Hacker News new | past | comments | ask | show | jobs | submit login
Tesla is shifting from nickel to iron cathode on standard range car's batteries (twitter.com/elonmusk)
102 points by giuliomagnifico 44 days ago | hide | past | favorite | 141 comments

Tesla is shifting nothing. Tesla is consuming all batteries that are available and, after starting operations in China they gained access to CATL who heavily invested in iron phosphate.

Thanks to high efficacy of Tesla drive train, base models can have decent range, around 250miles EPA, with iron phosphate battery packs.

However, iron phosphate battery volumetric density, at this time, does not allow for cars with what I would consider minimum for a single car family, in locations with real winters: 330miles EPA.

> does not allow for cars with what I would consider minimum for a single car family, in locations with real winters: 330miles EPA.

* in the USA

Our EV (our only car) has 100-150km real-world range (60-90miles). We have real winters. We can get everywhere we want to for daily driving. My grandparents has a cabin 2 hours away that we visit regularly in winter. We can get there one one fast-charge that takes less time than we use to buy groceries for the trip (fast-charger we use is next to a super-market).

I've borrowed my moms ICE car and a friends Tesla Model S for a road-trip across the country a couple of times though.

But Northern Europe does not have the same problem with suburban sprawl and lack of public transportation, people don't usually drive as far regularly, speed limits are lower (higher efficiency) and now there's fast-chargers everywhere.

I think this is a fundamental difference between European and North American perspectives: everything is so much closer in Europe, and so much farther away in NA.

Well, the saying is that Europeans think 100 miles is a long way and Americans think 100 years is a long time.

That’s certainly been my experience when I’ve visited the USA (I’m British and live in Berlin, have visited central and north CA, Nevada, Salt Lake City, NYC/Newark, Rhode Island and the Cambridge/Boston area).

For a point of comparison, in the state I live in in the US, you’re looking at 2-4 hours to a lot of vacation destinations. Many of which don’t have fast chargers along the way. In a normal pre-Covid year, I’ve driven upwards of 8-15 hours one way to go places, multiple times a year.

Public transportation is available but far less convenient, and charging stations would add possibly hours on trips that length.

To be fair, Lithium Iron Phosphate batteries (LiFePO4) are really good technology. Incredibly safe - they don't catch fire in the same way that standard Lithium Ion batteries do - they're better for the environment, have crazy discharge rates, etc. The only thing they're worse at is energy density.

They don't perform badly in winters, I characterized the cold-weather relative performance of LiFePO4 vs LiIon batteries for an aerospace engineering project I did in college.

Frankly I thought those characteristics would outweigh density in automotive applications, but I was mistaken.

I believe they are significantly better for cycle lifetime as well.

Difference of opinion, I think 200 miles is plenty of range even for most 1 car families.

That is basically 1 stop from LA to vegas, and 1-2 stops from LA to San Francisco.

I'd like to see focus on lowering the base cost of Teslas, and the lower the range target the easier to get there.

200 miles is not 200 miles in real winter though.

It's also not 200 miles anyway.

You can't use 100% of your battery.

Most people preserve their battery health, so will not charge to 100% or drain to 0%

You can't (practically) supercharge to 100%. It will probably take as long to get from 80 to 100% as it does to get to 80%. In many cases it's not nice to hog the charger.

And its really hard to get to 0% even if you tried. There probably won't be a supercharger located right where you hit 0% even if you risked it.

So.. you really use the middle 60% of your battery, between 20% and 80%

> Most people preserve their battery health

No, most people don't pay any attention to it, which is why all the Battery Management Systems in all EVs pay attention for you.

On a tesla you set it in the UI and it is an integral part of charging. There's a charging "set limit" percentage from 60 to 100%. 60% (daily) up to 100% (marked trip) I think most people choose 70 or 80%

On busy superchargers the limit is moved down to 80% if it is set above it.

Now other cars treat their battery differently. They may indicate 100% but they reserve a buffer and it's really some lower percentage. I suspect tesla has a buffer too, even if it is just for temperature fluctuations, which seem to move the percentage around.

200 miles is not even 200 miles in summer when doing a road trip. Highway miles are less efficient than the average of the test cycle, and much less efficient if you drive on the faster side.

It's now closer than you might think (except in extreme conditions).

The main driver of range loss for earlier tesla's was the fact that they used an resistivity heater for the cabin/battery. That changed fairly recently to a heat pump. The range loss now is far less than it used to be.

Extreme temps will render the heat pumps useless, but that's not the common case. You've got to be in northern Candida and Alaska before you are starting to talk about a useless heatpump.

No extreme conditions are needed. When taking relatively short trips in winter you may have to heat up the whole interior and turn on the seat heating every time. And until the battery is warm enough you also sacrifice some of the quick charging, regeneration, and performance. The BEV owners in my family all reported 20-40+% lower range whenever they can't preheat everything while still plugged in, and that's at relatively decent temperatures (-6/-10C). Of course this can be partially mitigated (although freezing in a 60k+E car doesn't seem like a decent option to me) and might happen only for several weeks every year but even a penalty that's smaller than what the average ICE will see in winter is still a lot given the typical shorter range and longer "refill" times for EVs.

How short is short? Throughout the winter I'll do 20 or 30 mile trips around town with no problem. I don't even think about range in those cases. It's not until I'm doing 200 or 300 mile trips that I have to start prepping for things.

