plj
Posts: 38
Joined: Sun Sep 03, 2017 9:52 pm
Location: Davis, California

Charge: As soon as possible, or as late as possible?

Sun May 20, 2018 6:54 pm

Top line: Charge before your trip, not after your trip.

Here's an article that caught my eye:
Enhancing electric vehicle sustainability through battery life optimal charging.
Transportation Research Part B 112 (2018) 1–18
https://doi.org/10.1016/j.trb.2018.03.016

The abstract:
"In this article, we investigate the potential for battery life prolongation through optimized charging under consideration of individual mobility requirements. Based on a comprehensive battery aging model we introduce a continuous quadratic programming model to derive battery life optimal charging (OPT). The strategy indicates when and how much to charge to maximize the potential range throughout the battery life. We find that OPT has the potential to more than double the expected battery life compared to simple and often abundant recharging activities as observable today. The degree of battery life prolongation strongly depends on the operating temperature. Since optimal charging would require deterministic knowledge of future trips and corresponding charging levels we investigate a more convenient charging heuristic derived from “As-Late-As-Possible” (ALAP) charging. ALAP charging considers range buffers between 5% and 60% over the range required until the next re-charging opportunity. We analyze the trade-off between (long-term) battery life and (short-term) range flexibility. We find that for decreasing temperatures the trade- off between battery life and flexibility is solved with increasing range buffers. From our results battery degradation aware charging heuristics can be easily derived and applied in real-world settings."

While I don't pretend to have read the article, let alone understand it, I interpret it to mean charge as late as possible. That is, minimize time the battery is at a high state of charge.

plj
Posts: 38
Joined: Sun Sep 03, 2017 9:52 pm
Location: Davis, California

Re: Charge: As soon as possible, or as late as possible?

Sun May 20, 2018 7:05 pm

Here's more from the paper:

6. Conclusion and implications
While latest studies on EV user behavior indicate that users prefer frequent and full recharging (AFAP), changing this charging behavior can tremendously extend battery life. Based on simulation results built on a comprehensive battery cell aging model and empirical mobility data, we show that a battery degradation minimal (optimal) charging strategy (OPT) extends battery life by a factor of two or higher. AFAP is especially harmful in cases of higher average operating temperatures. OPT is close to as-late-as-possible (ALAP) charging at high temperatures of 35 °C. OPT is highly correlated and converging towards ALAP for lower temperatures.

However, ALAP and OPT require full information about the next-range requirements, that cannot be expected to be available precisely in real-life settings. We therefore investigated the trade-off between flexibility and battery life by introducing flexible range buffers between 5% and 60% to ALAP, i.e. ALAP b .

We find that a lower range buffer of 30% is beneficial for high temperatures (35 °C). For decreasing temperatures the best trade-off between battery life and flexibility is achieved with increased range buffers, i.e. 35 and 50% for a temperature of 20 and 10 °C, respectively. For low temperatures, which can be achieved for example using battery cooling systems, ALAP b charging with a range buffer of 50% can be applied as an easy-to-use charging heuristic and allows for both battery life extension, flexibility and therefore user convenience.

In summary, while none of the presented ALAP b strategies including range buffers perform close to OPT, the harm of range buffers reduces with decreasing temperature such that this trade-off is less pronounced in climate zones with average (operational) temperatures around 10 °C or with active battery cooling systems that enable a performance in such a temperature range. However, ALAP b charging can be implemented as an easy-to-use smart charging heuristic, that leads to considerable battery life extension compared to the currently applied, naive AFAP charging.

GetOffYourGas
Posts: 1151
Joined: Tue Oct 20, 2015 5:25 pm
Location: Syracuse, NY

Re: Charge: As soon as possible, or as late as possible?

Wed May 23, 2018 2:25 pm

I find it amusing that they refer to "AFAP" (which is the same as the "ABCs" - Always Be Charging - you hear pronounced on forums) as "naive" charging.

I don't have access to the article, nor am I about to pay for it, but the gist is lower SoC is better, which is widely known. You have to balance that with the buffer you personally feel comfortable with. Unfortunately, GM has not given us sufficient controls to let us do this with our Bolts.
~Brian

EV Fleet:
2011 Torqeedo Travel 1003 electric outboard on a 22' sailboat
2012 Leaf SV (traded for Bolt)
2015 C-Max Energi (302A package)
2017 Bolt Premier

SparkEVPilot
Posts: 105
Joined: Thu Feb 02, 2017 4:42 am
Location: Manteca, California

Re: Charge: As soon as possible, or as late as possible?

Wed May 23, 2018 7:08 pm

plj wrote:Here's more from the paper:

6. Conclusion and implications
While latest studies on EV user behavior indicate that users prefer frequent and full recharging (AFAP), changing this charging behavior can tremendously extend battery life. Based on simulation results built on a comprehensive battery cell aging model and empirical mobility data, we show that a battery degradation minimal (optimal) charging strategy (OPT) extends battery life by a factor of two or higher. AFAP is especially harmful in cases of higher average operating temperatures. OPT is close to as-late-as-possible (ALAP) charging at high temperatures of 35 °C. OPT is highly correlated and converging towards ALAP for lower temperatures.

