## Your e car is increasing net carbon emissions – page 2 bp gas station

###

Something like each gallon of gas holds 35 kWh of energy but it takes 10 kWh to extract, 5 kWh to refine, 4 kWh to transport/distribute/store/pump, and 20% efficiency per mile driven. Versus an EV’s energy use based on 60% efficient natural gas combined cycle power plant, electricity transmission efficiency, all the conversion losses/heat losses associated with an electric drivetrain etc.

At the end of the day there are probably so many variables that it’s hard to say definitively but I suspect the basis for your EE prof’s position have changed and will change even more in the future. Petroleum extracted from Canadian oil sand, for example, consumes almost as much energy as it produces.

I’m totally certain the variables have changed. Batteries have improved since then and lithium vs. lead has advantages in energy density and efficiency of charge, discharge and storage, AFAIK (mangle can correct me, if not). So it might be a simple calc to repeat and I certainly see reason to hope we get there–EV motors are a lot less maintenance intensive, among other things.

Don’t overcomplicate it, though. The ICE side is simple if the energy spent getting gas to the station is roughly equivalent to the energy spent getting it from the same oil well to the power plant. Sure, there are a few more truck miles with gas stations as the end point, but if you think that doesn’t fall into the noise I’ll leave it to you to prove it–I’m sure Tesla will help with that. The comparison between levels of refinement might be of some interest, if you want to look at the final condition of all such fuels just before they’re burned. (What The Suit said.)

If you accept that trucks to gas stations are a minimal impact on total delivery efficiency, the comparison to EV’s is just straight efficiency. Just multiply the efficiencies of all the steps from the power plant to the motor and you have a good comparison:

E(petro power plant) x E(electric generator) x E(transformers) x E(power lines) x E(local transformers) x E(charging system) x E(battery, charge) x E(battery, discharge) x E(motor) = E(EV, net). (I might have missed a detail or two–obviously the net for all power lines has to go into that one value etc.) Pulling numbers out of my ass, if it’s 0.6×0.9×0.9×0.6×0.9×0.9×0.9×0.9×0.9 that’s 17.2% efficient from petro fuel to the car’s transmission, which should compare directly with the ICE efficiency in the low 20’s, before factoring in things like higher energy requirements for hauling more weight around (because gas still wins on energy density). Vehicle weight should not be underestimated, however, and the reality is that ICE suffers primarily from an image problem thanks to big vehicles always being ICE powered. Resist the temptation to compare F150’s to Volts.

I don’t think you can prevent oil sands extraction without reducing demand enough to make it a losing proposition, so it’s hard to have a clear path to comparing the next/marginal Joule of electricity either way. In any case, if those oil sands are used to power electric plants that need a higher net amount of fuel, then we’re better off burning less in ICE cars regardless. It’s really not a solution to try to talk about the problems on the supply side as if they are integral to the demand side.

I always find it interesting that most of the societal discussion of vehicles and emissions and fuel consumption revolves around personal transportation, when (in my estimation) commercial vehicles are a much more significant hurdle to the overall problem.

In my little corner of the stupid transportation world, I often think about how much more efficient it would be to subsidize a large fleet of Uber/Lyft type cars than to just drive behemoth 5mpg diesel pigs all around the county all day long just in case somebody wants to ride somewhere on the fixed route. We’ll burn 100 gallons of diesel to bring about 25 scuzzers from the methlands into town where they cause trouble for working people all day, then bring them home. The only way I can imagine this being justifiable, in a big-picture sense, is just a vague sense that by doing this with federal money we’re kind of stealing back some of the resources the rest of the country takes out of our local mines and forests. It’s still a wildly inefficient way of allocating those units of money/energy/human-endeavor…etc.

Does the EPA tell you what part of the system they’re considering for that equivalency? For instance, are we talking about just the fully-charged car or does that include losses between the meter and the battery? And hopefully this is obvious, but EPA FE numbers should not be compared with real world numbers at this point. Unpacking that seems like an even bigger can of worms than doing the math yourself.

I’ve been trying to find a good number for charge/discharge efficiency for lithium ions but haven’t spotted a recent one yet. The Tesla forums do contain some discussion of metered electricity required to fully charge a battery, but I don’t see where anyone has measured it, unfortunately. Been almost 20 years since I was involved with an EV project and at that time the controller and battery recharge efficiency were so bad that the electronics folks opted to skip regenerative braking. So I have to think modern controllers are better, but also expect that might vary wildly.

5% annual losses in transmission of electricity in the U.S. seems to be a commonly quoted figure. This might be higher for CA if a lot of the PNW’s hydro is making the long trip south–but of course the bitcoin miners will probably use that up near the source before long, so problem solved there.

Does the EPA tell you what part of the system they’re considering for that equivalency? For instance, are we talking about just the fully-charged car or does that include losses between the meter and the battery? And hopefully this is obvious, but EPA FE numbers should not be compared with real world numbers at this point. Unpacking that seems like an even bigger can of worms than doing the math yourself.

I’ve been trying to find a good number for charge/discharge efficiency for lithium ions but haven’t spotted a recent one yet. The Tesla forums do contain some discussion of metered electricity required to fully charge a battery, but I don’t see where anyone has measured it, unfortunately. Been almost 20 years since I was involved with an EV project and at that time the controller and battery recharge efficiency were so bad that the electronics folks opted to skip regenerative braking. So I have to think modern controllers are better, but also expect that might vary wildly.

5% annual losses in transmission of electricity in the U.S. seems to be a commonly quoted figure. This might be higher for CA if a lot of the PNW’s hydro is making the long trip south–but of course the bitcoin miners will probably use that up near the source before long, so problem solved there.

Yes, you can read EPA’s modeling assumptions and testing methodology for the MPG equivalency window sticker. It’s not a black box, and it is backed by empirical testing, though it is also subject to modeling assumptions to go from watts to gallons of gas.

I don’t think I agree that you are well-positioned to do better modeling than the EPA testers, though if you think you’re better at it than they are, have at it. Just keep in mind that if it’s not your field (and sub-field) of expertise, you are probably going in with a lot of false confidence.