Hacker News Reader

Comment by Robotbeat | Hacker News Reader

The climate argument is not "weak for electric cars." Electric cars are about a factor of 2.5 better in terms of emissions per mile even with the current US electric grid.

A 35mpg car emits about 250 grams of carbon per mile (a gallon of gasoline emits about 20 pounds of CO2).

US emissions per kWh using the most recent data is about 400-410 grams per kWh. A Model 3 gets about 4.1 miles per kWh. So each mile emits just 100 grams.

And in California (which has the most EVs of any state by far), where the electricity carbon intensity is more like 200 grams per kWh, it's just 50 grams per mile in the Model 3, or about a factor of 5 better than the gasoline car.

View on HN

I've seen compelling arguments that driving a Model 3 actually even has lower average CO2 emissions per mile than riding a bike, since riding a bike requires eating extra calories to power it, and food has a substantial carbon footprint.

Depends on the type of food you're eating, your local electricity grid, and many other factors of course. And obviously this ignores both manufacturing costs and the physical and psychological benefits to biking, so I'm not saying everyone should ditch biking, but it's an interesting comparison.

https://twitter.com/Robotbeat/status/1102249291568029696

That was me. :) The assumption was that if you have a particularly high emissions source of most of your calories (like virtually all meat especially ruminants or, interestingly, fresh fruit and not much grains, sugars, or oils), then biking could be worse. But it's just to point out how efficient electric cars can be and how diet can start to play a bigger role.

I wasn't super serious about the comparison. :) (And e-bikes are much, much more efficient than an electric car... if they last long enough).

> (And e-bikes are much, much more efficient than an electric car... if they last long enough).

Would this hold true for e-motorcycles as well, or are we really saying "slow moving vehicles are much more efficient than fast moving vehicles"?

I think in this case it's small vehicles (low mass, low aerodynamic cross section) that are more efficient. Electric cars would only be marginally more efficient (~50% better) at low speeds due to rolling resistance (whose coefficient is basically constant with respect to speed) and parasitic loss from the computer, infotainment, climate control, etc.

Per seat, electric cars are generally much better than electric motorcycles and about half as good as much slower e-bikes (although if ridden at the same speed, almost as good).

It takes less energy to accelerate less mass. So the lighter your vehicle the more fuel efficient it is for the same speed/acceleration.

Going a lower speed is also more fuel efficient (depending on the engine, but it takes less energy) as you have a lower air/road resistance.

In practical terms, while ebikes may have a slow top speed, they move through congested areas much more quickly, so for trip-substitution purposes, the efficiency is apples-to-apples. More direct routes at steady speeds is a real benefit to ebikes.

Much of the efficiency benefit is from the lower drag due to lower speeds, but if the purposed served is comparable, then it's valid. Just depends on what you're looking at.

FWIW, even at 30mph, ebikes are over an order of magnitude more efficient than a Model 3 at 70mph (<28 vs. 280 WH/mile). Zero (electric motorcycle) states 180 WH/mile at 70mph.

Agreed. That fits with my info. Pack a Model 3 with 5 people and go 30mph, and the efficiency per person is about the same as an ebike.

If you go steady speed the lower Cd and higher efficiency of the drivetrain might actually make it more efficient than the bike.

Two things I can think of is gasoline powered motorcycles aren't that much more efficient than a Prius. And a Prius seats four. I suspect the same applies when you switch to electric[1].

The other one is the injuries per mile are much higher for motorcycles. One aught to account for that as well.

Side issue: Motorcycles are terrible from an emissions standpoint. Though of course an e-motorcycle would be better.

[1] Scaling factor vs size is highly positive.

Oh my gosh, wow, what a strange coincidence! I totally did not notice that I was replying to the same person who wrote that post. Small world :)

And yeah, I wasn't necessarily reading too much into the comparison since of course there are so many different factors that can tilt the numbers different ways, but I found it very thought-provoking regardless. Great point about e-bikes too.

This is just stupid. Does this ignore the calories the driver uses up as well?

It's only counting additional calories burned on top of basal metabolic rate. Of course, the car will get you there faster (less time in transit), so basal metabolic calories will ALSO be lower in the car...

Speed greatly depends on your location, I highly doubt the majority of commutes in Berlin (where I'm at) will be significantly (if at all) faster by car than by e-bike - seeing as the ubahn tops at 45 mph & is quite often quicker to get around town than driving.

