yet. At some point EVs will create stress on the grid, slowing down the adoption and lea

by proc0 4y ago

The electric grid is a big bottleneck that I don't think has a solution yet. At some point EVs will create stress on the grid, slowing down the adoption and leaving the state of this transition in the worst of both worlds--costly electricity but many people still using gas. IMHO, the #1 priority should be cheaper electricity. Everything else follows from that, like more demand for EVs. It seems focusing on EVs too much is putting the horse before the cart, or w/e that saying is.

bombcar 4y ago
Plugin hybrids would bridge that gap pretty effectively- especially for short trips where it could go battery only.
- proc0 4y ago
  Yeah, I agree, hybrids could be a good transition while the grid is upgraded.
  - bombcar 4y ago
    What I'd like to see is a special panel (say when replacing your 100 amp service) that has one side defined as "always available" and another side defined as "utility brownout" and you could put things like the fridge and lights on "always available" and the AC and car charger on "utility brownout". They have these devices for AC units already, but it could be so much smarter.
bryanlarsen 4y ago
EV's mostly charge at night, and the grid is sized to handle air conditioners during the day time. So if the grid can handle air conditioning during the day time, it should be fine.
- CobaltFire 4y ago
  This is not the full story; the grid is sized to allow peak load heat shed during the cooler night hours; if we do not have those hours to cool it we start seeing a drastic increase in failures. I've heard some talk that some of the PG&E caused fires are due to the grids not getting enough time to cool and accelerating failures. Note that this doesn't excuse their lack of investment in infrastructure. Basically, we need to hugely over provision the grid to handle 24/7 loading without the current failure modes. That will require a complete rethink on what constitutes the base load, how solar can help and how it can't, etc.
  - KingOfCoders 4y ago
    "I've heard some talk that some of the PG&E caused fires are due to the grids not getting enough time to cool" Source?
    - CobaltFire 4y ago
      Nothing that I could reference or I would have put it in there. I have family in some of the effected areas (Marin and Navato) and that's one of the discussion points. It seems plausible but I have nothing more than that. Edit: I don't think the grid being hot would cause a fire, to be clear. I'm under the impression that it causes the lines to sag, increasing the occurrence of the primary mode of wildfire cause, vegetation contact. Further, running the transformers hotter shortens their life (as it does with most electrical equipment).
  - 0cf8612b2e1e 4y ago
    Do renewables also require cooling times or is that restricted to natural gas/coal plants? If we continue on this pace of installing so much solar will the peak plants get used less frequently? Or was that just sloppy language on peak vs base loads? I have not seen anything near-term realistic on removing fossil fuel base load generation.
    - CobaltFire 4y ago
      I think you may have extended what I'm talking about in ways I didn't intend; there was no sloppy language. The transmission infrastructure needs cooling off time, not the generation. Base load has to be completely changed if we consider charging EV's at a population level, which changes how we look at renewables that aren't around the clock (i.e. solar). To be clear I'm 100% behind getting rid of ICE, coal, gas, etc. as soon as we can. That doesn't mean there aren't assumptions that won't have to be challenged. As for my background and experience, I worked in Nuclear Power generation. I'm not a complete domain expert, but neither am I speaking from a place of no experience.
- eldaisfish 4y ago
  on what basis are you assuming that EVs charge mostly at night? Trucks, offices etc are easy examples to point to.
bunabhucan 4y ago
"If they perform scheduled charges for times in which electricity is cheaper, the US power grid could deal with 65 million EVs." https://insideevs.com/news/436665/24-million-evs-limit-current-power-grid/ https://insideevs.com/news/436665/24-million-evs-limit-curre... 2021 sales were less that 1% of that. At what point does the bottleneck kick in? I would imagine electric utilities have a huge incentive to make this work given that demand has been flat at ~4 trillion kWh for over a decade.
- KingOfCoders 4y ago
  Yes + people always forget second order effects. Increasing EV numbers will lead to changes in the power grid. It's like saying in 1886 "If all horses would be replaced by cars, the petroleum industry would collapse".
- asdff 4y ago
  That's a little concerning that its only 65 million. For comparison there are almost 300 million cars in the U.S., so this would be just a 20% adoption rate we are capable of. The reason why I am concerned is not for lack of power, its that I know for a fact the red states in this country will favor making up the slack with fossil fuel based power, because that's significantly faster and cheaper to construct than the best alternative, nuclear power. Even in california nuclear power can hardly catch a break, diablo canyon is limping along, they've already closed san onofre, and I don't think any new power plant is in the pipeline currently.
  - kevinpet 4y ago
    I think your arguments would be more effective if you removed the tribalism from your politics. Leftists are not consistent friends to nuclear power and in fact nuclear power generation does not seem to have any relation to a states partisan lean: https://www.statista.com/statistics/187270/us-nuclear-power-electricity-generation-by-state/ https://www.statista.com/statistics/187270/us-nuclear-power-...
    - asdff 4y ago
eldaisfish 4y ago
You make a good point but this sort of thinking is unfortunately dismissed here by those who view technology as the solution to all problems. Your observation is correct - the grid as designed is indeed a bottleneck and this will become increasingly obvious in northern climates as more services switch to electric - heat and transportation are good examples. An excellent case in point is the trans canada highway which has failed to electrify large sections due to insufficient grid capacity. Before people point to a map from an auto maker, do those maps show instances where the power feeding EV chargers has overloaded a circuit causing multiple hours of no charging? Cheaper electricity should indeed be a short term goal.
bluGill 4y ago
The grid operators are watching EVs coming and making plans. The limit is regulation, not technical. Often the law doesn't allow them to build the lines where they need to.
- eldaisfish 4y ago
  this is not complete. the limit is often technical because EVs and electric heating increase the energy and peak power demands on the grid. Grid operators also do not have infinite capacity to retire entire neighbourhoods to support the increased power demands of increased HVAC capacity combined with EV capacity.
  - KingOfCoders 4y ago
    EVs that are not used acting as local batteries decrease the demand on the grid. If every car in Germany would be an EV the battery capacity could power all of Germanys electric power needs for two days. Things are more complex than they seem to be at first sight.
    - eldaisfish 4y ago
      EVs = grid storage gets thrown around a lot. There is currently no viable pricing model for these services, they have not been trialed at any scale and there is little to no legal framework with either vehicle manufactures or battery suppliers in terms of the warranty. Your claim of batteries powering a gigawatt scale grid for two days does not pass the smell test. Something is off so it might help to show calculations or sources.
      - KingOfCoders 4y ago
        "Your claim of batteries powering a gigawatt scale grid for two days does not pass the smell test." What smell test? What doesn't specifically sound right? I might have miscalculated of course, 50M cars x 50kwh = 2500Gwh = 2.5Twh Germany Electricty usage: 561,8 Twh/ year -> /365 = 1.54Twh / day So two days ~3TwH, Battery capacity ~2.5Twh I would be interested in any corrections, especially as you are a renewable energy researcher and I'm just a coder.
benj111 4y ago
As discussed in the article these EVs only added 0.15% to electricity demand, so at 100% EVs were talking 30% extra demand. I don't know about the US, but in the UK electricity demand has dropped ~20% in the past 10 years. (10% in 7 years if you want to avoid covid etc) Adding in the fact that EVs are more likely to be charged off peak and I'd be surprised if it made any appreciable difference.

