How you use Carwings, and why you should, if you don't

[edatoakrun]

Back in August 2011, after I noticed that the update had made Carwings both useful and reliable, I posted the results in this thread.:

http://www.mynissanleaf.com/viewtopic.php?f=31&t=5423&hilit=+carwings" onclick="window.open(this.href);return false;

Since then, many have posted results showing how valuable CW can be, but a lot of “old-timers” who rejected CW after their initial look at it, still don’t seem to believe it is useful. So, I’m posting one example of how I use it, and hope others will make further examples, and suggestions.

To illustrate how useful Carwings is, to those who remain proudly ignorant, and also to show the limits of the GOM, I plan to repeat the the test run below, on the route I mentioned earlier, at different initial charge states.

I'm planning on 3 tests, with “80%” by timer, 100% charge, and the optimum for my purposes, the approximate charge given by 30 minutes of L2, following a 80% timer charge, which is shown as”92%” on the owners site, and 11 bars on the dash display. Those are the test results are shown in the table below.

When I first got my LEAF, I quickly discovered that, for my most common trip, I would get almost the same energy recovery from the approximately 2,000 ft. of descent, about 1500 net, by charging to a point beyond 80%. This was long before CW had become both accurate and reliable. My “test” was to simply stop after the large initial descent, about 1400 ft. net in the first 7 miles, and touch the discs, following charging to different levels. 100% charge = Ouch!, 80% warm disks, and 90%, just a bit warmer.

I generally charged only to 80%, but when winter came, with a little shorter range, my 8-9 bar “commute” often began to take 10 or 11 bars, so, on rainy days or when I had more errands planned, I got into the habit of starting charging before I left home, just to have more detour options, on the 50-60 mile trip.

I think CW will give me a much better understanding of the optimum charge level, and the larger question, the ratio of descent/ascent energy recovery, than the disc brake heat test, I did a year ago.

So, on the 30th, I charged to 80% by timer, then added 30 minutes of 16 amp L2. Temp was in the low 40’s F while charging, and in the low 50’s while driving. No heater use, no rain. The cars temp bar display started at four, and the fifth one came on just before the conclusion of trip # 4, below. I stopped at the 10-mile from-my–home point, at the beginning and end of this trip, to rest the CW trip report, as shown below on trips 1 and 5. As you can see, I made two stops in town, separated by trip #3.

I plan to repeat it at 80% charge, and 100%, and will post the results.

All reports below are from Carwings, except the GOM numbers, which I have added , for your amusement.

Mar/30

Total kWh use: consum. Regen distance Energy Econ GOM-47 miles at start-up

Trip 1: 0.4kWh 1.1kWh 0.7kWh 9.8miles 24.8miles/kWh 104 miles shown at end

Trip 2: 3.1kWh 3.6kWh 0.6kWh 14.9miles 4.8miles/kWh

Trip 3: 0.1kWh 0.1kWh 0.0kWh 0.4miles 4.4miles/kWh

Trip 4: 3.2kWh 3.8kWh 0.7kWh 15.2miles 4.8miles/kWh 34 miles shown at end

Trip 5: 4.2kWh 4.3kWh 0.1kWh 9.8miles 2.3miles/kWh 7 miles shown at end

Distance by Carwings total: 50.1 miles (2.5% constant under-reported)

Distance by odometer total: 51.4 miles

10:37 PS

I should add, that I have NOT had any updates since last August, So the erratic GOM behavior above, will, presumably become even more erratic, after I go in for the "notification" update, if I understand the "new and improved" GOM results, that others have posted, correctly.

 

[Ingineer]

To illustrate how useful Carwings is, to those who remain proudly ignorant, and also to show the limits of the GOM, I plan to repeat the the test run below, on the route I mentioned earlier, at different inital charge states.

I'm planning on 3 tests, with “80%” by timer, 100% charge, and the optimum for my purposes, the approximate charge given by 30 minutes of L2, following a 80% timer charge, which is shown as”92%” on the owners site, and 11 bars on the dash display.

FYI, a "full" charge is typically around 94-95% SoC, while an 80% charge is actually to 80% SoC. My last "full" charge stopped at just over 94% SoC which was 278 "Gids" or 22240 watt-hours. Also keep in mind, the battery contactor opens at around 2% SoC, so you only get about 92% usable per full charge. I've calculated that 100% SoC (if you could reach it) would be 23512 watt-hours. (on my car anyway) Knowing the top ~5% (1176wH) and the bottom ~2% (470wH) can't really be used, that leaves 21866wH actual usable from "full" to turtle, and 18340wH usable from 80% to turtle. Since there is a significantly wide variation from one "full" charge to the next, it's not an accurate test to charge to full and expect repeatable results. It would be more accurate to charge to 80%.