I think the recent experience in Texas shows that you have to plan for the worst case even if you don't think it will be necessary. Anecdotally, I know the manager of a car dealership here in Utah (we get real winters, at least sometimes and in parts of the state) who says he gets a lot of Tesla owners trading in their cars, and he implied it had to do with range issues.

Texas's winter was extreme for Texas, not extreme for what a heat pump will do.

I live in Idaho and get "real" winters. Range hasn't been a problem for me with my 2018 Tesla (Even though it has the resistivity heater). I've even visited Utah a couple of times in the winter.

And, no offense, but I don't really trust the manager of a car dealership to give me accurate information about what's going on with tesla trade ins. You wouldn't ask a Chevy dealer about a Ford.

I'm itching to heatpump across Northern Candida

If you frequently go on long road-trips, 200 miles is not sufficient in summer or winter.

If you don't, 200 miles is more than sufficient, regardless of the season. If you absolutely have to make a 190 mile trip in winter, consider stopping for a charge, or renting/borrowing an ICE car.

When the battery is also powering your car’s heater the range drops significantly. Especially with short trips that need to repeatedly warm up from cooler winter temperatures.

why not use a heat pump for heating given its >100% relative efficiency?

Edit: apparently some VW models include a heat pump for this reason.[1] Edit 2: Apparently some Teslas have it too. [2]

[1] https://www.speakev.com/threads/heat-pump-yes-or- no.16768/

[2] https://electrek.co/2020/03/13/tesla-model-y-has-a-heat-pump...

In my opinion, proper winter-ready BEV should have a heat pump for heating the interior and battery, battery insulation to reduce uncontrollable loss of heat from battery, rotary heat exchanger to further save on heating ventilation air and double-pane windows to reduce losses from interior.

Double-pane windows would increase weight by about 5%, while decreasing losses through windows about 5x, iirc.

Where are you pumping the heat from?

yea but many places in winter heat pumps efficiency goes to basically a electric heater level and if works at all.

IIRC, cold climate specific (Mitsubishi Zubadan series for example) have COP of about ~2 at -20C. Even cheaper cold climate pumps do function at -30, albeit with COP barely above 1. So resistive heater should also be present, but it won't be used often.

Not true. Inverter heat pumps can do 8 or more btu per watt at sub zero temps.

outside, just like residential heat pumps.

I think in a real winter the battery would need to be heated even when the car is parked overnight, no? Or store it in a well-heated garage overnight.

Nope, I walked outside in -10 degrees centigrade and had to choose between my diesel Sharan and my Renault Zoe. Easy choice: the Zoe had instant heat because of its heat pump. It worked fine even with with battery at -10, and the displayed range slowly rose, as the battery warmed up. Once the interior was comfortable, the BMS asked the heater to send some heat into the batteries. In the Zoe's air cooled batteries, the heating/cooling system is shared between the occupants and the BMS.

Stop. Just stop.

First off, brand ranges are rated at 100% capacity and many brands suggest to not charge to 100% most of the time if ever. Then throw in weather effects on range and this does not just mean winter but driving in rain can increase usage.

However the top two reasons to push 300 more as the baseline is simply convenience. The convenience of not having to worry about your charge or need to charge frequently. That removes the negative comparisons many try to make showing that BEVs are not ready for prime time. Once technology catches up and you can fix that range in ten minutes; not going to happen; then a lower range might be okay.

Second and more importantly the big automakers want you to buy into that idea so they can eek out small range improvements as a selling feature to get you into next year's car. Similar to how many of them still have heart failure of OTA in the definition Tesla uses it.

Now can there be a type of car where ranges are aimed at commuting only? Sure, but as a whole replacement vehicles for ICE vehicles should out perform on all numbers.

Don't let them sell you into an infinite cycle of range increases because they will. 250 is entry level now and 300 should be the minimum for any semi luxury and higher priced EV.

Hell look at the recently released Mini if you want an idea of where the industry hoped to be at but Tesla shot that idea full of holes and doubly so when the Bolt came along and now VW ID.* series.

For what it's worth, this is slowly changing. BattleBorn has certified their LifePo4 batteries to lower temperatures. Been a while since I watched this. [1] Just slower charge rates, but no longer unsafe due to charge controller changes. Previous versions would shut off the charge below 0C. So charging slowly over night may still be useful, but speed charging in cold weather would not.

[1] - https://www.youtube.com/watch?v=ywn-vBjKblI

-20C is not enough

Yeah, they are compensating with heating elements. This will of course use some of the capacity. Telsa is doing heating elements as well. Heating during overnight charging should be easier if you are on grid power. Charging batteries when they are cold can harm them. Discharging them when they are cold equates to more internal resistance, so you could lose up to 15% range without heating elements. No idea how much power the heating elements draw. I would bet that Tesla's battery acquisition Maxwell [1] could make some improvements in this area.

[1] - https://techcrunch.com/2019/02/04/teslas-maxwell-acquisition...

It’s not as if internal combustion engines work great in -20C temperatures either. Often they won’t even start, and diesels typically have an (plug-in electric!) engine block heater in such climates. So I just don’t see this as a problem. Also, what proportion of the world lives in climates where it makes sense to optimize for -20C weather? Northern Europe, parts of North America... and that’s pretty much it. So while Norway (which now buys mostly electric cars... and famously a LOT of Teslas) might use cold-optimized chemistries, it’s not a major consideration for electrifying the world.