by proc0 4y ago

bombcar 4y ago

Plugin hybrids would bridge that gap pretty effectively- especially for short trips where it could go battery only.

proc0 4y ago
Yeah, I agree, hybrids could be a good transition while the grid is upgraded.
- bombcar 4y ago
  What I'd like to see is a special panel (say when replacing your 100 amp service) that has one side defined as "always available" and another side defined as "utility brownout" and you could put things like the fridge and lights on "always available" and the AC and car charger on "utility brownout". They have these devices for AC units already, but it could be so much smarter.

bryanlarsen 4y ago

EV's mostly charge at night, and the grid is sized to handle air conditioners during the day time. So if the grid can handle air conditioning during the day time, it should be fine.

CobaltFire 4y ago
This is not the full story; the grid is sized to allow peak load heat shed during the cooler night hours; if we do not have those hours to cool it we start seeing a drastic increase in failures. I've heard some talk that some of the PG&E caused fires are due to the grids not getting enough time to cool and accelerating failures. Note that this doesn't excuse their lack of investment in infrastructure. Basically, we need to hugely over provision the grid to handle 24/7 loading without the current failure modes. That will require a complete rethink on what constitutes the base load, how solar can help and how it can't, etc.
- KingOfCoders 4y ago
  "I've heard some talk that some of the PG&E caused fires are due to the grids not getting enough time to cool" Source?
  - CobaltFire 4y ago
    Nothing that I could reference or I would have put it in there. I have family in some of the effected areas (Marin and Navato) and that's one of the discussion points. It seems plausible but I have nothing more than that. Edit: I don't think the grid being hot would cause a fire, to be clear. I'm under the impression that it causes the lines to sag, increasing the occurrence of the primary mode of wildfire cause, vegetation contact. Further, running the transformers hotter shortens their life (as it does with most electrical equipment).
- 0cf8612b2e1e 4y ago
  Do renewables also require cooling times or is that restricted to natural gas/coal plants? If we continue on this pace of installing so much solar will the peak plants get used less frequently? Or was that just sloppy language on peak vs base loads? I have not seen anything near-term realistic on removing fossil fuel base load generation.
  - CobaltFire 4y ago
    I think you may have extended what I'm talking about in ways I didn't intend; there was no sloppy language. The transmission infrastructure needs cooling off time, not the generation. Base load has to be completely changed if we consider charging EV's at a population level, which changes how we look at renewables that aren't around the clock (i.e. solar). To be clear I'm 100% behind getting rid of ICE, coal, gas, etc. as soon as we can. That doesn't mean there aren't assumptions that won't have to be challenged. As for my background and experience, I worked in Nuclear Power generation. I'm not a complete domain expert, but neither am I speaking from a place of no experience.
eldaisfish 4y ago
on what basis are you assuming that EVs charge mostly at night? Trucks, offices etc are easy examples to point to.

