High Voltage (HV) Battery - Best Charging Practice?

[Bricktop X PWR]

99% Tesla Superchargers to 97%

By the time I get home, the battery is at 80 to 85%

Balance once a month to 100% + balance on 7kw car park post. (Pub Lunch)

SOH after 1 year and 9000 miles is 100%

I never lost any sleep over my battery before, and I won't start now.

 

[MickeySw]

The car thinks it knows the SoC but it is a guess that could be 10% out. SoC is really difficult to measure, especially in an environment where the battery is constantly being discharged and charged at differing rates.

I would have thought that Coulomb counting is more accurate than that but regardless, the main ‘unknown’ variable in range calculation is future consumption..

 

[shpub]

Coulomb counting is not that accurate either if you want to calculate a battery SOC. It's accurate in that you can measure the current going in and out which will tell you how much can be extracted. Liion EV batteries are around 90% efficient at best, but it is very dependent on a lot of factors and that loss can be a lot bigger in extreme conditions: temperatures, discharge currents and rates, charging, cell balance and so on. 10 kWh may have put into the battery from the coulomb count but only 9 kWh can be extracted.

The usage or the mi/kW figure is pretty well known because it can be measured: current used and miles travelled. This gives a pretty good idea of what the average consumption would be. Yes it can vary depending on how it is driven but high discharge currents is one of the main contributions to the battery efficiency: drive it like you stole it will increase the energy consumption and also reduces the battery efficiency so that you can't get all the energy that the battery could offer. So we have further reduction in the range. A double whammy.

Most drivers don't do that and drive fairly consistently (MG4 Xpower owners may be an exception ) so the biggest unknown is really the battery and how much energy can be extracted which will not be the same as was measured going in. It is really hard to measure accurately especially as the future consumption has a double whammy: increase the consumption and not only does the mileage estimate goes down but the battery efficiency and thus the amount of energy that is available is also reduced, making the range even less!

As to which is the bigger figure? The future consumption is probable the most accurate providing the driving style doesn't change. The battery SOC though is probably the hardest to get any real idea of accuracy. That is why it is called a "Guessometer".

 

[jpvdheijn]

A Paper of Jef Dahn et al about the lifetime of LFP.

The lifetime can be drastically increased by cycling at a lower state of charge like with NMC or NCA batteries

The narative that a LFP battery does not suffer by charging often to 100% is wrong as found by this research.

The 100% story is solely to give the BMS a better estimate on the SOC of the battery

https://iopscience.iop.or...1149/1945-7111/ad6cbd/pdf

 

[shpub]

I am not so sure that what they found is actually relevant to EV use.

The full paper is here: Radware Bot Manager Captcha

Note the high temperatures used in the testing with battery temperatures of 55 deg in some cases and taking the battery voltage down to an SOC of 0. The batteries were abused and saw charging and temps that would not be seen by an EV as the BMS would start to have a fit. The battery tech is not one that I know off that has been used in EVs and their technique of using battery voltage to indicate SOC seems to be a bit suspect - they did select a cell that had a sloping voltage to help improve that. The long and the short of it is while they try and refer this to EV usage - Tesla were one of the companies that gave them the money to research it - it does not necessarily follow that the same effects would be seen in an EV application.

I wouldn't worry about it at all.