Bolt miles/kWh

Discussion in 'Bolt EV' started by Kevin C, Jun 1, 2018.

  1. Kevin C

    Kevin C New Member

    What kind of mileage are Bolt owners getting per kWh? I know, so many existential variables. From near hypermiling to lead foot mania.
    But on average and seasonably adjusted, what are you finding? Also, from those who drove a Spark and a Bolt, is it a night and day difference? (Looking at you Wade T.)
    Thanks in advance to those who care to share.
     
  2. rgmichel

    rgmichel Member

    From October 2017 to the May 2018, I achieved a median of 3.82 miles/kWh as read from the car's display. Reading from my ChargePoint charger, which reads the actual power used to charge the car, and accounts for charging losses, I achieve a median of 3.6 miles/kWh. I don't have enough records to cover all my summer 2017 driving, but it is likely a bit better. I am usually over 4.1 miles/kwh read from the car now. I am a fairly careful driver, but I do enjoy seeing the smallest possible images of cars in my rear view mirror every now and again.
    Of course the efficiency of the car is pretty fixed, and due to the range of the beautiful Bolt EV is hardly ever necessary to worry about trying to obtain the highest efficiency. Even in long distance driving, there are enough chargers, in the NE, prevent hyperventilation.
     
    Kevin C likes this.
  3. SWAMPLIFE

    SWAMPLIFE New Member

    4.1
    NOTHING WRONG
    NO COMPLAINTS
     
    Kevin C likes this.
  4. NeilBlanchard

    NeilBlanchard Active Member

    We are averaging 5.2 miles per kWh, on the dash display.
     
    Kevin C likes this.
  5. rgmichel

    rgmichel Member

    An instantaneous dash display is just a crude average that reflects only the last few journeys and has no real relevance to the efficiency of the car, which is generally around 4 m/kWh. As an example of this, the recent software upgrade messed with my dash display to about 2 m/kWh. This came back up to 4 m/kWh within a few charges. If the dash display can be manipulated this easily then it is a simple average and has no real meaning. So, if you are a very careful driver, then an easy drive can keep the average higher than other people rather easily. However, if you are a normal driver using heat or a/c and live on a hill that you always have to climb to reach home, then your are back at 4 m/kWh. The only statistic that really counts is keeping a long record of all journeys, to come up with an average that represents your driving only. Indeed, a better number would be to calculate the median of your data, rather than an average, as this will disregard most of the outliers.
     
    Kevin C likes this.
  6. NeilBlanchard

    NeilBlanchard Active Member

    That is the AVERAGE for the last ~1350 miles. I've been driving EV's for several years now. I have measured the charge, with our JuiceBox Pro 40, and that is always a bit more - I checked after a 197 mile trip recently, and the charge was about 8% greater than the dash display. The average at that point was 5.3 miles / kWh.

    On our e-Golf, the average is 5.6 miles / kWh. The e-Golf coasts by default, which is great. I shift the Bolt EV into neutral to coast, when possible, and this helps raise the average.
     
    Kevin C likes this.
  7. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    Yup. An accurate energy accounting would require you to have an EVSE which measures kWh drawn from the wall when charging the car, and you'd have to correlate that to the odometer reading on the car. That would give you the car's MPGe, which includes charging losses. There's not really any practical way for the average person to calculate actual miles/kWh without including charging losses. That would require measuring the actual state of charge of the battery pack, which is a task that challenges even experts.

    As Rgmichel said or implied, the dashboard readout of a plug-in EV isn't accurate enough to be a reliable measure of the amount of energy used per mile. I haven't seen any discussion saying the Bolt EV's dashboard display is wildly off, as happens in the Leaf and the Clarity, but even still, it's unlikely to be accurate enough to be relied upon.

     
    Kevin C likes this.
  8. rgmichel

    rgmichel Member

    Its not a matter of accuracy. I am sure the dashboard is accurate, but it measures an average. What you need are long term data, and a calculation of the median not the average, as I said above. Accuracy is quite another matter and should not be confused with the concepts of "average" or "median". If you use the word "accuracy" you are saying the numbers are "wrong" or "right". I am quite certain the number on the dash gives you an accurate estimate of the average right from when the car is new. Mine presently reads 4m/kWh over 11k miles.