Speaking as someone who's car is currently plugged in, and who just went through a couple weeks of below -40c mornings, this is bunk. Modern IC engines start reliably at -20 without preheating. Mine started fine this morning at -22c even though I had not plugged it in. It isn't a great idea for the engine/battery in the long term, I am kicking myself a little for not plugging it in, but I had no doubt it would start (honda civic). I've even cold started it at -36 in an emergency. I generally wait for my brake/clutch/PS fluid to warm up but they have nothing to do with starting.

In a modern IC engine, the real issue isn't the block heat but the battery. A bad/old/tired battery won't provide the amps when cold. For all practical purposes, anyone worried about cold starts would be better served by a battery heater/blanket than a block heater.

For gasoline vehicles, you’re correct (although it certainly helps to have a block heater). Block heaters are essential for cold weather diesels. But all combustion engine vehicles need a way to start, and they rely either on muscle power (eg pull-start lawnmower) or a battery (okay, I have hill-started my car quite a few times...). In practicality it means internal combustion engines rely on batteries just as much for starting. And the chemistry they used (lead acid) wasn’t terribly good in cold weather, either. Nowadays, they sell nice compact lithium chemistry battery jumpstart packs the size of a paperback novel which will start your car. Regardless, batteries are still the solution.

Just remember that modern car electronics stress the hell out of a car battery during car use and they don't last as long as they used to.

If your engine is running, your battery won't even feel those electronics. Your alternator/rectifier is pumping 13.9v so long as the engine is turning. Your 12v battery definitely should not be drained by anything unless at a very low idle/off.

>It’s not as if internal combustion engines work great in -20C temperatures either.

Pretty much every ICE made in the last 40yr, save your $110 lawnmower that was made without a choke to save $0.38, is fine at -10 to -20f so long as it has some semblance of compression. If anything the biggest problem is that batteries don't work so well in the cold so they'll have a really tough time starting an old engine with poor compression (which will be even harder to start in the cold).

It hits -20 all the time where I live and many/most people in the neighbourhood park outside. Modern cars with healthy batteries absolutely will start.

So will an electric vehicle. Of course, it’s recommended to let your internal combustion engine run a bit to warm up before highly stressing it to prevent damage.

LiFePO4 batteries simply want to be warmed up before accepting a significant charge rate to prevent damage. That’s comparable.

> Of course, it’s recommended to let your internal combustion engine run a bit to warm up before highly stressing it to prevent damage.

I don't have a reference handy, but I read a decade or so ago that the best way to warm up and engine is actually just to drive in a reasonable fashion.

Yup, that would be fine. Driving carefully would be acceptable. Just don’t gun it. (I have a lead foot, so I’ve had to watch myself...)

Not sure what it’s being downvoted if you street park a car at -20 it ain’t gonna start if it doesn’t have engine block heating.

If your car was sold in a region that gets that cold it was likely pre-installed if not you had to take care of that.

The block heater heats up the oil which heats up the engine block it’s usually electric that can run off the battery or an external outlet.

If you buy a car which was intended for warm or moderate climates and drive it in Norway or Minnesota during peak winter you won’t be starting it in the morning.

This isn’t true. Very few vehicles have engine block heaters and it’s really more of a diesel thing because the fuel gels.

My car happily starts at -40, if a tad reluctantly at times.

If your car isn’t old or diesel, it will probably start at any outside temperature.

About the only difference is the engine coolant needs to be rated to expected low temperatures.

Eh? In Northern Europe new petrol cars are sold with block heaters, the recommendation is to plug in the car when if gets to -10-15c.

Not only does it ensures your car can start but it also prevents damage to the engine.

This is how outside parking in Northern Europe during the winter looks like: https://www.bigstockphoto.com/image-78137918/stock-photo-blo...

Growing up in central Alberta (with routine -25C daytime temps) we always plugged in our cars, but now almost nobody does. There's no need anymore. Modern cars start fine.

Even my diesel VW Jetta had no problem.

It is a good idea, but there are lots of good ideas when it comes to cars. Do you ever start your car when parked on an incline? That can play havoc with oil delivery but we all do it. Park a car in the cold with a not-full gas tank? That isn't recommended either.

Cold-starting engines 30 years ago was a big problem. They had lots of different metals that would flex against each other. Modern engines are built to tighter tolerances and that means metals that expand/contract at more similar rates. And our oils do not thicken as easily thanks to improved chemistry. A cold start really isn't going to destroy your engine. It will still probably outlive the rest of your car.

30yr ago was 1991. Pretty much anything with fuel injection (i.e. basically everything in 1991) will start just fine at low temperatures.

The one OEM who's processes I have knowledge of with was doing their testing down to -40F in the late 1980s (which coincidentally is the temperature at which their electronics system starts throwing codes for misbehaving temperature sensors).

Partially agree, but...

- Some of the manufacturers at times have moved back towards Iron-Block/Aluminum-head designs (rather than all aluminum.) Thankfully yes modern gasket technology and metallurgy has improved but there is still extra wear and tear as a result.

- Depending on the way the ECU is set up, there may be other advantages to pre-warming the block. As an example, my WRX sometimes gets -really- cranky starting in cold temps. But it's not the start I'm worried about.

    - Below ~25F, the fluid in the clutch gets to the point where I can sometimes pull my leg off the clutch and wait at least a half-second before the pedal 'thunks' up. An engine block heater would probably help with that a bit.

