marginal electricity overnight?

derekt75 · Feb 1, 2016

When I plug my car in overnight in California, what is the carbon emissions of the energy that comes on line to provide energy to my batteries?

One can look up what the carbon emissions per MWh are for California where I live.
but using that to talk about my car's emissions are a little goofy. When I plug in my car, no more power is generated from large hydro. We get what we can get from the dam. Similarly, my roof's photovoltaics generate what they generate so I think it's inaccurate to say that my car is powered by my PV.

So what I want to know is if I telecommuted instead of driving to work, how much carbon would actually be saved?
Do they run a nuke 10 kW hotter overnight to meet my car's demand? or do they run a natural gas plant 10 kW hotter? or something else?

AntronX · Feb 1, 2016

Its hard to know exact amount, but we can approximate.

Look at CAISO system status page. There you will see two graphs:

1. Total demand:

2. Renewable contribution:

Lets say you charge your car between midnight and 5am. During that time, approximate average demand was 21GW. During same time, average renewable power was 2.2GW wind, 1.05GW geothermal, 0.2GW biomass and 0.35GW small hydro. Total renewable 3.8GW. Other 0-carbon sources are large hydro and nuclear. Nuclear runs at constant power or about 2GW. Hydro is tricky, but lets average yearly generation data from here and get 1.4GW (in 2014).

Add it all up: 3.8+2+1.4=7.2GW non-fossil sources. Then divide this by total demand: 7.2/21=0.343 or 34.3% of electricity was from non-fossil sources during 12am-5am on February 1st 2016. The other 65.7% came from natural gas and imports. Electricity from natural gas has carbon intensity of about 600 grams CO2 per KWh. Imported power is a mix of everything, so its safe to assume national average of about 600g/KWh as well. Then simply multiply 600g by 65.7% and get 394g/KWh average carbon intensity of electricity delivered to your outlet during the time frame above. From here, just do the math of the miles you driven and you get your carbon emissions avoided by telecommuting from home.

nwdiver · Feb 1, 2016

This is a bit of a complicated question... it's not going to be the same everyday. Nuclear power is base load and isn't likely to vary much. The only real way to charge your car without increasing the use of fossil fuels would be if wind, solar, nuclear or hydro was curtailed due to lack of demand... something that is currently unlikely unless you're in Texas. Even if you're charging it directly from a PV array you're not displacing fossil fuels that would have been displaced if that power had be exported.

IMO... don't worry about it. The day will come that there will be solar curtailment during the day and wind curtailment during the night. At that point it will become important to charge your car at the appropriate time. As romantic as the notion that the energy your car uses is clean the real objective needs to be displacement. If you're displacing more fossil fuels with PV exports than you're burning with grid imports then your net effect is <0.

CHG-ON · Feb 1, 2016

I think we are mainly powered by nat gas in the south bay. So more emissions. But a heck of a lot better than much of the country that is on coal.

derekt75 · Feb 1, 2016

AntronX said:
Its hard to know exact amount, but we can approximate.
...

Then simply multiply 600g by 65.7% and get 394g/KWh average carbon intensity of electricity delivered to your outlet during the time frame above. From here, just do the math of the miles you driven and you get your carbon emissions avoided by telecommuting from home.

Thanks AntronX,
but I'm not that interested in the "average"; I'm interested in the marginal usage.
If I drive to San Francisco and back, how much carbon have I emitted compared to not making the drive.

From the looks of it, we're always using as much of the non-emitting power as we can. We can't get any more from existing nuclear plants, we can't get any more from existing wind turbines [even if I don't understand why half the Altamont Pass turbines seem idle -- disrepair?], and we can't get more energy from hydro (c'mon El Nino). So any marginal usage seems to be natural gas usage.

On the whole, I'm not super-worried about it.
If I really cared, I would have bought a small house with no pool and sent a huge check to Terrpass.com rather than buying a huge house + pool + PV.
but I am interested in a fair comparison against HFCV advocates.