    The ChargePoint charger that I use includes charging losses and I repeat what I said above about my winter driving:
    From October 2017 to the May 2018, I achieved a median of 3.82 miles/kWh as read from the car's display. Reading from my ChargePoint charger, which reads the actual power used to charge the car, and accounts for charging losses, I achieve a median of 3.6 miles/kWh.
    When I say "as read from the car's display", I mean I read the display before EVERY charge and entered all numbers into a spreadsheet so I could calculate average, and median energy used as reported by the car. Charging power was calculated from the kWh used at each of every charging session reported by my ChargePoint charger and entered into a spreadsheet. This allowed me to calculate average and median energy charged, while the difference between the dashboard data and the ChargePoint data reflects the charging loss.
     
    Last edited: Jun 5, 2018
    ChargePoint and Kevin C like this.
  9. Pushmi-Pullyu

    Pushmi-Pullyu Well-Known Member

    Well, if LG Electronics and/or GM has engineered the Bolt EV to actually report energy usage and range estimates from the instrument panel display accurately, then they certainly deserve kudos! Elsewhere on this forum you can read about wildly varying range estimates reported from the instrument panel readout by several Clarity drivers, and comments on other forums about the Leaf's so-called "guess-o-meter" certainly don't lend any confidence either!
     
    Last edited: Jun 5, 2018
  10. rgmichel

    rgmichel Member

    The calculation reported on the dash is very simple. It computes the number of miles driven and the kWh used over some interval, which I suspect is over 5 mile intervals, and averages these numbers over time. This is accurate, but it is not the useful number that people think it is. You have to think about what an average means, and how it will behave over time. Take a look at the bar graph in the energy dashboard, and see how the 5 mile intervals behave over time. They are very variable due to the nature of the simple "average" calculation. What people really want is a more sophisticated calculation, which is why I collect detailed data before and after every charge to find more meaningful numbers. The median is one good measure that is more meaningful than the average for the way you drive, while the charger's readout of power actually absorbed by the car is a more meaningful way of looking at the efficiency of the car; again the median is a better measure than the average. Having said this, if you keep your average going right from new, the m/kWh over many thousands of miles is going to be a good indicator of the efficiency of the car and the way you drive. However, it is not the true efficiency of the car because of the charging loss. If you think that the dash readout out is not accurate because it is not the true efficiency then that is true.
     
  11. NeilBlanchard

    NeilBlanchard Active Member

    Listen - it is the long term average. The displays are actually quite accurate, and only are not "accurate" when you change how you are driving.
     
  12. rgmichel

    rgmichel Member

    The display is accurate at calculating the average no matter how you drive. You can't have it both ways. The display is always accurate. Its variable because an average is not a very good way of answering ALL the questions that a driver may have. There is no equivalence between accuracy and variability. They are quite separate things and should not be confused.
     
  13. NeilBlanchard

    NeilBlanchard Active Member

    Understood - the whole thing about it being a guess-o-meter is because folks expect it to predict range - before they head up a mountain, or when they turn on the heat, or start driving 75 when they had been driving 35.

    That's why I put the quotes around accurate.
     
  14. rgmichel

    rgmichel Member

    My understanding of the prediction of range is the car uses a shorter term average to come up with the range numbers at the left side of the speedometer. The number we have been talking about above is the one in the center of the display when you select the vehicle information menu, which I have always assumed is a straight long term average. I actually record the middle "range left" number whenever I record my charging information. One day I will try to draw some conclusions from those data too. Its funny really that the GM engineers understand all these things completely, but we have to try understand by studying it all after the fact.
     
  15. NeilBlanchard

    NeilBlanchard Active Member

    Right, I am going by the average miles / kWh in the center of the display.
     
  16. rgmichel

    rgmichel Member

    Some miles/kWh data from my charging history, between 6463 and 11466 miles. The data were read from the middle display just before charging the car and before each charge. The graph does not account for charging losses, and represents the variability of day to day driving, such as use of heater, and A/C, outside temp and so on. The data run from Oct. 2017 to June 2018, so the earlier part of the graph, after 6000 miles, shows the climb down into winter, while the higher odometer readings show the climb back out into summer. The dip mostly represents use of the heater in the winter. The two big dips at 10k and 11k represent the software update done by GM Plot of m per kWh 7 June 2018.png Plot of m per kWh 7 June 2018.png
     

Share This Page