    - The way the ECU is programmed, there is a -long- delay from when the car switches from open loop (just working off a MAP or MAF sensor) to closed loop (looking at the O2 sensor). In cold enough weather, If the temp gauge doesn't reach a certain point before I hit the highway, it -never- hits closed loop and my mileage is absolute trash. Not every car is set up like this, but more than you think. It's an emissions thing.

Most of your clutch's working fluid is nowhere near any block heater. It is in your clutch master cylinder right above your clutch pedal. It is essentially inside the cabin with you rather than under the hood with the engine. You might have a fluid reservoir under the hood but, warm or cold, that fluid isn't being used unless you have a leak.

As for running temperatures, the standard trick is to reduce airflow across the radiator. Trucks put on "bras" or you can rig up something made of carboard in front of the radiator. Always cardboard because it is soft enough not to damage anything if it gets loose.

> This is how outside parking in Northern Europe during the winter looks like: https://www.bigstockphoto.com/image-78137918/stock-photo-blo...

Only the fancy ones, most people don’t have parking spots like these.

The building I used to live in Finland had outlets for maybe 20% of the parking spaces. Mostly people wanted the pole if they wanted to run an interior heater so their windows wouldn’t be frozen in the morning.

No such thing in the coldest metropolitan area in the US (Minneapolis/St. Paul, at least there are none which are larger and colder). Avg daily high/low of -5C/-13C in January over the last ten years. City streets are lined with cars, and only very rarely will you see an extension cord pulled to a street parked vehicle.

Diesel at least was much more popular in Europe which would explain the prevalence of block heaters and the popularity of their use in gasoline/petrol engines, but that doesn't mean they are necessary.

Growing up we had a block heater on our large diesel farm truck and on diesel tractors, I have never in my life used a block heater on a non-diesel vehicle having spent the first >25 years living in climates which regularly hit -30C.

These are also quite common in Canada.

Diesel is common in Europe but so is petrol.

Plugs for block heaters are very common in Northern Europe, Finland is turning them into charging spots now https://insideevs.com/news/332283/finland-has-a-genius-charg...

The heater also does plenty of other things including defrosting windows.

This is why garages (including heated garages) are super common in Minnesota. It’s not just an extravagance. And, conveniently, garages tend to have outlets in them, very often better than just 120V 15A, too!

It'll start fine if you fill it with 0W or 5W oil and didn't compromise on the battery. Even -30C should be fine.

Some people fill their car with 10W oil or don't buy a battery with adequate CCA, but that's on them.

Also, what proportion of the world lives in climates where it makes sense to optimize for -20C weather?

Dismissing problems or desires that don't affect people you know is pretty shortsighted.

Also, skiing.

It’s dissmissible if the comparable solution (internal combustion engines) already have to use the exact same intervention (engine block heater). I grew up in Minnesota (and visit regularly). If anything it affects me MORE than most people, and I’ve been stranded with a non-starting internal combustion engine car in the cold winter many times which is how I know the existing internal combustion engines already don’t do well. It is not uncommon for parking lots (especially in the Northern parts) to be equipped with outlets for this precise reason (and people often leave their engines running while they go into the store on particularly cold nights so they won’t be stranded). Makes electrification there actually easier in some ways as there’s already widespread infrastructure for trickle charging.

Concerns which are not that well-informed are, indeed, dismissible.

If you drive an ICE car bought in say California to where it gets to -20c and park it outside it won’t start either since most likely the dealership didn’t include a block heater in your package.

not true, i don't see anyone on my street plugging in and we get -30 or lower some days, though probably not good for the engine but yes they will start no problem.

Sure, maybe if the car is 40 years old. But every modern gas car (and all the diesels I've owned) will handle it just fine.

Diesels need plug-in electric engine block heaters at anything like those temperatures as well. This would be the same thing.

No they do not. They need winterized diesel fuel (kerosene added), but that's it. -30C was a "nothing special" for BMW N47 engine to start. -37C is a pour point of 0w40 engine oil, but higher temps are fine in terms of cold engine startups

Here’s a map of the minimum yearly temperatures in the US: https://www.reddit.com/r/MapPorn/comments/73h7eo/winter_low_...

(note C vs F on the legend)

As someone who lives in California, and likes snow sports, it looks like I could run into -20C conditions in the Sierras, and certainly the Rockies (although driving that far with an electric car would probably be too annoying to attempt).

It’s probably not a concern for the average Bay Area family.

I use a rough metric of 0°F as the lowest low to generally worry about (Fahrenheit is nice that the 0-100°F range is pretty close to the typical ambient temperature range in temperate zones). Despite living around the 5b/6a zone boundary (that's -10°F on the map), I have yet to see my car's temperature gauge report 0°F or below, so the conditions are a bit lower than I'd gauge as "lowest typical low."

But that said, zones 6a/6b are definitely in the region of "yeah, I'd worry about 0°F overnight", and people in 7a will definitely worry about temperatures around 10°F if they're going to start their car. 6a-7a covers most of the population corridor in the NE US, about ⅓ the US population. Additionally, much of the Midwest is in the 0°F-is-possible territory, from St. Louis and Chicago straight through to Detroit and Pittsburgh.

The general point is that starting in -20°C isn't some "well, sucks to live in Canada/Minnesota/North Dakota" concern, but rather "gee, a significant chunk of the US population has to do this on an annual basis."