From my math, a Model S powered by a natural gas power plant is roughly as good in CO2/mile as a Prius powered by gasoline. A HFCV running on steam reformed natural gas has about twice the emissions per mile as either the Prius or Tesla (worse than an accord), and a HFCV running on electrolysis is around 4X the emissions (worse than an Escalade). The canard is that HFCVs can be powered by "clean electricity", but as far as I can tell, there is no marginal clean electricity in California. If you want an extra kWh of electricity, you need to cause the emission of 600 g of CO2. Sure you can build more wind/solar, and every bit helps, but you'd need to build a lot more wind/solar/nuclear before using an extra kWh of energy didn't result in a natural gas plant working harder. and as long as that's true, electrolysis of water is dumb.
[and if we know that hydrogen must be generated from methane, why do we bother stripping the carbon off of the methane when CNG is a perfectly viable fuel... but that's another topic].

I think it is interesting, though, that the choice to take to lunch my Tesla or my friend's Prius probably has no real effect on CO2 emissions.

My "marginal" discussion might have a logic hole:
my increased usage might cause them to build a new wind/solar/nuke somewhere.
but I suspect that's unlikely if new power plants are built to meet peak demand, and my car is intentionally charged at trough demand.

nwdiver · Feb 1, 2016

derekt75 said:
..... but I am interested in a fair comparison against HFCV advocates.

I think it is interesting, though, that the choice to take to lunch my Tesla or my friend's Prius probably has no real effect on CO2 emissions.

My "marginal" discussion might have a logic hole:
my increased usage might cause them to build a new wind/solar/nuke somewhere.
but I suspect that's unlikely if new power plants are built to meet peak demand, and my car is intentionally charged at trough demand.

More important that where we are now is the direction in which we're heading... from an energy use perspective.

If HFVCs are to be sustainable then they must shift from gas reforming to electrolysis... a 'sustainable' HFVCs will use >2x as much energy per mile compared to an EV... that's an immutable part of the underlying physics.

EVs will play a crucial role in expanding wind and solar beyond what they would be capable of without a source of 'dispatchable demand'; That's a role that a Prius can never play... unless they get a much bigger battery.

The day that driving the Tesla vs a Prius is indisputably better is not far off.... EVs are the only cars that get cleaner as time passes :wink:

Robert.Boston · Feb 4, 2016

From a truly short-term marginal perspective, you can safely assume in California that the marginal source of electricity is natural gas. The advantage of charging overnight is that the energy will come from highly efficient units that are underutilized at night. If you charged during the morning or evening "ramp" period, you would be using some of the least efficient generators in the state.

From a longer-term perspective, sustained growth in energy usage from EVs triggers more purchases under the Renewable Energy Portfolio (which is a % of usage). So, although your energy tonight comes from natural gas, eventually the system will catch up and build more renewable energy sources.

Jeff N · Feb 4, 2016

I would assume both overnight and daytime marginal use comes from NG in California (which uses almost no coal) so as a rough approximation your use at night is being offset by the use avoided during the day. Of course, that assumes you only use your electricity for charging your car and your solar panels generate exactly the amount that you use to charge your car. In reality, you use electricity for tasks around the house and your solar panels may generate less than your total usage.

techmaven · Feb 4, 2016

Yeah, I looked at this issue over and over. Hydro and nuclear are tapped out and run at 100% overnight and demand far exceeds their production in almost all areas. There are certainly exceptions. For many people, the marginal production is going to be either coal or natural gas. Now wind is interesting. There are apparently times where wind produces a lot of electricity, but the grid can't absorb it and the spot price goes negative. There is some push back on adding wind in some places as a result even with high wind potential. Therefore, adding widespread EV charging would provide a reliable demand for additional wind power. Also, there is an issue of idling. Apparently the vast majority of coal plants can only bring down their output generation so far or risk much higher damage as a result. There is a lot of literature on cycling coal plants and the costs associated with it. What I can't find is whether or not BEVs are such a low power draw right now that we are soaking up the power produced by coal plants that would burn that coal anyways... in other words, they have to run that plant at a certain level, and if we don't use that electricity, it just gets wasted. I can't find any metrics on that, but it does appear to happen.

