How accurate are the "energy economy" figures?

[evnow]

Fifteen miles after the low battery warning without even losing bar 1??

Is this with the old firmware ?

[planet4ever]

No, the new firmware. Note my delivery date on the left.

Ray

[DaveEV]

I am coming to the conclusion that everything -- except the three odometers -- is subject to random voltage variations, or other non-linear analog measurements. I've driven the car pretty hard today, starting with a 100% charge and traveling 87.1 miles, mostly freeway, more than a third of it at 65-70 mph, almost all of it with A/C on. That's not too shabby, and I must say I am pleasantly surprised. Oh, my energy economy for the day? Only 4.0.

Hmmm, so 87 miles and 4.0 mi / kWh? My 82 mile trip today the dash read 4.3 mi / kWh so in theory I should have been able to drive 93 miles - but made it home after 2 hours of trickle charging with 2 bars and 13 miles remaining...

This was actually the first time I've charged to 100% since I got the car 2 weeks ago - now have a bit more than 500 miles on the odometer but have to wonder if the battery or car will "break-in" more as some have suggested that the range seems to get better with some time. I know that new tires tend to be less efficient for at least a thousand miles or so before the loosen up.

My mi / kWh from the wall is very close to ttweed's - right around 3.6. I pretty consistently get 4-4.2 mi / kWh on the dash. Carwings currently indicates 5.5 mi / kWh over 464 miles.

[evnow]

No, the new firmware. Note my delivery date on the left.

Ray

That is strange - and also the fact that your miles / kwh was low. Was your m/kwh changing in the last few miles ?

I got a total of some 18 miles from battery low to turtle. But I lost the last bar at the same time I got the battery low.

[planet4ever]

No, the new firmware. Note my delivery date on the left.

That is strange - and also the fact that your miles / kwh was low. Was your m/kwh changing in the last few miles ?

I got a total of some 18 miles from battery low to turtle. But I lost the last bar at the same time I got the battery low.

The last mile was in town, mostly 35 mph, but the 10 miles before that was a four-lane country highway (US 101 many years ago) that I mostly drove at 40 mph, some 45. It looks flat, but actually rises 120 ft very gradually. I think m/kWh was 4.1 when I started that section, but I'm not sure. I seem to remember seeing 4.1 somewhere along there. It may have been 4.0, or even 3.9 when I got the battery low warning.

My car's "normal" behavior (based on a small sample - perhaps four times) is to give low battery somewhere in the middle of the last bar, with the bar disappearing 2-5 miles later. Something definitely seems wrong for the bar to hang on for another 15 miles (at least, since it never disappeared). It also seems odd to go 15 miles after 'low battery' without hitting 'very low', though I guess my very cautious driving during that time might explain that. The bottom line is I have to believe that either 'low battery' was signaled early or 'zero bars' wasn't signaled when it should have been, or both.

Ray

[aqn]

I've been keeping record of, among other things, energy economy figures (miles/kWh, from LCD display and from center display), tripmeter miles, and starting & ending SOC bars. From those figures, I calculate kWh's per SOC bar: (tripmeter miles / energy economy) / (staring SOC bars - ending SOC bars). So far, over 16 "tankfuls", leaving out the two largest and two smallest values and averaging the rest, the average is 1.44 kWh per SOC bar. The big unknown here, as this thread's title suggests, is the accuracy of the car's energy economy figures.

It would be great if more than one person collect this data; it'd be interesting to compare notes.

Note 1: it only took a few "tankfuls" to converge on the figure of 1.44 kWh/SOC bar: 12 "tankfuls", ignoring four largest/smallest values, averaging 8 values.)

Note 2: 1.44 kWh/SOC bar is different than the 1.7 kWh/bar (Boomer23) and 1.6 kWh/bar (TonyWilliams) arrived at via reverse engineering some CAN bus data. Why? I dunno, but we will need to see if their results converge on a different value if/when they collect more data.

[aqn]

I've been keeping record of, among other things, energy economy figures (miles/kWh, from LCD display and from center display), tripmeter miles, and starting & ending SOC bars. From those figures, I calculate kWh's per SOC bar: (tripmeter miles / energy economy) / (staring SOC bars - ending SOC bars). So far, over 16 "tankfuls", leaving out the two largest and two smallest values and averaging the rest, the average is 1.44 kWh per SOC bar.

It's in no way rocket science, but just in case someone is interested: I've uploaded my spreadsheet to a Google doc.

The columns are in the same order as the LEAF's displays, left to right. I go left to right copying down the numbers (and resetting them where possible) so as to not forget anything, since there are eight numbers/time. Every time I charge and enter them into the spreadsheet later.

The columns are:
_ start - guess-o-meter miles
_ start - bars *
_ mi/kWh LCD display *
_ average speed
_ odometer
_ tripmeter *
_ end - guess-o-meter miles
_ end - bars *
_ mi/kWh center display *
_ time charging start *
_ charge to 100%? *
_ time charging end *
_ notes

The pink shaded columns (marked with * in above list) are necessary for calculations
The unshaded columns are for record keeping only.
The blue shaded columns on the right are calculated from the pink shaded columns on the left:

_ charging time
_ charge time per bar (minutes)
_ miles/ bar
_ % diff bet. LCD disp and center disp
_ kWh/bar, according to LCD disp
_ kWh/bar, according to center disp