Two weeks ago I had -32C in MN and it didn’t set any records.

Also my family back there told me it didn't get above -20C for like a week or two in a row.

Not a lot of people live in climates where it goes under -20C. Folks in Eastern Canada seem to be doing fine with 250 mile EVs as well.

What are you talking about? It gets under -20C routinely each winter for a couple nights every year through the whole of the northeast; upstate New York, Vermont, Maine, New Hampshire, Ontario, Quebec.

Toronto is the 3rd/4th most populous city in North America (depending on how you count it). Almost every winter there's a -25C overnight. Not for days on end like where I grew up in Alberta, and not every year, but it definitely happens. There's 7 million people in the Greater Toronto / Hamilton corridor alone. 2 million people in the Montreal area. Almost a million in Quebec City, 1.4 millionish people in the Ottawa/Gatineau area. Not to mention Detroit, Windsor, Buffalo, Albany, Syracuse, Burlington, etc. etc. etc. Oh, and I'm seeing a mean minimum of -19C in Chicago for January, too, so throw in a few more million people there because that's close enough.

"Not a lot indeed." Only maybe a couple dozen million people.

EDIT: Oh yeah, I hear the US has a "midwest", too. I hear there's people there, too.

Minimum single families drive 200 miles per day? Is 'minimum single family' a family of 12 average commuters?

Single-car family. As in, this has to serve the transportation needs of a family who relies on it as their only vehicle. That entails far more than their daily commute, and has to support a reasonable worst case, which is what his number represents. A car that only supports your average case is not very useful. And range is diminished in winter.

In other words, if I’m living in Dallas and I have to choose 1 car to meet all of my use cases, I’m not going to pick one that precludes me from ever being able to take the wife and kids to visit relatives in Houston, even if I only make that trip occasionally.

Tesla has superchargers exactly for situations like this. Lots of people do this trip (or longer ones) all the time in Teslas.

Anticipating your next complaint (“It takes so much longer to recharge!”) - You also save a lot of time by charging at home and never having to stop at a gas station. Overall, I think you save time because of this. It’s not exactly 1:1 equivalent because maybe long drive waits are more costly than frequent short gas stop visits and waits when you aren’t on a long road trip , but it’s really not the end of the world. Gives you some time to stretch your legs.

> Anticipating your next complaint (“It takes so much longer to recharge!”) - You also save a lot of time by charging at home and never having to stop at a gas station.

This is not really a valid argument. The point isn't how much time you spend refueling over a lifetime, but rather how many miles you can travel in a reasonable day for a long road trip (and how flexible your route can be). If long road trips are important to you and you only have one vehicle, of course you probably don't want it to be an EV.

In my view, EVs are incredible, but they're still a hard sell for most single-car households in America. But people should buy the car that matches their needs! No one is better off when fans of a particular type of car deliberately try to convince people to buy that type of car against their interests.

Did you read the rest of my comment? Quite an uncharitable response. I address exactly that...

I still think EVs are okay for a family that does a lot of road trips. You are describing... me. I routinely make a ~12 hour drive. We take a bit of a break every few hours to walk around and play with the dog. The rest of the time, car drives itself.

Hacker News is really getting worse. Taking me out of context by quoting the first line of my comment and ignoring the rest (where I specifically address your comment) is inappropriate. We are better than this.

I read the entire comment. I’m not talking about relative costs of waiting to refuel. I’m just talking about actual effective daily range for a road trip. Refuel times directly affect daily range.

It is obvious that "refuel times affect daily range." The question is whether slower "refuel" times on long trips are a net negative to ownership of an electric vehicle. Most people would say no.

It’s not exactly 1:1 equivalent because maybe long drive waits are more costly than frequent short gas stop visits and waits when you aren’t on a long road trip

As I said...

> It’s not exactly 1:1 equivalent because maybe long drive waits are more costly than frequent short gas stop visits and waits when you aren’t on a long road trip

Your comments have been quite low quality, I think kind of emblematic of how Hacker News has declined throughout the years. It's very easy to be contrarian if you can pick upon a thread, but as far as I remember HN guidelines discourage that.

I'm sure you have not formally crossed Amy HN guidelines, but it kind of sucks just seeing the current state of discourse (eg you) here compared to what it used to be.

Finally - I’d add that “refuel” times DO NOT affect daily range for an electric vehicle. You refuel at home, while you’re asleep. It’s irrelevant in those circumstances.

Okay. I’d prefer you discuss whether my comments are representative of HN’s decline elsewhere. I’m not qualified to judge that.

It may be that, at this stage in the technology, EVs may not suit all families as a single car. But there are many families that do have 2 or more cars and it totally makes sense for one of those cars to be and EV. Eventually, as battery/charger tech matures it becomes easier to satisfy those single car families, too.

even with a 200 mile range, ignoring the actual 330, you are making it from Dallas to Houston with a 15 minute stop in corsicana or huntsville.

I haven't downvoted you, because that doesn't make sense when you merely disagree with a person.

I’m a big fan of the LiFePO4 chemistry. Super Cheap per kWh, abundant raw materials, extremely long cycle life (even without state of the art manufacturing), and usually a lot more stable and much less likely to start on fire. If you want to DIY a lithium battery pack, LiFePO4 is the least suicidal way to do it.