However, ALAP and OPT require full information about the next-range requirements, that cannot be expected to be available precisely in real-life settings. We therefore investigated the trade-off between flexibility and battery life by introducing flexible range buffers between 5% and 60% to ALAP, i.e. ALAP b .

We find that a lower range buffer of 30% is beneficial for high temperatures (35 °C). For decreasing temperatures the best trade-off between battery life and flexibility is achieved with increased range buffers, i.e. 35 and 50% for a temperature of 20 and 10 °C, respectively. For low temperatures, which can be achieved for example using battery cooling systems, ALAP b charging with a range buffer of 50% can be applied as an easy-to-use charging heuristic and allows for both battery life extension, flexibility and therefore user convenience.

In summary, while none of the presented ALAP b strategies including range buffers perform close to OPT, the harm of range buffers reduces with decreasing temperature such that this trade-off is less pronounced in climate zones with average (operational) temperatures around 10 °C or with active battery cooling systems that enable a performance in such a temperature range. However, ALAP b charging can be implemented as an easy-to-use smart charging heuristic, that leads to considerable battery life extension compared to the currently applied, naive AFAP charging.

OK. Now explain this in layman's terms so an old geezer like me can understand what they are suggesting.

SparkEVPilot
Posts: 105
Joined: Thu Feb 02, 2017 4:42 am
Location: Manteca, California

Re: Charge: As soon as possible, or as late as possible?

Thu May 24, 2018 3:47 pm

SparkEVPilot wrote:
plj wrote:Here's more from the paper:

6. Conclusion and implications
While latest studies on EV user behavior indicate that users prefer frequent and full recharging (AFAP), changing this charging behavior can tremendously extend battery life. Based on simulation results built on a comprehensive battery cell aging model and empirical mobility data, we show that a battery degradation minimal (optimal) charging strategy (OPT) extends battery life by a factor of two or higher. AFAP is especially harmful in cases of higher average operating temperatures. OPT is close to as-late-as-possible (ALAP) charging at high temperatures of 35 °C. OPT is highly correlated and converging towards ALAP for lower temperatures.

However, ALAP and OPT require full information about the next-range requirements, that cannot be expected to be available precisely in real-life settings. We therefore investigated the trade-off between flexibility and battery life by introducing flexible range buffers between 5% and 60% to ALAP, i.e. ALAP b .

We find that a lower range buffer of 30% is beneficial for high temperatures (35 °C). For decreasing temperatures the best trade-off between battery life and flexibility is achieved with increased range buffers, i.e. 35 and 50% for a temperature of 20 and 10 °C, respectively. For low temperatures, which can be achieved for example using battery cooling systems, ALAP b charging with a range buffer of 50% can be applied as an easy-to-use charging heuristic and allows for both battery life extension, flexibility and therefore user convenience.

In summary, while none of the presented ALAP b strategies including range buffers perform close to OPT, the harm of range buffers reduces with decreasing temperature such that this trade-off is less pronounced in climate zones with average (operational) temperatures around 10 °C or with active battery cooling systems that enable a performance in such a temperature range. However, ALAP b charging can be implemented as an easy-to-use smart charging heuristic, that leads to considerable battery life extension compared to the currently applied, naive AFAP charging.

Is this what they are suggesting:

I arrive home with 30 miles on the GOM and park my EV for the night. The next day I expect to drive 50 miles. So, knowing the average mi/kWh for my vehicle, the expected air temperature the next day and the road over which I will travel, I calculate how many kWh I need to add to my HV battery to cover the mileage I expect to drive plus an acceptable buffer. The initial GOM mileage may be considered the "buffer" for air temperature effects plus any unexpected factors that may impact my average mi/kWh. And, this number may have to be adjusted up or down depending on environmental and road concerns. Once I have determined how many kWh I need to add to my battery, plus any charging losses, I determine the required number of hours of charging and program my EVSE to have the EV charged sufficiently to cover my trip when I am ready to leave the next day. Then, if I do not plan to use my EV for a period of time, I leave the battery at a low SOC and do not recharge the battery until just before my next trip.

In short, do not charge more than what is needed to cover the trip and return home with something still in the battery and do not recharge the battery until just before you plan to use the EV again for another trip. Is this what the article suggests?

GetOffYourGas
Posts: 1151
Joined: Tue Oct 20, 2015 5:25 pm
Location: Syracuse, NY

Re: Charge: As soon as possible, or as late as possible?

Thu May 24, 2018 4:51 pm

That was how I read it too. Sounds like a lot of work, but they are claiming double the battery life over "naive" charging. I don't know under what conditions, though.
~Brian

EV Fleet:
2011 Torqeedo Travel 1003 electric outboard on a 22' sailboat
2012 Leaf SV (traded for Bolt)
2015 C-Max Energi (302A package)
2017 Bolt Premier

SparkEVPilot
Posts: 105
Joined: Thu Feb 02, 2017 4:42 am
Location: Manteca, California

Re: Charge: As soon as possible, or as late as possible?

