Hacker News Reader

Yeah, it was a pretty silly and annoying opening to the article. Power is the only reason there's any difference in charge time for any of these devices. The reason their laptop is taking so long to charge is because they are likely using a phone charger on it. The reason an EV can't change in an hour is because nobody installs L3 chargers on their homes.

The only reason it's related at all to battery size is because larger batteries need higher power levels in order to charge in an hour.

I'm hand-waving a bit over chemistry because you are entirely right about the C rating.

> The reason an EV can't change in an hour is because nobody installs L3 chargers on their homes.

The reason my EV can't charge in an hour is because it won't accept more than around 60KW, which charges it in about 3 hours.

You have a 180KWh battery that only takes 60KW? What are you driving? That's a weird configuration.

Most of that time could well be the last ~20%

My car takes 240KW (It still feels insane to be delivering that much power) and 10-80% (56KWh) is about 15 minutes and 80-100% is another 15-20 minutes

Mine is the same.

That's why I find a car that only accepts 60kW and charges for 3 hours to be odd. Even if I limited my car to only 60kW, it'd take about 1 hour and 4 minutes to get 80% and another 20 minutes for the remaining 20%. So around 1 1/2 hours

(edit: fixing numbers)

Mine takes around an hour and a half from zero to 80% and just over an hour the rest of the way. In general, I find it rather unpleasant when the last 20% takes almost as long as the preceding 80%. I usually only charge to just above 90% though because going above disables features.

I usually charge from less than 10% to near 90%. Lower than 10 and the motors reduce output, higher than 93 and regenerative braking shuts off

2025 Subaru Solterra. It's essentially the same as the 2022 model, and I juust managed to miss the new 2026 model that is better at everything except physical controls

Oh wow. I didn't know Subaru made EVs... and I can see why. That's horrible.

It's really strange how much these older manufacturers struggle to make a good EV.

This was one of Toyota's first BEVs (Subaru's EVs are all based on Toyota). It keeps trying to pretend to be an ICE vehicle and that pisses me off. (forward creep, no one-pedal driving, accelerator constantly modifies or ignores my input...) If it weren't a lease, I would pull it apart and hack the control units.

Rule of thumb: ANY li-ion cell (or battery of cells) can charge from ~20 to ~70% in under 20 minutes. If not, the charging is not done correctly (either you're limited by the charger or the cooling of the battery)

> If not, the charging is not done correctly

For some value of "correct" but not others. For EV's, most of the time, the "correct" thing to do is to plug it in overnight, and let the software do the rest to optimise charging. Which might look something like:

Charging to start at midnight (or whenever the cheap rate starts) and complete by 7am (or usual wakeup time) and minimise wear by charging below a fastest rate, and take the battery level from where it is now (e.g. 50%) to the configured stopping point (e.g. 80%)

This is clearly different from "charge as fast as possible and be done in 20 minutes". Although that is occasionally the requirement. But not most of the time.

I miswrote. I meant that it's not the battery being the bottleneck, but rather something else - the charger, the cooling, the grid, the user/app setting a non-maximum power etc.

Agreed, though the other point is that the larger charging system (battery, charger, grid etc) being the bottleneck in need of optimisation is not typical. It happens, but not every time. For overnight charging, you want L2 and usually the rest doesn't matter so much.

This is true. Additionally, configuring them in series, they can all charge at the same time, it just requires a higher voltage charger.

> If you configure them in parallel, they can all charge at the same time. Increasing the number of cells doesn't increase the time to charge them.

Correct. In principle if you have enough current available you can charge several thousand 18650 cells in the same time it takes to charge one. But you also need low-loss switching to reconfigure all the batteries to be parallel and monitor them all for rate of charge, voltage, and temperature. The electronics needed are buildable but not trivial.

There is one other consideration. Sure we can charge some batteries at C rates of 10 - but should we. Often the cell will take it, but you damage the cell and so you lower your lifespan. If you are on a road trip once in a while this isn't a big deal, but if you travel constantly (that is a delivery driver) this becomes an important economics question since the battery will wear out faster.

> but if you travel constantly (that is a delivery driver) this becomes an important economics question since the battery will wear out faster.

Delivery drivers are single shift and usually only drive very short distances, which means these vehicles mostly see a slow trickle charge over night.

That is one type of delivery. I know someone who specializes in cross US deliveries. There are some niches where only the original counts and they don't trust it to a standard delivery, so it gets driven in a car. truck drivers also count as a delivery. I used to have a friend who installed paint booth to automotive body shops - a special skill that had him drive all over the US to wherever the next job was