I see LiFePO4 as enabling Africa to electrify quickly and cheaply. It’s feasible to hook up solar panels to an inverter/charger, a cheap and/or DIYed LiFePO4 battery pack, and combine it with a backup generator to have affordable (less than 15¢/kWh... potentially a LOT less) and (just as important) consistent power without needing access to the grid and that’s like 95% carbonfree.

That's great news! Lithium iron is a much safer chemistry, can handle high discharge rates very well, and has a much greater cycle life. Hopefully, the higher demand will bring the same benefits as with lithium cobalt batteries, improving energy density and bringing prices down.

Yes. Lithium iron phosphate batteries can pass the "nail test"[1] and ordinary water sprinklers will extinguish fires.[2] Unlike ordinary lithium ion batteries.[3]

I'll bet Tesla omits the titanium skid plate under the battery for those.

Lithium iron phosphate also makes sense for fixed applications. BYD says they've been able to get the energy density per liter up near regular lithium-ion by improved packaging, but lithium iron phosphate is still about 2x heavier.

[1] https://youtu.be/rb_J2QQ0k-4

[2] https://youtu.be/NeaK9V69Xks

[3] https://youtu.be/f30fBFitkSM

Retail LFP in the US quite a bit more expensive per KWH than LNC though apparently not in China or if you order from there. I wonder why that is.

FWIW I'm a one-car "family" and I wouldn't be bothered by a BEV with 200 mile range if I had a way to charge it on normal nights. I can deal with an occasional car rental or a few stops if I go on a trip. The main thing putting me off BEV right now is that I live in a city without good access to daily charging facilities. If I were in the burbs the satisfaction of charging it from solar panels instead of paying for fuel would make up for quite a few occasional range inconveniences.

The advantage in BYD packaging you can also get to the same extend with other batteries. As Tesla does with their next generation platform.

So that argument doesn't really make sense.

Seriously! I'd be more likely to buy a LiFePO4 car because I know the pack will last basically forever. I don't need incredible range, especially since the supercharger network exists, so that's a pretty sweet deal.

Be nice if it was an option you could order, but whatever.

Be even nicer if some non-Tesla cars would offer it.

I confess I'm more excited for the hobby market getting access to hobbyist-friendlier secondhand batteries with tons of life left. Handling the electricity's just as risky as any other high capacity pack (electrocution and dropped wrenches exploding into plasma when they short the pack and so forth) but at least making a battery management mistake or accidentally drilling a hole in a cell isn't automatically an unstoppable fireworks show in your garage.

BYD, the world's largest electric car maker, offers it.

I was reading a Chinese EV review comparing BYD's Han to Tesla M3.


The Han uses iron cathode whereas the Chinese-made M3 uses the nickel cathode.

The Han is ~6000 USD cheaper, but gets slightly higher range than the M3 (Han has longer city driving range, but worse at high way). This is especially impressive when the Han is ~400KG heavier than the M3.

Both cars have comparable charging speed (31%-100% for 1hr 15min).

The reviewer concluded that the Han is a worthy competitor, but Tesla still has an edge at tuning and design. With Tesla switching to iron cathode, I think Tesla can eek out comparable battery performance while maintaining their edge in other areas.

They are already using LFP chemistry on Standard Range Model 3 sold in China and Europe. (Batteries produced by CATL I believe)

They also called this out during their last battery tech presentation:


I find it interesting that a commercial truck is 'mass sensitive'. Presumably it's because they need a larger battery capacity due to the work needed to haul the load?

Yeah. Consider that your whole truck is limited to 80,000 pounds weight and a typical semi tractor might weigh 20,000 pounds (less for a day cab). Anything above that 20,000 pounds eats directly into your payload.

Still, Tesla should be able to hit their 500 mile range figure with a full typical load and without exceeding the 80,000 pound total weight limit. Takes a lot of work to improve the efficiency of the drivetrain (low battery internal resistance, low power electronics losses, very high efficiency brushless motors, low gearbox and bearing resistance), improving aerodynamics (surprisingly many tractors still seem to have the aerodynamics of a barn), extremely low rolling resistance tires (while maintaining traction), and reducing weight of the tractor through improved and optimized materials, optimized design, etc. ...and then optimizing the battery to have both high specific energy and long cycle life (even while Megacharging).

It’s actually possible to do this, but it requires careful attention to inefficiency and mass throughout the entire vehicle. A lot like designing a reusable orbital rocket.

It's mass sensitive because of regulation. Of course you have to haul your battery around and that impacts range, performance, charging, etc. but not wanting to waste money and time while the DOT writes you a ticket is the primary driver.

The big problem with electric commercial trucks is that the local delivery stuff that is most readily electrified is also the stuff where weight regulations are your biggest thorn in side (for a variety of reasons).

The step van that delivers bread would already be electric if it weren't for the fact that the extra 800lb would push them into a different more expensive regulatory class where they become unprofitable or displace so much cargo they become unprofitable.

Don't forget road deterioration goes like something between the 3rd and 4th power of mass. If one of your concerns is environmental impact, then minimizing the amount of road being torn up is something you care about. Whether that's on the actual list of reasons is unknowable to us.

Yeah, to the third or fourth power assuming the same width, diameter, and number of wheels. So there are ways around that from an engineering perspective. The Tesla Semi has baselined Super Singles (a single tire as wide as a double tire, more efficient with equivalent or better traction and lower overall weight), I believe, which should help a bit with this.

But this is why road regulations limit weight and weight per axle.