-Phil

 

[DeaneG]

Sounds great, but I just drive the car rather than the website ("proudly ignorant"). If there aren't a comfortable number of remaining miles showing on the Leaf (1.5 or 2x the trip distance), we take the gas car. Destination flexibility is more fun than watching the GOM.

 

[edatoakrun]

To illustrate how useful Carwings is, to those who remain proudly ignorant, and also to show the limits of the GOM, I plan to repeat the the test run below, on the route I mentioned earlier, at different inital charge states.

I'm planning on 3 tests, with “80%” by timer, 100% charge, and the optimum for my purposes, the approximate charge given by 30 minutes of L2, following a 80% timer charge, which is shown as”92%” on the owners site, and 11 bars on the dash display.

FYI, a "full" charge is typically around 94-95% SoC, while an 80% charge is actually to 80% SoC. My last "full" charge stopped at just over 94% SoC which was 278 "Gids" or 22240 watt-hours. Also keep in mind, the battery contactor opens at around 2% SoC, so you only get about 92% usable per full charge. I've calculated that 100% SoC (if you could reach it) would be 23512 watt-hours. (on my car anyway) Knowing the top ~5% (1176wH) and the bottom ~2% (470wH) can't really be used, that leaves 21866wH actual usable from "full" to turtle, and 18340wH usable from 80% to turtle. Since there is a significantly wide variation from one "full" charge to the next, it's not an accurate test to charge to full and expect repeatable results. It would be more accurate to charge to 80%.

-Phil

Thanks.

I can look at CW reports of ABC from 80%, but:

I can't conveniently take my car to "turtle", due to steep hills, no alternative charge locations, and limited cell phone coverage, near my home. What do you consider ABC is, in kWh, from 80% to LBW, and VLBW? And could you also please correlate this number to a testable temperature?

As you may have noticed, CW seems to give me slightly lower ABC that your figures. my "hottest while charging" battery test, using the VLBW + 8% to dead (not turtle) as my estimate method, only showed about 20.4 kWh, but that was with a highly uncertain 60F (+ or - 10F?) battery temp.

Since August , I have been hoping other would test CW kWh ABC to a lower, more accurate battery capacity level.

And given your comments, it would probably be best for them to check CW, for calibration to your measurements, in tests from 80%, rather than 100% charge, to the lowest level they can take it to.

I don't doubt that there might be a error in CW ABC, but there seems to be the exactly same corresponding error in m/kWh reported both by CW, and the 2 different dash/screen reports from the car of m/kWh, which I do not use, but that that others do.

So, for me, CW reports have provided a reliable and accurate basis, when adjusted correctly for variations in future use, in calculating future range, since last August.

 

[edatoakrun]

Sounds great, but I just drive the car rather than the website ("proudly ignorant"). If there aren't a comfortable number of remaining miles showing on the Leaf (1.5 or 2x the trip distance), we take the gas car. Destination flexibility is more fun than watching the GOM.

If you want to drive an ICEV when you don't need to, fine.

If you take a look at my post above, I'd never be able to make that trip, in my LEAF, without lining up a recharge at the halfway point, using your method.

And that 50 mile trip, is pretty much the minimum I ever drive, in my LEAF, so I'd wind up not driving it, at all..

And my advice, is never take the GOM too seriously.

Sooner or later, you will get an erroneous estimate, even using your "1.5 or 2x the trip distance" fudge factor, that will leave you stranded, on the way up a big hill...

 

[Ingineer]

Thanks.

I can look at CW reports of ABC from 80%, but:

I can't conveniently take my car to "turtle", due to steep hills, no alternative charge locations, and limited cell phone coverage, near my home. What do you consider ABC is, in kWh, from 80% to LBW, and VLBW? And could you also please correlate this number to a testable temperature?

As you may have noticed, CW seems to give me slightly lower ABC that your figures. my "hottest while charging" battery test, using the VLBW + 8% to dead (not turtle) as my estimate method, only showed about 20.4 kWh, but that was with a highly uncertain 60F (+ or - 10F?) battery temp.

Since August , I have been hoping other would test CW kWh ABC to a lower, more accurate battery capacity level.

And given your comments, it would probably be best for them to check CW, for calibration to your measurements, in tests from 80%, rather than 100% charge, to the lowest level they can take it to.

I don't doubt that there might be a error in CW ABC, but there seems to be the exactly same corresponding error in m/kWh reported both by CW, and the 2 different dash/screen reports from the car of m/kWh, which I do not use, but that that others do.

