ZOE Charging Curve (7kW)

ZOE Home Charging (Image: T. Larkum)
ZOE Home Charging (Image: T. Larkum)

The ZOE is unique among all electric cars on the market in its ability to fast charge with AC current, a feature of its patented Chameleon charging system. However, as with all battery charging, the charging rates don’t just depend on the power supplied but also vary with the level of charge so it can be difficult to predict exactly how long a particular charge will take.

In general there are three factors in play:

  1. As the battery fills up the charging rate will slow down. Some people have likened this to filling up a bucket of water so that it is as full as possible with nothing spilt. It gets harder to put charge in as the battery approaches being full, and it has to be put in more carefully.
  2. Related to the first issue, the charging power is typically reduced as the charge level increases. Anecdotal evidence shows that fast charging of a ZOE slows down at rates above 80% and this is presumed to indicate that a 43kW charge point will drop to 22kW (and perhaps again to 7kW) as the charging proceeds (this is something that testing will hopefully show in detail).
  3. When the battery is nearly fully charged the battery management system (BMS) will conduct ‘battery balancing’. Essentially the system will check the voltage of each cell that makes up the battery and adjust each individual cell’s state of charge (SOC) to even them out. This process can take some time, and it appears as though the charging has paused, though it will resume on completion.

To gain a better understanding of these charging processes it is useful to record charging rates for different amounts of power and levels of charge. To start this off I recently did it for my ZOE charging on its home charge point, the simplest scenario.

I began by running down the charge to a low level by driving it at speed along a local dual carriageway. Then I parked up on my drive in front of the garage where I usually charge (the cable is just long enough to reach when passed under the garage door) and plugged in. I recorded the charge level and the time at the start of the charge, and returned to the car every 5 minutes until it was charged, recording the charge level on each occasion (I used the timer on my mobile phone to remind me as each 5 minute period was up). I suspect the neighbours were wondering what was going on!

To cut a long story short, the charging ‘curve’ at 7kW turned out to be remarkably simple – essentially a straight line up to 99% SOC when battery balancing apparently took place (see Figure 1).

Figure 1: ZOE 7kW Charge Curve (Image: T. Larkum)
Figure 1: ZOE 7kW Charge Curve (Image: T. Larkum)

The balancing seemed to take a remarkably long time (about 2.5 hours, after 3.5 hours of charging) which may reflect the fact that the car was outside in cold weather. Similarly a total time of 6 hours is unusually long; usually it takes less than 4 (at least when it’s in the garage).

While it was charging I had the idea to see how accurate its predicted end time was, and I started recorded its ‘time remaining’ figure from about 40 minutes in. Bearing in mind that it was 99% charged after 205 minutes (and 100% after 355) it’s interesting to see that the prediction of 220 minutes it gave for most of the time was about right (see Figure 2). This is if one considers the duration to be the time to 99% ‘plus a bit’, where the ‘bit’ allows for cell balancing but is probably not an easily predicted amount. Conversely, however, for the first quarter or so of charging time its prediction was very poor, seemingly increasing at first (though I don’t have much data for the earliest period) then dropping towards a better estimate.

Figure 2: ZOE Charge Completion Prediction (Image: T. Larkum)
Figure 2: ZOE Charge Completion Prediction (Image: T. Larkum)

Most EV owners will be careful to allow time for the cell balancing phase to extend the life of the battery. However, I have often skipped this stage when short of time (e.g. when fast charging at motorway services) since the battery isn’t mine. I suspect Renault didn’t think of that when they chose to lease rather than sell the battery to drivers. Whether it matters very much long term we may not know for some years.

Anyway, with regard to home charging we can say:

  1. At 7kW the charging is very consistent and essentially linear up to 99%.
  2. At 99% cell balancing occurs, and its duration appears to be unpredictable (and presumably depends on many factors including temperature, and time since last cell balancing).
  3. Unfortunately, ZOE’s display of the predicted time to charge (even to 99%) is largely useless in the early stages.
  4. Fortunately, it is easily calculated by the driver. The rate of charge is about half a percent (actually 0.475%) per minute to 99%.

In summary, as a rule of thumb:

At 7kW a Renault ZOE charges at about 4 miles (6km) every 10 minutes.

 

Home Forums ZOE Charging Curve (7kW)

Tagged: 

This topic contains 4 replies, has 3 voices, and was last updated by  timbo 5 years, 4 months ago.

Viewing 5 posts - 1 through 5 (of 5 total)
  • Author
    Posts
  • #7465

    Trevor Larkum
    Keymaster

    The ZOE is unique among all electric cars on the market in its ability to fast charge with AC current, a feature of its patented Chameleon charging sy
    [See the full post at: ZOE Charging Curve (7kW)]

    #7479

    donald
    Participant

    I would doubt that so much time at the end of the charge cycle is cell balancing.