I suspect the same is similar to any major (and therefor congested) urban area.

The real killer is the American-style sparse suburban living that requires long commutes on highways...

I'd say depending on where you park the bike can be substantially faster (it's actually door-to-door, a car is very rarely).

I think so too, i was trying to be careful and make the less drastic bet that it's at the very least not slower :)

> I've seen compelling arguments that driving a Model 3 actually even has lower average CO2 emissions per mile than riding a bike, since riding a bike requires eating extra calories to power it, and food has a substantial carbon footprint.

Color me skeptical. A Model 3 has ~20x the mass of a rider & bike, give or take. Producing it, delivering it, and moving it around thus takes ~20x the energy in the best case, energy that has to come from a CO2 emitting source in both cases. As bad as meat & dairy might be, being 20x worse than automobile and electricity production combined is a really high and unlikely bar.

This is where I'm tripping up on this idea, too. The 20x difference in mass of a Tesla - but still more efficient than a human pedaling a bike may in fact violate the Laws of Thermodynamics. Also, I'm not sure if this is truly taking into consideration of the resources needed to make a Tesla vs. a bicycle. Or I guess the big one: the # miles/year driven by a usual car owner vs. # miles/year cycled by someone using purely human power to move oneself.

The car can absorb energy when braking, so the minimum expenditure based on the laws of thermodynamics is minuscule and irrelevant.

That’s an extremely tall claim, that minimum expenditure is minuscule and irrelevant, maybe taller than the original unlikely claim that a car might be more efficient than a bike rider. Regenerative braking does not invalidate the laws of thermodynamics, we just have to identify the regeneration efficiency.

So how much of the energy expenditure does regenerative braking in a Model 3 recover? From what I’m reading, it may be in the single digit percentages for highway driving. Tesla’s own web page says the theoretical max is 64% [1], and that’s only in a perfect situation (accelerate & immediately stop), it’s not accounting for driving for a long time without braking or using the electric features inside the car, e.g. aerodynamic loss, rolling loss, & heating & a/c. Wikipedia says including the accessories, typical best-case efficiency is around 50% [2]. But if you accelerate and drive for a mile without braking, you’ve already lost most of your initial energy to air & tires & battery conversion loss, the brakes can only capture energy from what’s left.

[1] https://www.tesla.com/blog/magic-tesla-roadster-regenerative...

[2] https://en.m.wikipedia.org/wiki/Regenerative_brake

You're proving my point. That's a big list of all the ways the car wastes energy that have nothing to do with the thermodynamic minimum to move the mass from point A to B. Those factors absolutely dominate the energy use.

Maybe an analogy is useful? There's a claim that you can cook 20 pounds of food in one type of bonfire setup with less fuel use than if you cooked 1 pound of food in a different type of bonfire. The amount of energy needed to actually perform the chemical changes on a pound of food, even multiplied by 20, is negligible compared to the vast quantities of energy that are going to other things.

Oh, you're right, I didn't understand your point before. You're saying that the thermodynamic minimum is a theoretical limit not approached in practice, right? It's not related to regenerative braking. The connecting "so" made me think you were saying that regenerative braking is the reason the thermodynamic minimum is much smaller.

It's more of "With thermodynamics we care only about the thrust to get up to speed. If we add regeneration then we can even factor most of that out, getting us to energy expenditures so very tiny that they're irrelevant to the total."

I could have phrased it better.

Got it, and likewise I could have not made assumptions and jumped to the conclusion that I understood your comment. I got lost between theory and practice. I think I agree with all of what you said when we're talking about theoretical limits and theoretical near 100% efficient regeneration. It's just that in practice, regenerative braking on a Tesla is not that efficient, and it isn't something that recovers most of your energy spent accelerating, it only recovers a minority share normally, and often only a very small minority.

It's also ignores the fact that simply driving a car contributes to congestion, and the various network effects of that. More traffic means more pollution, more cars means less people are willing to walk or ride bikes due to safety, etc.

Food in general is a massively inefficient way to convert solar energy into muscular exertion. If you compare all the steps that occur from farming all the way to getting the food on your table, and then add in how inefficient our bodies are at converting food energy into mechanical energy, the results of the comparison becomes a lot less surprising. Electrical transmission, batteries, and electric motors are all very efficient by comparison.