When you are all-in on electric, I wonder if it could make sense for a truck to accept the unsprung weight penalty of wheel motors and go for uniform, independently steerable, suspended, self-monitoring and motored wheel-units that are simply repeated under the vehicle as often as needed for the load/power requirements and controlled with clever software that takes care of driving dynamics etc. Like a lazy copy/paste design, like all post-Tesla EV batteries are designed. It would be terrible for center of gravity because all the room between where conventional designs have their wheels would be filled with more wheels (and the accompanying suspension) but it would be a one size fits all design that I'm sure could have many benefits beyond just reducing per wheel load if the controlling software is smart.

Tesla’s Semi prototypes do use 4 BLDC motors (well, more complicated than straight BLDC) from the Model 3. But I think optimization is really key for making it economically viable for mass cargo transport. Lazy copy-paste is great for prototypes and really shows how awesome electric is for making new vehicle configurations, but you really HAVE to optimize to be competitive at scale. Tesla has succeeded over the last 9 years against all comers BECAUSE they’ve optimized. Of course, manufacturability is one of the optimization constraints...

Super singles are for rolling resistance.

For pneumatic tires on highway speed vehicles your ground pressure is approximately your tire pressure. Super singles use pressures in the same ballpark as the duals they replace.

Well yes, reduced road wear is not the primary advantage of super singles. The contact area is slightly larger, which can help slightly (things such as total load and outside dimensions and speed being equal—if you like, it can allow a slightly lower tire pressure). It’s still in the same ballpark, as you say, but it should slightly improve road wear characteristics.

Why not have a three-wheel axle, with two wheels at the edges like in conventional vehicles, and the third wheel riding in the middle of the lane? That would transfer ~30% of the load to a portion of the road which is not wearing down anyway.

Stability. Roads very often have curvature built into them for rain/runoff/erosion mgmt. If your road surface is an arch, having a wheel at the apex is not gonna go well. You gon' tip over.

Interesting, thank you. I wonder how well a reduced spring pressure would work for that case, though it would negate much of the advantage of transferring load to the center of the lane.

Likewise, an interesting potential solution to the problem. I hadn't thought of a lower spring rate. It would certainly reduce the braking distance. A lot more complexity, cost, and weight, though.

How do Super Singles deal with punctured tires? I'd think one advantage of dual tires is that if one pops, there's a higher likelihood of the truck remaining drivable?

>How do Super Singles deal with punctured tires?

You sit on the side of the road waiting for the tire service to show up.

>there's a higher likelihood of the truck remaining drivable?

"DOT would like to know your location"

You're technically not supposed to be operating with a blown tire. That said, driving to the nearest place you can reasonably get or wait for a replacement is the correct course of action.

Minimizing environmental impact requires considering the vehicle count as well.

Having a more light duty trucks zipping around because we've regulated fewer medium duty truck out of profitability for that use case is likely not environmentally efficient.

When you're talking about roads expected to last X years with a certain % traffic of semi trucks in the 80k ballpark pretty much nothing you do in the 40k ballpark and below matters.

Also, you have to differentiate between wear on the road surface and the road bed. The road surface doesn't care how much weight you have. It just cares about contact pressure. The road bed is going to care more about overall tonnage because by the time the force gets to the road bed it's spread over such an area the contact pressure is low regardless.

It's hard to generalize these things because mother nature plays a large role what the depreciation curve of a road looks like and local wealth plays the deciding role in where in the depreciation curve you justify replacing it.

> Having a more light duty trucks zipping around because we've regulated fewer medium duty truck out of profitability for that use case is likely not environmentally efficient.

Really? I thought it was personnel-inefficient but pretty much ideal for the roads, for the aforementioned power-law reason. Lighter trucks are gentler on the roads even if there are more of them.

Do I misunderstand?

There's carbon, particulate and road wear (and noise and traffic). You can prioritize whatever you want in whatever order you want to make whatever state you want seem like the best.

If the road has to be built to handle X% semi trucks for 10yr and mother nature will destroy it in 20yr no matter what you do you lose nothing by driving a bunch of medium duty trucks on it.

If I understand correctly then it will be lithium iron phosphate? If so that seems like the best chemistry in my limited understand. I've got a huge Lifep04 battery I use to run equipment and the cycle, power density, failure mode, etc are all amazing.

The only problem for vehicles is that LiFePO batteries are heavy. (IIRC ~60% the energy density of Lithium ion)

This is where Tesla's structural pack comes into play. They drop a bunch of weight by turning the pack into part of the car structure (rather than something it carries).

But that same technology is also available for nickel based batteries. So the relative performance doesn't change when comparing chemistry.

Certainly, to me it is more that they can maintain the total range of the car even with a decreased energy capacity. Because they are the only manufacture perusing this (AFAIK) this gives them a competitive edge.

My 16 272 Ah LFP cells arrived today (February 26) in France from China (shipped January 11 - 47 days ago).

13.9 kWh for $1248 + $416 shipping and tax, so just shy of $120/kWh delivered ($90/kWh before shipping).

Will be used with an hybrid inverter and solar panels to reduce my electricity bill, UPS my home, and charge my Tesla Model 3 SR+ (2019, USA built so not LFP).

The 3 to 5x difference between those cells and commercial offerings make me hopeful we'll see the market for home batteries take off relatively soon. LFP are ideal for the application and are already cheap. With the cost of rooftop solar continuing to dive, getting a house to be much more energy independent, or even fully off-grid will become almost easy in the next 5 years.