Therefore, one of the big take aways is that by adding reliable consumption at super off-peak hours, we can help transform the grid with better, more sustainable baseload power. It makes economic justifications to keep open nuclear in some cases, to add wind power, to add pumped-hydro, and even if that's coal, there is a lot to be gained by reducing cycling and installing scrubbers.

Jeff N · Feb 4, 2016

derekt75 said:
From my math, a Model S powered by a natural gas power plant is roughly as good in CO2/mile as a Prius powered by gasoline. A HFCV running on steam reformed natural gas has about twice the emissions per mile as either the Prius or Tesla (worse than an accord), and a HFCV running on electrolysis is around 4X the emissions (worse than an Escalade). The canard is that HFCVs can be powered by "clean electricity", but as far as I can tell, there is no marginal clean electricity in California. If you want an extra kWh of electricity, you need to cause the emission of 600 g of CO2. Sure you can build more wind/solar, and every bit helps, but you'd need to build a lot more wind/solar/nuclear before using an extra kWh of energy didn't result in a natural gas plant working harder. and as long as that's true, electrolysis of water is dumb.

A Tesla Model S powered by NG electricity is probably somewhat better than a Prius on gasoline.

A 2016 Prius on gas is about 47 pounds of CO2 for every 100 miles when you include GREET upstream emissions from refining etc. If a Model S is about 100 MPGe then it takes about 33.7 kWh to drive 100 miles. If you take that 47 pounds and divide it by 33.7 kWh that works out to 1.4 pounds per kWh or a little more than 600 grams per kWh. 100 MPGe is about 3 miles per kWh so we are talking about roughly 200 grams per mile. NG electricity with upstream emissions probably isn't worse than that.

A HFCV like the Toyota Mirai running on steam reformed NG hydrogen is not twice that level. According to a recent NREL report, it is roughly about the same (240 grams per mile).

http://www.nrel.gov/docs/fy14osti/60528.pdf

Here's a handy graph from that report:

A Mirai on electrolysis hydrogen is about 570 grams per mile or roughly 3 times worse than a Prius on US grid electricity or about the same as a Cadillac Escalade at 17 mpg. Presumably, the emissions of a distributed electrolysis hydrogen system (made at the local "gas" station) might be at least partially offset by renewable generation like wind or solar installed offsite.

Modeling all of this with marginal vs average electricity generation is a bit of a philosophical topic. Once plugins become widespread, why would you consider their energy use to be generated at the margin while other household use like furnace fans, electric ovens, and lighting are presumably considered to use non-marginal generation. All electricity use can't be at the margin....

derekt75 · Feb 12, 2016

yeah, so your numbers agree that a BEV running on US mix or on Natural gas (which are similar in carbon intensity at 550-600gCO2/kWh) emits around 240gCO2/mile while a Prius emits around 214gCO2/mile, right?
I'll grant that your report shows that a Mirai running on SMR H2 is much better than I was thinking, but I think that graph still shows it to be worse than a Prius, so I still don't see why Toyota is bothering to make an expensive hard to refuel vehicle that's strictly worse than a Prius.
and the graph agrees with my general conclusion that electrolysis from the grid or from methane is stooopid. However it then seems to argue that electrolysis from wind is good, which I think is a completely specious argument because that wind electricity could be used for y'know powering electric stuff so you don't need to burn more CH4 to power your electric stuff.

As for what is marginal and what isn't, when considering changes (e.g., driving to work or telecommuting), I do tend to think that all usage is marginal. When our BEV energy consumption exceeds our fossil fuel electricity consumption, I'll grant that it's no longer marginal. but this is a very long time out. and yes, when deciding what appliances to turn off, I think about all of my electric usage at my marginal electric rate and the marginal CO2 contribution: if I'm considering the carbon savings of a more efficient air conditioner, I'm going to look at my marginal rate of CO2 contribution.

rolosrevenge · Feb 12, 2016

Sometimes at night the wind will be blowing and force other generators into suboptimal operating points, or the wind will be slightly curtailed. Charging your EV on those nights actually ends up reducing green house gases.

jerry33 · Feb 13, 2016

Jeff N said:
A 2016 Prius on gas is about 47 pounds of CO2 for every 100 miles when you include GREET upstream emissions from refining etc.