How inefficient, what does massively mean? How do we demonstrate electric is 'very' efficient by comparison? Your list of electrical losses notably left out the power generation, which should be solar or wind, but in practice in the US is still a third coal, which does not compare particularly favorably to food. There are a whole lot of inefficiencies from end-to-end in both systems we're comparing here, many of which are really difficult to measure. What we do know for certain is that the amount of work that has to be done with a car is twenty times the amount of work that has to be done with a bike per mile of consumption, and a much much higher factor than 20x for production, which makes the suggestion that the car could be more efficient than the bike quite unlikely to be true, even though food production is inefficient.

Well okay, so I'm getting a bit off-topic with the whole "food is inefficient" argument, but if you look at the efficiency of photosynthesis it only harvests about 3%-6% of the light energy hitting the plant. A lot of that energy then goes to growing other parts of the plant that don't contribute to its caloric content (stems, roots, etc., or in the case of fruit, the entire rest of the plant besides the fruit). To get a ballpark idea of the numbers here, even biofuels have been found to be only about 0.3% efficient at converting sunlight to chemical energy, compared to solar panels which can typically harvest about 10% of the incoming sunlight into electrical energy. Plants are just not that efficient in general.

But efficiency of converting sunlight into energy isn't really what this argument is about, since we're more concerned with the CO2 output than the energy efficiency. The efficiency of human muscles factors in though, as it is only around 20%, compared to electric motors and batteries that are over 90% efficient. Regenerative braking is a big win for the electric car too, since it reduces the impact of the car's higher mass a lot. And when you look at the CO2 per calorie emitted by coal (and of course the US grid emissions are much better than this since we use many other better energy sources than coal), you get about 1.1g per calorie for coal compared to about 4.8g per calorie for fruit. So in a carbon emissions sense, food can also be much less efficient than electricity, even when that electricity comes from fossil fuels.

No single one of these factors is enough to tip the scales on its own, and of course the numbers can skew either way depending on what kind of food you eat, what kind of electricity you use, and many other factors. But I don't think the comparison is a totally crazy one.

Great point that solar efficiency is almost irrelevant to food's efficiency or pollution. The sun is shining anyway, whether or not we use it. Better to do anything with it than not. For that matter, CO2 is a big problem but not the only problem. I have no idea how electric and food compare there.

Regenerative braking is fantastic as an idea, but it does not actually offset the mass in practice. Tesla's regen braking has all kinds of limits - 64% electric conversion efficiency right off the top, but it also caps the regen rate, it doesn't regen when it's too cold, and regen can't recapture a/c or air friction losses, which actually add up to the majority of a car's power consumption for highway driving. I backed that up with data nearby. Regenerative braking is a great example of how cherry-picking efficiency percentages from the components of a system doesn't really help us understand the overall system efficiency.

Anyway, I don't know what the end-to-end system efficiency of food is, and to be honest I'm skeptical of anyone who claims to, the few stats we have are already gross approximations that tend to lose their context and their error bars. Combining them is extra problematic. My really one and only point is that the scales are pre-tipped 20:1 because of the mass on the consumption end, so no matter what happens in between, in order for the scales to balance, the complete end-to-end efficiency of electric system plus electric car manufacturing combined has to be more than twenty times that of food + bike manufacturing. No matter how good electric is, the savings are very much hampered by automobile manufacturing and by power generation. So it's just not that likely that the complete system for e-cars is a factor of 20 more efficient. There are multiple parts of the process that are the same or worse.

You know what's even better? Me. I drive an SUV in San Francisco and I offset all my carbon usage every year. I also eat meat. I also fly. I'm also better than nearly every electric car driver.

I know you're being tongue-in-cheek, but it's worth making the point.

Carbon offset prices are determined by current carbon offset capacity. Like, number of acres where trees can be planted, number of carbon-scrubbing machines that can be installed at a certain location, etc.

Carbon offset supply is limited. Meaning, it cannot scale to the level the planet needs, at current prices. If we gave these carbon-offset companies X billion dollars tomorrow at current prices, it's not like our problems would be solved.

Therefore, carbon offset prices will go up as demand (read: more people seeking to offset carbon) goes up.