Thu May 24, 2018 9:37 pm

GetOffYourGas wrote:That was how I read it too. Sounds like a lot of work, but they are claiming double the battery life over "naive" charging. I don't know under what conditions, though.

There may be something to this if you consider that every electronic device powered by a lithium battery is usually only partially charged [ low SOC] when it is first purchased. Is that done to prolong the battery life between the time of manufacture and the time of purchase? I wonder, given we did not fully recharge our lithium battery powered device every night, if the battery would last longer?

GetOffYourGas
Posts: 1151
Joined: Tue Oct 20, 2015 5:25 pm
Location: Syracuse, NY

Re: Charge: As soon as possible, or as late as possible?

Fri May 25, 2018 12:53 pm

SparkEVPilot wrote:
GetOffYourGas wrote:That was how I read it too. Sounds like a lot of work, but they are claiming double the battery life over "naive" charging. I don't know under what conditions, though.

There may be something to this if you consider that every electronic device powered by a lithium battery is usually only partially charged [ low SOC] when it is first purchased. Is that done to prolong the battery life between the time of manufacture and the time of purchase? I wonder, given we did not fully recharge our lithium battery powered device every night, if the battery would last longer?


I have an iPhone 5C that I try not to charge above 80%. After almost 5 years, the phone still holds a charge pretty well. Others who plug in their phone all night every night (so it sits at 100% for 10-12 hours/day) have seen their battery life go to nothing. Anecdotal, but true.
~Brian

EV Fleet:
2011 Torqeedo Travel 1003 electric outboard on a 22' sailboat
2012 Leaf SV (traded for Bolt)
2015 C-Max Energi (302A package)
2017 Bolt Premier

SparkEVPilot
Posts: 105
Joined: Thu Feb 02, 2017 4:42 am
Location: Manteca, California

Re: Charge: As soon as possible, or as late as possible?

Fri May 25, 2018 8:12 pm

GetOffYourGas wrote:
SparkEVPilot wrote:
GetOffYourGas wrote:That was how I read it too. Sounds like a lot of work, but they are claiming double the battery life over "naive" charging. I don't know under what conditions, though.

There may be something to this if you consider that every electronic device powered by a lithium battery is usually only partially charged [ low SOC] when it is first purchased. Is that done to prolong the battery life between the time of manufacture and the time of purchase? I wonder, given we did not fully recharge our lithium battery powered device every night, if the battery would last longer?


I have an iPhone 5C that I try not to charge above 80%. After almost 5 years, the phone still holds a charge pretty well. Others who plug in their phone all night every night (so it sits at 100% for 10-12 hours/day) have seen their battery life go to nothing. Anecdotal, but true.

The Chevy Spark EV and the Chevy Bolt EV both have a charging option called "Programmed Departure Time". This option allows one to specify the departure time at which the car will be fully charged. So, with a little calculation work, a driver can set the programmed departure time at a time later than the actual disconnect / departure time so the car is only 80% fully charged when the car is disconnected. You just need to know the amount of time it takes to charge from 80% to 100% for whatever charging voltage and current you are using. For example, it takes my Spark EV 4.5 hours to charge from 80% to 100% full charge when using an L1 EVSE at 120 VAC and 8 amps. So, if I want to leave at 8 am with 80% in my battery, I can set my departime time to 12:30 pm and just disconnect at 8 am when I am ready to leave. I will be trying this during the next week or two using L1 at 120 VAC at 8 amps and 12 amps and L2 at 240 VAC at 16 amps.

trevmar
Posts: 25
Joined: Sat Jan 21, 2017 8:08 pm

Re: Charge: As soon as possible, or as late as possible?

Fri May 25, 2018 11:32 pm

FWIW I'm an electrochemist who spent (or wasted) over 40 years trying to make electrons swim. In the 1970's we found users of nickel-cadmium batteries in walkie-talkies got much better life when the devices were being heavily used (and abused) as opposed to spending all their lives on "float" charge. I didn't work much on lithium-ion but the principles are the same - when the electrodes are fully charged it is very stressful on them and the separator, you are driving the material lattice structure to a state that is not 'normal'.

Think of metals - very few (gold, platinum) are found native in nature, the rest always appear as salts like oxides, sulfates, etc, and we then put enormous amounts of energy into them to smelt them into useful base metals like lithium, iron, aluminum, etc. What's the first thing iron (steel) does when it sees the real world? Rusts! (i.e. reverts to the state it was originally in the earth)

So in my view a fully charged battery is inherently unstable - clever design will help get the maximum amount of useful electrical energy out of it, but it doesn't enjoy being in that state, and storing it (especially at high temperatures, which accelerate all chemical reactions) makes it worse

Entropy rules our universe (apparently), even though I probably made more money knowing how to galvanize a bucket than actually understanding the theory.

Return to “Batteries and Charging”