Before electric cars it was hard to really use solar PV produced energy.

But with one or two electric cars and the appropriate amount of battery you'll get way higher PV production use by just charging after work and week-ends.

If my 14 kWh pack does 4000 full cycles that's 56 MWh so 0.03 USD/kWh out of the battery.

$600-700 for a 5kW hybrid inverter (stackable)

PV panels are about $0.4/Wpeak and depending on location, in France 1 Wp will produce about 1kWh/year.

Which brings price under $0.10/kWh for a DIY solar + battery system of 4 kWp PV and 14 kWh battery.

Tesla apparently voids your battery warranty if you use the car as a stationary power source. That's mentioned in tiny print in the warranty documentation someplace. I haven't seen this myself, but people mentioned it during the recent Texas blackout where many people had the idea of using a BEV as a battery bank.

Not sure what you mean by needing an EV to use the energy. With an A/C system you can heat and cool the house with electricity efficiently. Since that works even if your car is away with you at work the battery size is reduced making that load a better business case. The same for heating water, which is a cheap battery in itself.

If anything EVs seem particularly unsuited for home solar because the solar panels should instead be wherever the car spends its day so the car battery can be charged directly instead of needing another battery to buffer the energy for it.

In France you don't need heating/cooling for most of the year with a moderately insulated house.

Thats's why electric car charging changes the equation.

And about no heating/cooling at all with a Passive House :).

I agree about charging during the day when the sun shine is way better but it's under the control of your employer which may or may not provide chargers, and might not install solar PV at all for various reasons.

PV and battery at home, you can do it right now.

For a long time LFP prices stayed at $200-$300/kWh but mid-2020 new vendors with $100-$150/kWh delivered appeared.

I don't know the story behind these cells but so far no bad experience reported on youtube or forums.

Here is one of the first buyer who just measured his pack after 6 month of use:


Did you validate capacity? $90/kWh looks great.

Not yet, will do in the coming days.

Where did you order them?

There are a few known good vendors based on feedback from https://diysolarforum.com

I chose this one:


Maybe this? https://news.ycombinator.com/item?id=26278184

Haven't looked yet.

The tweet talks about swapping iron for nickel, not iron for lithium.

Yup. In spite of the “lithium is the new oil” meme, lithium is actually super abundant and is often extracted from brines the same way as sea salt is. Lithium isn’t where to focus replacement efforts on.

Well, TIL. Though it might not be so clear cut: https://www.wired.co.uk/article/lithium-batteries-environmen... (article from 2018).

Well sure, even salt extraction can have an impact. Have you seen the salt extraction facilities at the Dead Sea? Someone could write an exposé about how our demand for salt is a “big problem” illustrated with dramatic pictures. But we need some perspective here, and anecdotes are way too easy to cherry-pick. The amount of lithium needed is minuscule compared to the amount of oil. 5 kilograms of pure lithium for a Model 3 versus like 20000kg of gasoline for an equivalent fossil fuel car. And the lithium can be recycled at end of use. But what’s 3 or 4 orders of magnitude between friends?

There is lots of lithium. But lithium is kind of a tricky, its almost more chemical then a base metal. Qualifying a new lithium extraction technology is tricky, and every brine is different.

Lithium is not tricky to find, its trick to extract and purify and the it takes a while to qualify it with battery companies. Doing all that takes a while and doing it cheap is not easy.

So lithium prices, might still go up because getting high quality supply into the supply chain is not easy.

Check out:


The have a large number of videos, where many of the upcoming lithium producers present their projects.

The economical ways currently are:

- Spodumene (lithium in hard rock minerals)

- Brine (lithium in a salty water underground pond)

What is being developed for next generations:

- Clay (sedentary deposits)

- Deep Geothermal Brine (like Brine but much deeper down)

What is being worked on is Direct Lithium extraction, that means to get lithium directly from the brine (or leached from the clay) rather then putting it into evaporation ponds.

If you want real detail from an expert on lithium, check out:


here in Alberta E3 metals will extract it from old oil wells, direct extraction no ponds.

from lithium-ion to iron-cathode

Edit: article source was wrong https://www.carscoops.com/2021/02/tesla-shifting-battery-typ... I corrected the tile

They aren't switching from lithium. In the tweet he even says there's plenty of lithium. They are switching from a nickel cathode to an iron cathode due to the scarcity/cost of nickel.

The title should definitely be changed. It makes it sound like they are going to a completely different battery technology.

Oh sorry, my mistake, I copied-pasted fromn this article: https://www.carscoops.com/2021/02/tesla-shifting-battery-typ...

Edit: corrected the title

LFP on standard range is not news, they stated it on battery day I think. Especially since allegedly CATL is making good gains to LFP density.

LFP is cheaper, has insane cycle endurance, is safer, and has great temperature tolerance.

LFP reaching the 200 mile range for a car is a watershed engineering moment for humanity. NMC and others can be dedicated to other tasks.

Out of curiosity, are there any online courses that one can take/watch to learn more about battery chemistry and/or battery economics?

Youtube channel The Limiting Factor is going deep on battery technology and economics:


Fun fact: you can get an LFP battery for your house, https://enphase.com/sites/default/files/downloads/support/En...

A smaller, lighter, and slightly cheaper car (which presumably would need fewer batteries) would be a great match for Europe. Driving through and parking in these ancient cities is a real pain with a big car.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact