Limiting charge to 80%, Why (or why not?). Some JuiceNet features

There is some debate out there whether charging to 80% (as opposed to full) gives you better battery health in the long run. Various factors include a potentially dodgy battery chemistry in the first generation of LEAFs, a class action lawsuit in the US, and general theories on lithium battery technology.

However, the following factors make me believe that in general, the 80% setting is not necessary for battery longevity. It may make a difference, but should not be a significant factor for more recent models. When you charge to a full 100%, the actual cell voltages get to around 4.1V, which actually corresponds to around 90% according to typical lithium battery discharge curves. (A really full battery should hit 4.2V). So if you aimed for 80% charge on the dashboard, you're actually more like 70% or so. Heat and time still seems to be larger factors, from most sources I've seen. Another factor that I think plays a huge role is how aggressively you cycle the batteries, which means a.) Quick charge use and b.) aggressive acceleration and regen. Of course, if you look far enough there'll always be anecdotal evidence to the contrary, but most general information out there seem to be relatively consistent.

So, why would we want to charge to 80%? Actually, after some further testing, we've started going towards 90%... For us, there are the following reasons:

1.) Given our typical commute of less than 20km per day, we do not need the full capacity of the battery

2.) Perhaps more importantly, having a full charge limits the amount of regeneration available. Since we do mostly city driving with a lot of start and stops, and it is not always feasible to coast towards every stop, being able to regen means that in the long run, we will be saving energy.

3.) Another energy saving factor is that since there is a fixed charging overhead of around 300W (to run the cooling pumps and other charging circuitry), charging beyond 80% or so becomes significantly less efficient due to the charge rate slowing down as the cells are being topped up and balanced fully. Basically at that point, you continue to consume ~300W but the % gained by the battery slows down significantly. Having said that, the process of cell balancing is important and should be done once in a while. Our current habit is to do a full balance every 3 or 4 charges, which works out to be around once every two weeks (depending on weather and driving needs).

So, how do we do it?

Without a smart EVSE, one can simply set the car's charge timer to a slightly later time than the actual departure. This will hold off charging when you plug the car in and only start X hours before your departure time. X is calculated based on the power of the supply you're plugged into, and ranges from about 5-7 hours on full Level 2 (240V 27A) to 14+ hours on Level 1 (110V 15A) for a typical charge. This requires some guesswork which gets better with experience as you have to figure out how much exactly in advance you set the time to hit the charge level you need. However, since we already invested in a JuiceBox, it's a lot more convenient to use the built in features which are accessible from the smartphone app and web consoles. Here's what the main charging screen looks like:

As we've set the vehicle to a 2014 Leaf, it knows that 24kWh of energy is a full battery. From there it allows you to set a target value for each charge session. Above we see the small triangle on the gauge set to 60%. What this will do is limit the amount of energy piped to the car to 24*.6 = 14.4kWh.

It should be noted that the amount of energy that is metered through the EVSE will not be exactly what the battery gains, due to a variety of loss factors and estimation of actual charge levels on the dash. In our experience, what you ask the JuiceBox to give to the car, compared to what the car actually gains, is actually 5-10% more when aiming for a target in the 80-90% range. I expect the value to be closer when the charge level is in the more linear voltage vs energy level range (i.e. below 80%).

In addition to the charge limit, we can also apply the "Time Of Use" feature to start charging at a later point in time:

Here in Quebec we don't have time of use billing so the consumption rate is same any time of the day, but this scheduling can be used to control when your car is fully charged, similar to the car's built in charge timer. The convenience once again is that you can set this from anywhere via the app or web portal.

All in all, a "smart" EVSE is probably not essential, but for people who like to look at data, it's probably a great compliment to the telemetry data that you get with the SV/SL models!

Another LEAF! (kinda)

So after about 3 weeks or so, we got a very cute Tomica die-cast model of the LEAF from a Japanese seller on eBay:

Pros: Matches the car very well, as you'd expect from Japanese attention-to-detail

Cons: Right-hand drive, and kinda expensive for a die-cast at nearly $20 after shipping (it's almost double the price of the current 2018 model! :P)

The kid loves it!

Exploiting charging station for free parking (This is why we can't have nice things)

Circuit Electrique runs the biggest public charging network in Quebec. The company behind it, AddEnergie, may be better known by residential EVSE users elsewhere as the maker of the Flo residential units. They also operate the L3 DCFC stations. While there are very few free public stations around these parts, we are to have a good number of them!

Here the L2 stations are billed in two ways: most are $1/hr, billed by the minute. A few are $2.50 per session, which make it a great deal for a full top-up of larger battery vehicles like the Bolt. They are available both at free and metered locations. Some metered parking locations will have free parking for EV charging (but you pay for the charge).

During the two days of going back and forth to the hospital after the birth of our second daughter, we made use of a Circuit Electrique charging station that is conveniently located across the street from the Glen superhospital site. It is an area that is relatively busy, and otherwise you'd have to pay to park. However, the charging spot is free to park when charging.

Here I bumped into a SmartED, which appeared like this on both days:

Whats interesting is that the charger lights up as green and shows as "available", but there is actually a car plugged into it. The first time I didn't think much of it, but the next day I saw the owner leave the vehicle with the charge active (indicated by a white light), but 5 minutes later when I went past the light was green again. So this was no accident...