One way to battle that is to reduce carbon emissions while buying offsets. That will reduce the required "supply".

People who buy offsets without reducing their emissions, on the other hand, are effectively stealing resources from future people, by buying low and making it more expensive for the people buying later.

(If you really want to be a jerk about it, you can calculate your lifetime carbon expenditures and just offset all of it now while prices are cheap.)

Haha, you’re right, of course. My real goal is to popularize the notion of paying for your carbon debt so that one day when we know we only have a certain carbon budget we’re acclimated to the idea of paying to pollute a little more by buying some of someone else’s budget. Also, I’d rather bootstrap them now rather than later.

I have to buy the offsets now though or no one will listen. Cool idea to buy the entire lifetime’s offsets.

The average American probably already eats enough calories to ride a bike to and from work multiple times a day.

The climate argument is not factually weak... but when it comes to actually persuading lots of people to change their ways... its just so weak. It's just a sad fact about our political reality today.

And even among those who say they are concerned about climate change and the environment - it still can be very difficult to actually inspire some sort of productive urgency in them. Climate change is the most gradual disaster to ever to befall humanity and our psychology isn't built for that at all, en masse.

On the other hand, when you see the air is turning black from the bumper to bumper traffic in your city, and then you notice yourself coughing, and your lungs hurting, at the same time... well that's a different situation.

I think people just need to get good at making the argument, starting from premises. Simple deductive arguments that end up with a conclusion that people need to cut their emissions in half, spend $200/year on carbon offsets, and plant 20 trees / year. That's about right for an average US citizen, and if you show the reasoning then it's the sort of thing people can grasp and in turn share with other people.

What percentage of emissions is usage vs manufacture? I've looked, but I can't get a figure.

The main difference between EVs and standard cars is the battery. Estimates of battery emissions vary widely, but reviews show an average of about 110kg of CO2 per kWh of capacity: https://theicct.org/sites/default/files/publications/EV-life...

(Although this is improving with time and may be as low as 56kgCO2/kWhcapacity).

So a 50kWh Model 3 has about 5.5mT of emissions from its battery. The rest of it may have lower emissions than a conventional car (since there's no engine, exhaust, fuel tank, etc), but pessimistically we can just assume that the difference in manufacturing emissions between EV and standard is just the battery.

Over a 175,000 mile lifetime, a 35mpg gas car will emit 45.4 tonnes of CO2. And a Model 3 (4.1kWh/mile) will emit about 8.5 tonnes with Californian electricity or 17 tonnes with US average grid electricity. In either case, the vast majority of emissions is in the actual energy to fill up the vehicle, not the battery, and electric cars are FAR better (even with a >200 mile range battery which is considered very large in much of the literature) in this respect.

A factor of 3.2 times better in California (including the battery) and still a respectable factor of 2 times better in the US with current electricity mix. But of course, if you buy the EV now and it lasts 10-15 years (including secondary and tertiary owners), the US and Californian electric grid will be much better as coal (and later natural gas) is phased out.

> But of course, if you buy the EV now and it lasts 10-15 years

With companies like Tesla... they are the D-Link of vehicles.

They absolutely don't want long term stable good cars. They want you at the 4-5 year mark to buy a new one. And given the company has harassed engineers poking at under the hood, they will highly likely damage your car via remote disables or other 'trickery'.

I don't trust anything that Musk puts his hands on. There's a good reason for that.

Any source on them not wanting long term stable good cars or is this just speculation, and if so what makes you speculate this? Other than body panel gap problems i'm unaware of any real build quality issues and apparently the motors are designed for a million miles before service.

All the older batteries are holding up very well, so the admittedly limited data we have directly contradicts that. I'd like to see a better right-to-repair stance taken by Tesla, but their products have thus-far proven durable.

You can dislike Musk all you want, but it doesn't change reality.

If anything it’s the other way. The software updates provide a huge boost to the existing fleet, features that you would normally have to trade-in and get a new car to enjoy.

In some cases you can even upgrade the hardware (e.g. the AP and MCU) with the latest shipping parts to keep your model fresh. This is not universally true - there is some drama over a promised Model S MCU upgrade which a lot of people want only apparently only a handful of owners have gotten, but overall the approach is very refreshing.

