Cars (all fuels and electric)

maxbc wrote:

This is starting to look like VFR dispatch rates, which is fitting :)

:-D except for VFR I have a weather forecast and can plan accodringly, so it is actually better!

maxbc wrote:

the average capacity degradation is 2.3% per year (1.6% under ideal temperatures), and is not very significantly dependent on the amount of miles driven

Long term statistics on the Tesla fleet have given 15% degradation after 9 years and/or 200,000 miles, so even less than that.

I had around 2% degradation after 4.5 years, nothing in year 1 and 2 and then 1% per year which is consistent with the lower numbers. But perhaps Britain with the lack of extreme hot or cold weather is “ideal conditions” for batteries.

I wonder sometimes how this is measured – does the BMS actually measure charging resistance and report that, or is there guesswork involved?

the average capacity degradation is 2.3% per year

This is about 10x better than LIPO batteries in anything I have ever used, so we have a few options

• the battery is oversize and they fake the degradation, only to allow it to catch up after say 10 years by which time it will be out of warranty
• the battery capacity calculation is simply faked (the EV forums are full of people saying they get much less than the claimed endurance) and frankly it would be very hard for an individual owner to check this because he has no way to tell the energy drawn by the vehicle etc… IOW how is this calculation made? The only proper way is a full discharge, a full charge, and a load test.
• they have invented a new type of electron which is heavier and orbits slower around the nucleus

For me, the sanity check is against all the other consumer appliances.

Cobalt wrote:

Long term statistics on the Tesla fleet have given 15% degradation after 9 years and/or 200,000 miles, so even less than that.

That is 1.8% per year… Closer to the “ideal temperature” that maxbc gave.

They’re not LiPo batteries but Li-Ion, the difference may come from there.

Using phones at 1 cycle per day (my previous phone really did that, like 15-100% every single day), it seems about right to be at 80% capacity after 3 years. So the 1000-1500 cycles seems about right, and if you do a cycle per week (in this way a big capacity EV is actually a good thing even if you don’t do long trips at max range), you would reach 1000 cycles after 19 years, which means some kind of time or thermal (passive) degradation takes the lead compared to cycle degradation. My previous laptop was in the same ballpark, reaching (from memory) ~90% capacity after 3.5 years doing about a cycle a week.

So using this sanity check it’s very reasonable.

Sure, some cars “fake” it, sometimes ecplicitly so (“108 kWh battery (95 usable)” and that buffer between 108 and 95 gets eaten into). But not by much because the WLTP range is very important for sales so doing anything that reduces the range they can publish will backfire. Any “test cycle detection devices” will probably have been disabled at this point after the VW emissions scandal.

The “my range is worse than advertised” is just a simple fact of life as it is in petrol cars – nobody drives like the test cycle. Quite interesting to watch on longer trips; lowering the speed from 80 to 70 mph and cruising along does wonders for the range.

The main reason that car batteries last longer than the LiPO batteries you find in drones and model is that they are different types of batteries; the electrons are the same but the eletrodes are not.

• Laptop, mobile phone and other batteries tend to be Lithium Cobalt Oxide batteries (“LCO”, LiCoO2 cathode, graphite anode) and can deliver 200-250 or more Wh per kg. The trade-off is that they don’t like charging or discharging at high currents and are prone to thermal runaway when mistreated.

• Car batteries tend to be lithium-nickel-cobalt-manganese oxide (NCM) or Lithium-nickel-cobalt-aluminum oxide (NCA). They have lower energy densities (150-200+) but can stand higher charging and discharge currents and (if treated well by the BMS) live longer than LCO batteries, although NCM is better than NCA.

• Solar batteries tend to be lithium-iron-Phosphor 0xide batteries (LiFePO4, short LFP), they have even lower energy densities but where mass does not matter they are great as they live the longest of them all, and are also very hard to set on fire.

There is progress all of the time increasing the capabilities of these technologies, so this changes all the time; the same model of the same car can have different chemistries depending on where and when it was built. Many other chemistries exist. In general this progress will only slightly increase energy density which is simply limited by the chemistry, but longevity and our understanding how to best manage, charge and discharge batteries has advanced quite a bit.

maxbc wrote:

but Li-Ion

I am not sure people are actually interested, so if not someone please tell me to stop…

All rechargable lithum batteries are Li-Ion. They work by moving electrons “the long way round” through the wires while the Lithium moves through the electrolyte in a reaction that can be reversed. The main differences are the material of the elecetrodes and the electrolyte (polymer or liquid/gel).

For example LCO batteries start with Lithium and Graphite (Li C6) at the negative electrode (cathode) and with Cobalt Oxide (CoO2) on the positive electrode (anode).

The Lithium has one valence electron and a fairly low “electronegativity” compared to oxygen and cobalt which means that forming a bond with the Cobalt-Oxide will release energy, and this creates a voltage across the two electrodes.

As the battery discharges, an electron separates from the Lithium and goes through the wire, and the positive Lithium Ion that was created moves through the electrolyte to the anode.There it “reunites” with the electron and forms part of LiCoO2. You could grind down the electrodes and mix them as a powder and would get the same reaction, but that would just generate heat.

When the battery is charged, the whole process is reversed.

Batteries degrade because these reactions are not perfect, and overcharging and deep discharging creates undesirable compounds where the reaction CANNOT be reversed.

There are several ways these batteries differ

• The exact materials used for each electrode [as in my list above]; this can include adding silicon to the graphite to have more lithium in a lower volume at lower mass
• The electrolyte
• The physical properties (surface area / porosity, distances, etc etc) of these materials

But the chemistries are all based on oxidizing Lithium, and the theoretical maximum capacity is exactly one electron per Lithium atom contained in the battery

Peter wrote:

This is about 10x better than LIPO batteries in anything I have ever used, so we have a few options

LiPo batteries are probably the worst performing (in terms of life) type of Li-Ion battery around. We also tend to use them the hardest – we want them to take as little volume as possible for our devices and they have to endure full charge and discharge cycles to give the best time in a very small device.

I used to use A123 LiFePo batteries for my model helicopters, and they were at least in this ballpark in terms of degradation over time. In terms of degradation/cycle, I can’t say, but they were meant to be capable of 1000s of cycles back when other lithium batteries degraded far more quickly.

I’d suggest the truth probably lies somewhere in the middle.

My natural scepticism suggests to me that the software probably fakes the degradation to some extent, and in terms of the figures being published I wouldn’t trust anything from a manufacturer.

The proof of the pudding will be in the eating, and it’d be interesting to see what sort of range a 15-20 year old EV can get. Even on the Tesla data, the annual drops after 7 years are getting pretty significant and I suspect the annual drop gets worse over time.