Hacker News Reader

Comment by cheschire | Hacker News Reader

If you’re going to factor in the refining process for the fuel, wouldn’t it make sense to factor in the refining and manufacturing process for the battery?

View on HN

Yes, it would, but that’s part of making the overall car and it depends on lifetime. As EV batteries can now last 500,000 miles (vs 250,000 miles for first generation long range EVs), that’s not as big of an impact as many assume. (A lot of papers assume really low EV lifetimes, like just 100,000 miles, and no secondary reuse of the battery and they assume really outdated and inefficient numbers for manufacturing the cells.)

Only tangentially related, but I wonder what the expected battery lifetime drops to when you factor in crashes. You mention that the cells might last 500,000 miles but is that really relevant if you're likely to have your car totaled before 200,000?

Are there diminishing returns if you are a company working on a million mile battery since virtually no vehicles will make it that far?

I am speaking as someone who has had three cars totaled before 100,000 miles (none of which were my fault!)

I imagine most of the longevity research is actually being done for the stationary battery market.

The fact it carries over to vehicles is just a bonus.

Besides the salvage value of a battery pack would be huge, so I suspect a battery would be about the last piece of a car to end up on the scrap heap.

The point is that the worst quality studies do assume that the battery ends up on the scrap heap, way before the car itself does.

Totaled EVs are a great source of second-hand battery modules, used either for fixing broken batteries or even for grid storage.

Lots of commercial vehicles make it to a million miles. Those are the ones that will want the longest lived batteries.

> You mention that the cells might last 500,000 miles but is that really relevant if you're likely to have your car totaled

The market for batteries pulled from wrecks is hot! For my own personal interest I wish it wasn't so I could batteries for cheap.

Why are batteries typically measured in miles, rather than years and or recharges? Age and charge cycles are the two hardest things on batteries, not distance traveled. I for example will not drive 500k in my life, but I'm sure the battery is going to have much less capacity and life left in 10 to 20 years regardless.

It depends on what you are talking about. If you are talking about a battery car, then it is better to use a number that makes sense for the end user, and number of charge cycle is not an easy one. If you recharge from 70% to 90% every day, is that a cycle, is that 1/5th of a cycle?

The main problem is the number of full capacity cycling a battery cell can do can change widely based in how it is used. So it depends on how it is integrated into a car and how the car is being used. A typical Li-Ion cell will last a few hundred cycles if cycled completely at a high rate without proper thermal control. But this number can easily be multiplied by 10 with proper thermal management, not charging it to max voltage and doing shallower cycling. If you want to give a number for a final product, you are bound to make assumptions on how it will be used. And at this point you just have an equivalence between distance driven and cycling count.

Also, age has less impact than cycling count in the context of a properly designed battery pack. The two main factors for battery degradation are time spent at high temperature and high state of charge, and cycling count (with speed and depth being the key elements). EV manufacturers are keeping a buffer for that purpose and their batteries do not spend a lot of time at high state of charge where the chemical degradation is more effective.

That would be comparing two completely unrelated things.

If you want to factor in manufacturing emissions, you would have to do so for both. Unlike the manufacturing process for the battery, fuel refining is not a one time thing, but part of the emissions associated with operating the vehicle.

What about emissions required for generating electricity?

That’s already factored in.

I think the only thing in question is the co2 cost associated with the creation and consumption of the fuel, not the storage of the fuel or the parts that consume it.

But if https://climate.mit.edu/ask-mit/how-much-co2-emitted-manufac... is a good benchmark, we can probably assume the battery's manufacturing process releases between 3120kg and 15,680kg of co2, though that does describe a tesla model 3 battery and I don't know how similar it would be. For the sake of expedient math, I'll average them and say it emits 9,000kg of co2.

If we assume a lifespan of 200k miles, no part replacements, and no change in co2 costs/mile for EVs, then the total emissions would be:

~68,200kg for the EV (341g/mile) + the battery = 77,200kg total

~177,800kg for the non-EV (889g/mile), not including the manufacturing emissions associated with the ICE

Ultimately it's not an insignificant amount of co2, but in context it is actually pretty unimportant. It turns 341g/mile into 386g/mile. Of course, it's incorrect to assume an EV will have static "emissions" since they all come from the production of electricity, and given current trends it would be fair to assume those numbers will trend downwards. Gasoline, however, can probably be expected to have fairly static emissions over the life of the vehicle, likely actually getting worse as parts wear.

> If we assume a lifespan of 200k miles, no part replacements, and no change in co2 costs/mile for EVs

I am under the impression that battery trains do not last anything like that long - am I misinformed?

Doesn't change the maths much, but curious

Most EV manufacturers are offering around 8 year/160,000-200,000km warranties on the batteries.

So not 200,000miles. But definitely covering the average usage pattern.

And not all will last that long, but the majority will. And potentially the worst case is just degradation - I.e. 200km range instead of 400km range, so still quite useful for a variety of use cases even outside of cars.

Yes, current EV batteries (not the old Nissan Leaf ones without active cooling) are projected to last as long as 3 million kilometres: https://youtu.be/7DYknTVskTw?t=325

https://www.greencarcongress.com/2019/09/20190927-dahn.html Dahn’s “million-mile battery” detailed in open-access paper in JES

Warranty is to some percent of original capacity, like 80 or 90%. Then you have another few hundred thousand miles for the next quantum of degradation.

Not really because this about emissions per mile driven. It takes energy to refine and deliver energy (both electricity and gasoline) so it makes sense to take that into account when comparing total emissions (comparing local emissions just makes ICE vehicles look bad). If you included the battery pack you would need to include the drivetrain of the ICE vehicle to make it commensurate. At that point it is just a different analysis, a fine analysis to do of course, but a different one.

Reply on news.ycombinator.com