bunabhucan 4y ago

"If they perform scheduled charges for times in which electricity is cheaper, the US power grid could deal with 65 million EVs." https://insideevs.com/news/436665/24-million-evs-limit-current-power-grid/ https://insideevs.com/news/436665/24-million-evs-limit-curre... 2021 sales were less that 1% of that. At what point does the bottleneck kick in? I would imagine electric utilities have a huge incentive to make this work given that demand has been flat at ~4 trillion kWh for over a decade.

KingOfCoders 4y ago
Yes + people always forget second order effects. Increasing EV numbers will lead to changes in the power grid. It's like saying in 1886 "If all horses would be replaced by cars, the petroleum industry would collapse".
asdff 4y ago
That's a little concerning that its only 65 million. For comparison there are almost 300 million cars in the U.S., so this would be just a 20% adoption rate we are capable of. The reason why I am concerned is not for lack of power, its that I know for a fact the red states in this country will favor making up the slack with fossil fuel based power, because that's significantly faster and cheaper to construct than the best alternative, nuclear power. Even in california nuclear power can hardly catch a break, diablo canyon is limping along, they've already closed san onofre, and I don't think any new power plant is in the pipeline currently.
- kevinpet 4y ago
  I think your arguments would be more effective if you removed the tribalism from your politics. Leftists are not consistent friends to nuclear power and in fact nuclear power generation does not seem to have any relation to a states partisan lean: https://www.statista.com/statistics/187270/us-nuclear-power-electricity-generation-by-state/ https://www.statista.com/statistics/187270/us-nuclear-power-...
  - asdff 4y ago

eldaisfish 4y ago

You make a good point but this sort of thinking is unfortunately dismissed here by those who view technology as the solution to all problems. Your observation is correct - the grid as designed is indeed a bottleneck and this will become increasingly obvious in northern climates as more services switch to electric - heat and transportation are good examples. An excellent case in point is the trans canada highway which has failed to electrify large sections due to insufficient grid capacity. Before people point to a map from an auto maker, do those maps show instances where the power feeding EV chargers has overloaded a circuit causing multiple hours of no charging? Cheaper electricity should indeed be a short term goal.

bluGill 4y ago

The grid operators are watching EVs coming and making plans. The limit is regulation, not technical. Often the law doesn't allow them to build the lines where they need to.

eldaisfish 4y ago
this is not complete. the limit is often technical because EVs and electric heating increase the energy and peak power demands on the grid. Grid operators also do not have infinite capacity to retire entire neighbourhoods to support the increased power demands of increased HVAC capacity combined with EV capacity.
- KingOfCoders 4y ago
  EVs that are not used acting as local batteries decrease the demand on the grid. If every car in Germany would be an EV the battery capacity could power all of Germanys electric power needs for two days. Things are more complex than they seem to be at first sight.
  - eldaisfish 4y ago
    EVs = grid storage gets thrown around a lot. There is currently no viable pricing model for these services, they have not been trialed at any scale and there is little to no legal framework with either vehicle manufactures or battery suppliers in terms of the warranty. Your claim of batteries powering a gigawatt scale grid for two days does not pass the smell test. Something is off so it might help to show calculations or sources.
    - KingOfCoders 4y ago
      "Your claim of batteries powering a gigawatt scale grid for two days does not pass the smell test." What smell test? What doesn't specifically sound right? I might have miscalculated of course, 50M cars x 50kwh = 2500Gwh = 2.5Twh Germany Electricty usage: 561,8 Twh/ year -> /365 = 1.54Twh / day So two days ~3TwH, Battery capacity ~2.5Twh I would be interested in any corrections, especially as you are a renewable energy researcher and I'm just a coder.

benj111 4y ago

As discussed in the article these EVs only added 0.15% to electricity demand, so at 100% EVs were talking 30% extra demand. I don't know about the US, but in the UK electricity demand has dropped ~20% in the past 10 years. (10% in 7 years if you want to avoid covid etc) Adding in the fact that EVs are more likely to be charged off peak and I'd be surprised if it made any appreciable difference.