    Li charging for all chemistry types runs along a typical recipe:

    First stage is that a constant current is applied to the battery. The battery, being a load, will cause the charging supply to be ‘pulled down’ to some voltage. As that fixed current is applied to the battery, the voltage will rise as the cell charges and the cell voltages rise. At some point, that voltage hits a level corresponding to, perhaps, 4V per cell (or whatever proprietary level Renault have chosen as an engineering margin for protecting the cells).

    Li cells are tricky to know what SOC they are at. It is probable, IMHO, that when the charge level hits 99% this is not 99% SOC but is the end of the ‘constant current’ phase of charging. This is simply because the BMS can’t tell what SOC the cells are at, at the end of that constant current phase.

    So the second stage then begins which is a constant voltage phase. In this phase the mode of charging changes and that 4V per cell (whatever the algorithm picks) is held steady. Because the cells are still charging, the current will now begin to drop. The constant voltage phase continues until the drawn current drops to 10% (or, again, whatever the algorithm is set to) of the constant current phase.

    Throughout this constant voltage phase, the BMS doesn’t really know the SOC because the cell voltage is being dictated by the applied power from the charger. There’s no way of knowing the cell voltage, except by turning off the charging and allowing the battery to rest for a while until the open-circuit cell voltage can be measured. This would take too long to keen drivers wanting to drive off, so I am fairly certain that the long dwell at ‘99%’ is simply the constant voltage phase (sometimes called the ‘saturation phase’).

    Interestingly, the slower you charge in the constant current phase, the shorter the saturation phase is. At 0.1C charge rate (which means a charge rate that takes 10 hrs of charging, e.g. 2.2kW for a 22kWh battery) there is almost no saturation phase. The faster you charge, the longer the saturation phase. There is a point where there is no advantage to trying a faster charge at constant current if you are aiming for 100% saturation charge as it simply extends the saturation charge period.

    If you were to plot the drawn current during charging, you should probably find that at 99% the 30A charge current begins to drop, and will fluctuate downwards in a series of ups-and-downs as the BMS monitors whereabouts the SOC is. I expect it will probably terminate charging once it drops down to under 5ish Amps.

    Cell levelling shouldn’t happen with every charge. If it does, then the battery is towards the end of its life, or at least some of the cells are. Cell levelling is wasteful and generates heat as the ‘high’ cells are shunted and the ‘low’ cells are differentially charged. Only if it is a very unbalanced battery pack will this need to go on for a long while. I have only observed it twice with the Fluence, once when I first charged the car at 3.5kW, and again some months later. Of course, you may well have caught an occasion when it was doing some cell levelling, but if it does it on every charge then, AFAIK, that would suggest the battery pack needs attention.

    #7480

    timbo
    Participant

    Donald
    Thanks for this explanation. I have a plot of drawn current during charging from a recent overnight charge on my home 7kW charger. As I constantly monitor electricity usage and generation on my home pc, I know the shape of this graph is typical of an overnight charge. The only variations being the length of time of the constant current and the length of time of the constant voltage stage (it seems to take longer on this part in cold weather).

    The two stages are pretty clear:
    – constant current for about 3 hours
    – then constant voltage (reducing current) which sees the rate dropping from 7kW/hr to c. 2.5kW/hr over the course of about an hour.

    Attachments:
    You must be logged in to view attached files.
    #7482

    Trevor Larkum
    Keymaster

    donald, thanks for the insight. timbo, what system are you using to monitor your power usage? I have a Wattson system for my solar generation and home usage, but I find it pretty un-user-friendly. I tried monitoring it during this test but since it was done during the evening (I wasn’t going to get out of bed every 5 minutes!) there were lots of other things turning on and off during the same period (washing machine, tumble dryer, kettle, etc.) so I couldn’t come to any conclusions.

    #7521

    timbo
    Participant

    Trevor, I use pvoutput (www.pvoutput.org). This is a great site that allows you to upload consumption and generation data every 5 mins 24/7. Once you install the software and set it up to talk to your electricity monitor, you’re away. It’s compatible with quite a few different types of monitor. I use the simple CurrentCost Envi monitor (http://www.currentcost.com/product-cc128.html) with cable clamps around the relevant solar pv and consumption cables. The monitor is plugged into my (always on) pc and it’s so good that I never look at the meter itself.
    If you wanted to monitor usage of your car charger, you’d simply need another clamp around the relevant cable. I haven’t bothered with this as I have so little else on during the night such that when I charge the 7kw/hr drawn by the car swamps the small additional usage.
    I’ll email you a link to my webpage so you can see how versatile it is – copes with peak/off peak tariffs, costs etc.

Viewing 5 posts - 1 through 5 (of 5 total)

You must be logged in to reply to this topic.