What I then realized was that this individual was actually exploiting a loophole in the billing procedure to get free parking! Here the stations charge by the minute (at $1/hr), and in places like this one, you get free street parking while charging. However, in this case there is a flaw where the billing stops when the car stops charging, and the station display shows the message "please return plug", but somehow the billing stops and the station availability clears up! I confirmed this as I was leaving the station by stopping the charge through hitting the release button on the J1772 plug, but did not actually unplug it. I left it for another 2 minutes, but the bill timer had already stopped and the station returned to "available" status. 

I wrote an email to AddEnergie and they said it is clearly a violation of parking regulations and a billable offence. However, I'm not sure if the meter attendants are aware of this particular detail, so folks like this one are slipping by. Eventually I expect such loopholes will be closed, but in the mean time, you might run into this kind of behaviour once in a while. (On the other hand, having chargers blocked by gasoline vehicles happens a LOT more often... :P)

New addition to the family

We are now officially a 2 child household! How well does the Leaf fit child seats? Better than our '99 Corolla (and much easier to get in and out due to the increased height of the vehicle). Its definitely not a huge car by North American standards, but being used to compacts most of our lives, the Leaf is actually very spacious inside. Here is a infant bucket seat next to a front facing toddler seat:

An adult can be squished in the back, either on the side, or even in the middle between the two seats. The limiting factor is the width of the carseat themselves. For the toddler we picked up a used Clek Foonf that is much slimmer than the EvenFlo (still in the Corolla), and it is much more space efficient while feeling a lot more solid too. The EvenFlo bucket infant seat is actually the wider one that takes up more space. It is possible to wedge in a (smaller) adult on the side, or even between the two car seats in the middle. Getting out is a bit of an issue in the latter case. We've since traveled with the full allotment of 3 adults and 2 kids (as well as the 4 adult 1 kid permutation). Definitely not ideal, but doable if needed.

Internet hearsay suggests its possible, with the right combination, to have THREE child seats in the back. I would not think of such a thing, for safety and many other reasons :P. At that point I would definitely pine for the passenger version of the e-NV200, which has yet to cross the pond...

Here's the Leaf parked outside the Glen super hospital site with a conveniently located public charging terminal. At $1/hr for charge and no parking fees, it beats the other pay parking spots around there!

As someone not accustomed to publicly advertising my family composition or vehicle contents, I reluctantly made this addition to the car after being tired of getting tailgated and honked at in the 30km/hr zones in our neighbourhood:

Time will tell how well it works...

Feb 2018: Grid to Road Efficiency. (Spoiler: its about 80%!)

After a few complete months of driving, I've gathered some interesting numbers for consumption. In this case, we're looking at the conversion from the electrical grid to what the battery provides for the motor. The ultimate figure in terms of distance can be easily found on the odometer.

To measure the electricity from the grid, we use the data logging features of the JuiceBox that tallies up all the charging sessions. Here's our charging log for the month showing a total of 161.1kWh:

Circuit Electrique, the only public charging network we've used so far, provides energy consumption info as well (despite billing by time only).

From the above two sources, we have a total of 182kWh of energy provided to the car from the grid, not counting the energy consumed by the EVSE to facilitate the charge. I expect this ancillary value to be negligible considering the simple electronics involved... the public station might be a bit more power hungry but as you can see above, we don't use them much.
What might also be interesting is we charged a total of 14 times (not counting the small blips shown in the JuiceNet chart above, which is probably from pre-heat after a night of charging. this value actually *slightly* overstates our energy input, but we can ignore it for this rough analysis... :P). This is basically once every 2 days, on average. I expect things will be better once the weather gets warmer.

On the car end, Nissan Connect/CarWings contains detailed info under the confusingly named "electric rate simulation" menu. Here we can find the net power output measured from the battery. The actual number is the total energy consumed by the car, minus the amount recovered through regeneration, and measured 146kWh.

Putting the two together, we find an overall efficiency of about 80% from grid input to battery output. This source measures L2 efficiency to around 90%. However, this is what the battery gets on the INPUT, after PDM conversion and charging overhead only. It doesn't account for how much the battery gains from that input power, and as we know, we always lose some due to heat. Additionally, when charging the car to a full 100%, after about 80% or so the charge rate slows down to balance the cells, and in this situation, the relative efficiency will go down further as the charging overheads stay the same. So all in all, a not-surprising value. Finally, one more obvious assumption we make here is that the roll-over effects at the beginning/end of the month are negligible.

One conclusion worth considering at this point is if you're on level 1 charging exclusively, you would see an even more significant loss. Because electricity is still so much cheaper than gas, its likely not a huge deal unless you drive a lot so from a purely financial perspective, you might not actually need to install a level 2 at home. On the flip side, if you drive a lot, you probably want to have a L2 charger at home to gain the mileage back at a faster rate anyways.

Putting Some More Numbers Together

Some other interesting numbers to keep track of, which I will try to do on a more regular basis:

Efficiency: 80% Average temperature: -5C
Kilometers driven: 835km
Grid to road km/kWh: 4.58 km/kWh
Battery to road km/kWh: 5.72km/kWh
Charging days: 14 (once every 2.0 days)

I will try to compile these values every month, as soon as the data is available.