So, for me, CW reports have provided a reliable and accurate basis, when adjusted correctly for variations in future use, in calculating future range, since last August.

ABC? Automatic Battery Cutoff?

-Phil

 

[edatoakrun]

ABC? Automatic Battery Cutoff?

-Phil

Available battery capacity.

I mean, the amount, in kWh out of the total battery capacity, that Nissan allows drivers to use. Use another term, if you have a better one.

 

[TonyWilliams]

...remaining miles showing on the Leaf (1.5 or 2x the trip distance), we take the gas car. Destination flexibility is more fun than watching the GOM.

I don't watch the GoM at all, therefore I must have the most flexibility!

:shock:

 

[TomT]

I simply use Gary's SOC meter and I don't worry about the GOM or Carwings...

 

[ebill3]

Proudly ignorant using the Gary's GID Meter and Tony's range chart.

Bill

 

[TonyWilliams]

Proudly ignorant using the Gary's GID Meter and Tony's range chart.

Bill

Welcome to the club!!! Can't wait for LEAFscan!

 

[edatoakrun]

I just repeated the test from the first page of the thread, at 80% charge. A Bit warmer today, while charging and driving, so I expect CW will post a slightly higher m/kWh, reflecting the LBW that I got, just before reaching my carport.

So, if Phil, or anyone else, can post a Gid-projected battery capacity from a 80% charge (at about 45 F) to LBW, I will post CW kWh results, as soon as I get them.

BTW, this AM I said:

...Since August , I have been hoping other would test CW kWh ABC to a lower, more accurate battery capacity level.

And given your comments, it would probably be best for them to check CW, for calibration to your measurements, in tests from 80%, rather than 100% charge, to the lowest level they can take it to.

I don't doubt that there might be a error in CW ABC, but there seems to be the exactly same corresponding error in m/kWh reported both by CW, and the 2 different dash/screen reports from the car of m/kWh, which I do not use, but that that others do...

There is also the possibility, I would think, that both a gid-count battery capacity, and the lower capacities reported by CW and dash miles, might both be "right".

If I understand correctly, the gid count, and Phil's kWh numbers using those values, represents stored energy.

There is an unknown inefficiency in discharge.

The CW reports, as well as the dash/nav screen m/kWh X miles driven reports others have posted, which mostly appear to be lower than Phil's kWh numbers, could be reflecting total kWh, at some, point past the battery.

The kWh used to heat up the battery during discharge, must coming from somewhere, right?

 

[planet4ever]

that leaves 21866wH actual usable from "full" to turtle, and 18340wH usable from 80% to turtle.

Wow! This is significant news that I had not heard before, and never suspected. I had always assumed that "80%" meant 80% of usable capacity, and scoffed at the people who said it was really 83%, just because Carwings calculated 10 bars out of 12 = 0.833.

Now you are telling us that "80%" is really 18340/21866 = 0.839, or 84% of usable capacity! Please pass the crow.

This also means we have some 'splaining to do about Tony's chart and ten bars vs. 80% charge. I thought we had all agreed that the tenth bar wasn't full after an 80% charge. Yet the chart says it is full at 84% of usable capacity.

Ray

 

[TonyWilliams]

that leaves 21866wH actual usable from "full" to turtle, and 18340wH usable from 80% to turtle.

Wow! This is significant news that I had not heard before, and never suspected. I had always assumed that "80%" meant 80% of usable capacity, and scoffed at the people who said it was really 83%, just because Carwings calculated 10 bars out of 12 = 0.833.

Now you are telling us that "80%" is really 18340/21866 = 0.839, or 84% of usable capacity! Please pass the crow.

This also means we have some 'splaining to do about Tony's chart and ten bars vs. 80% charge. I thought we had all agreed that the tenth bar wasn't full after an 80% charge. Yet the chart says it is full at 84% of usable capacity.

Ray

Well.... Let me first say that the data in my chart is based on (largely) observed performance. Much like Nissan publishes the battery is 24kWh, our empirical data reflects something closer to 21kWh available to us, plus or minus whatever corrections for temperature, degradation, cell imbalances, el al.

The same is true of Gids. The range chart is not based on a 80 value per Gid, but instead is based on whatever value is produced from measured performance to the operator. Reverse engineering, of you will.

So, when our simple observations reveal that 281 Gids is present at 100% charge, as is customary when "all is well", and we drive at a 4 mile/kWh rate for 84 miles, where the car comes to a halt with a depleted battery, then we can divide 4 into 84 and get 21kWh.