Note that this is only applicable when the Prius is new. As it ages, it gets worse. The Tesla would get better as it ages.

Jeff N · Feb 13, 2016

jerry33 said:
Note that this is only applicable when the Prius is new. As it ages, it gets worse. The Tesla would get better as it ages.

Not within typical vehicle "lifetime" like 150,000 miles or whatever. Prius owners do not generally talk about noticeably reduced mpg as their cars get older. Consumer Reports did an article a few years ago comparing an older 2001 Prius against their original mpg tests when it first came out and they found no significant reduction in mpg results on the used Prius cars tested. Of course, as the Tesla ages, it may need a new battery eventually which will imply a CO2 manufacturing overhead.

On the other hand, over time, we all think gasoline will be more likely to be made with harder to refine crude while the electricity grid will be using less coal and more renewables.

techmaven · Feb 13, 2016

Jeff N said:
Not within typical vehicle "lifetime" like 150,000 miles or whatever. Prius owners do not generally talk about noticeably reduced mpg as their cars get older. Consumer Reports did an article a few years ago comparing an older 2001 Prius against their original mpg tests when it first came out and they found no significant reduction in mpg results on the used Prius cars tested. Of course, as the Tesla ages, it may need a new battery eventually which will imply a CO2 manufacturing overhead.

On the other hand, over time, we all think gasoline will be more likely to be made with harder to refine crude while the electricity grid will be using less coal and more renewables.

At 300,000 to 500,000+ useful miles, the replacement of a Tesla battery is too far out to really model effectively especially compared to equivalent gasoline cars. How many people go through an engine overhaul with a BMW M5? That cost is roughly $10,000 to $15,000 and so many folks don't. It is likely that a Tesla battery will outlast the competitor engines for the most part.

Of course, I assume jerry33 is talking about how the grid is getting cleaner with less coal plants over time. The issue with coal electricity production for BEVs is actually not CO2 - compared to an equivalent V6 or V8 engine powered German or American sports sedan, even at 80% coal, a Model S has less CO2 emissions. The problem is actually NOx and SOx emissions and good filtering on coal plants is necessary to bring those emissions down. The U.S. national average is quite poor compared to the best coal plants in the nation. In some cases, the dirtiest coal plants are the ones getting shut down since they aren't economically feasible to retrofit with better filtering/scrubbers.

One of the problems with natural gas production is the amount of methane that is leaked and is often uncounted in GHG calculations. But certainly, the other pollution emissions are far better with NG.

jerry33 · Feb 14, 2016

Jeff N said:
Not within typical vehicle "lifetime" like 150,000 miles or whatever. Prius owners do not generally talk about noticeably reduced mpg as their cars get older.

Having owned a couple of Prius for 100K to 160K miles I wasn't referring to mpg (which on the 2004 got better each year). I was referring to pollution (not CO2, which is based mostly on mpg). Towards the end the 2001 used quite a bit of oil. The 2004 used some.

omgwtfbyobbq · Feb 14, 2016

Was it leaking oil or burning oil? I don't think oil consumption creates that much more pollution than burning any other hydrocarbon.

jerry33 · May 22, 2016

Jeff N said:
Not within typical vehicle "lifetime" like 150,000 miles or whatever.

I was referring to overall emissions, which will go up as the car ages. CO2 is not the only pollutant.

Search

marginal electricity overnight?

derekt75

Member

AntronX

Member

nwdiver

Well-Known Member

CHG-ON

Still in love after all these miles

derekt75

Member

nwdiver

Well-Known Member

Robert.Boston

Model S VIN P01536

Jeff N

Active Member

techmaven

Active Member

Jeff N

Active Member

derekt75

Member

rolosrevenge

Dr. EVS

jerry33

(S85-3/2/13 traded in) X LR: F2611##-3/27/20

Jeff N

Active Member

techmaven

Active Member

jerry33

(S85-3/2/13 traded in) X LR: F2611##-3/27/20

omgwtfbyobbq

Active Member

jerry33

(S85-3/2/13 traded in) X LR: F2611##-3/27/20

Similar threads