In my opinion the carbon cost to manufacture is mostly relevant when you get into arguments about overall lifespan— like, can an EV get a battery change every 100k km and otherwise drive until the wheels fall off? If so, then it amortizes that initial cost to a much lower point than a gasoline car which is usually toast around the 200-300k mark.

When comparing the per-distance numbers, though, the initial manufacture can be treated as a sunk cost; you would have paid it regardless of whether you bought an EV or conventional car.

For the most part, long range (>200 mile) EVs have batteries that will last at least ~200,000 miles before range is reduced significantly. Figure about 1000 cycles. It's possible to do even better if you're careful with charging style (and larger batteries in the same vehicle tend to do better as their discharge and charge rates are proportionally less).

I've been working off the belief that the limiting factor for a gasoline car is the engine which only lasts about 4-8000 hours before it's kaput. Electric motors and simple fix gear drivetrains should last much longer.

EV's might well double the age of the average car on the road.

I think they will, especially once the battery gets slightly larger and chemistry gets further refined. 1000 equivalent cycles times 300 miles of range per cycle is 300,000 miles. Takes about 20 years to get there, so for commuter vehicles, will almost certainly want to replace by then.

For places in the US that use salt on the roads (upper Midwest / northeast), vehicles suffer from corrosion before drivetrain issues in my experience.

I would never buy a used car that has had lived through winters in places where roads are frequently salted.

The way I estimate this is just assume about 2 tons of CO2 per ton of steel[1]. Based on that it's hard to see a normal car requiring more than 5-6 tons of CO2 to produce.

[1] Chemical reduction dominates everything. The energy required for forming and assembly would be much less.

Some metals can undergo a bunch of reheats in processing (3D printing of metals is particularly bad) which could approach chemical reduction energy, but you're right to zeroth order (which is good enough in these conversations, IMHO).

Good rule of thumb.

One other point is most automobiles are now recycled. Going what one reads on the internet about 25% of the steel in cars comes from recycled steel. And 95% of cars are fed into the recycling stream when they are scraped.

> The climate argument is not "weak for electric cars." Electric cars are about a factor of 2.5 better in terms of emissions per mile even with the current US electric grid.

Isn't a large benefit also that the emissions generated by electric cars is centralised at the places producing the electricity, which makes it easier to clean up the pollution?

This is all true, though much of the impact of cats is in how they affect urban design patterns and result in cities where it's impractical and dangerous to walk, cycle, etc. , and huge amounts of land are lost to asphalt.

I like EV's but they still encourage auto dependence and its associated ills.

Now I'm considering whether cities would be nicer if we would let cats influence their design... probably yes (more nature?), but there might be other problems.

It's something to explore. More sunny places to nap for everyone! But the toll on bird life would be substantial.

Im confused by the numbers cited. Gas only weighs 6 lbs per gallon. Assume a car gets 20 miles per gallon. How does a car produce 20 lbs of CO2 per mile?

Not 20 lbs of CO2 per mile. Most of the mass of gasoline is carbon. But most of the resultant CO2 mass is oxygen. So each gallon (6 lbs) of gasoline produces about 20 pounds of CO2. So in your example of a 20mpg car, each mile is a pound of CO2.

This just reinforces how ridiculously unfamiliar I am with the actual working details of internal combustion engines - quite interesting.

C weighs 12

O weighs 16

And there's two Os in a molecule of CO2, and gas is mostly a bunch of Cs. The Os come from the air.

(and it's 20 pounds per gallon, not mile)

Combustion is a reaction between fuel and oxygen. Part of the weight comes from the fuel, the other part comes from oxygen in the atmosphere.

Its 23 pounds of CO2 per 7 pound gallon of gasoline. Simple chemistry. A (CH2)x unit in gasoline combusts to CO2. Thats an increase from 14 atomic units to 44 atomic units by replacing H by O.

Since the average internal combustion vehicle uses 500 gallons a year, that is about 6 tons of CO2.

This is a great question. I had exactly the same thought some time ago and was scratching my head where the extra mass was coming from.

Short answer, it pulls oxygen from the air to make the CO2. Most of the mass of CO2 is from the O2 that came from the air, not the tank.

A gallon of gasoline by itself only weighs ~7 pounds, but the reaction with the atmosphere’s oxygen results in ~20 lbs output.

Reply on news.ycombinator.com