Therefore, a simple 21 divided by 281 results in 74.7331 watt/hours observed for our use, with the available tools. That is what the chart is based on, and 80 watt/hours does not fit into the model.

I think you're mixing data points to arrive at the 84% available, for an 80% charge, since it very clearly does stop at 80% Gid, or about 225 on a good battery.

How about we end the back and forth on the issue with Phil correctly stating the Nissan published data EXACTLY as they claim, 80 watt/hours, and a Gid will be the same piece of data that we use with a 75 value. Everybody wins.

Finally, the 80% - ten bar issue. Normally, it's not full since the tenth bar would be 78% to 84%. However, even in normal circumstances, temperatures change, and I've had 11 bars and 9 bars at 80% charge.

Everybody wins.

 

[edatoakrun]

Proudly ignorant using the Gary's GID Meter and Tony's range chart.

Bill

Well, the gid meter alone won't allow you calculate future range, and the OP should tip you off to why Tony's range chart, IMO, still miss-states the ascent/descent effects on range, preventing accurate range estimates, on drives with large changes in elevation.

I got the figures for the 80% charge to compare to the OP, and will try to put them in a easier-to-read format, but you really should look at your own CW, if you haven't yet, to see all how all the reports are correlated.

Meanwhile, here a a few CW features you should check out:

Complete accounts of energy efficiency for each day of the month. Check you odometer. My CW under-reports
"miles driven" by about 2.5%, and some other LEAF drivers have reported the same thing. More on this later.

It was a January-like March, and, as you can see, I didn't get out much. All the trips shown were to different valley locations, including the one on the 30th, the "electric rate simulation" for which, including energy use by trip segments, is on this thread's OP.

Similar driving leads to similar daily m/kWh. The only days I got below 4 m/kWh, were days with lots of heater/defroster use, or more than a few freeway miles.

The "regional rankings" page reports your M/kWh by monthly total, since delivery.

In my case, you can see the correction that occurred with the update at the beginning of August. I t sure looks like the earlier 3 months may be correctable by a common coefficient, but I never have gotten around to trying it.

Month and Year Grade Rank Energy Economy
May/2011 Gold 492 5.4 miles/kWh
Jun/2011 Gold 695 5.7 miles/kWh
Jul/2011 Gold 1651 5.3 miles/kWh
Aug/2011 Bronze 3622 4.2 miles/kWh
Sep/2011 Silver 3457 4.5 miles/kWh
Oct/2011 Silver 3640 4.2 miles/kWh
Nov/2011 Gold 2792 4.2 miles/kWh
Dec/2011 Gold 2827 4.0 miles/kWh
Jan/2012 Gold 2547 4.2 miles/kWh
Feb/2012 Gold 2354 4.4 miles/kWh
Mar/2012 Gold 2975 4.3 miles/kWh
Apr/2012 Gold 2982 4.3 miles/kWh

I strongly suspect all my CW m/kWh reports are off by the same percentage as miles driven, but 2.5% is just too small an error, for me to find, by looking at total battery capacity and range. Anyone else figured that out?

My Dash has shown 4.3 m/kWh , and my NAV screen 4.4 m/kWh, for all miles since delivery, since settling there, on the drive down from Mt Lassen, last July.

These calculations are for all miles since delivery. Some have reported (never seen it myself, I never reset) the daily CW reports and dash daily readings are identical.

It would be interesting, if it turned out that a 2.5% corrected m/kWh from CW corresponds to the Nav screen m/kWh.

Anyone checked that out?

How do those who calculate battery pack capacity by odometer and the cars m/kWh, choose one screen over the other?

 

[edatoakrun]

I just repeated the test from the first page of the thread, at 80% charge...

So, if Phil, or anyone else, can post a Gid-projected battery capacity from a 80% charge (at about 45 F) to LBW, I will post CW kWh results, as soon as I get them...

There is also the possibility, I would think, that both a gid-count battery capacity, and the lower capacities reported by CW and dash miles, might both be "right".

If I understand correctly, the gid count, and Phil's kWh numbers using those values, represents stored energy.

There is an unknown inefficiency in discharge.

The CW reports, as well as the dash/nav screen m/kWh X miles driven reports others have posted, which mostly appear to be lower than Phil's kWh numbers, could be reflecting total kWh, at some, point past the battery.

The kWh used to heat up the battery during discharge, must coming from somewhere, right?

CW reported 11.1 kWh use, from an 80% timer charge, to the LBW, which occurred 0.5 miles after losing the second range bar.

And the gid count says...?

 

[Ingineer]

Interestingly the "Gids" have a resolution no better than 80 watt-hours by definition, but in practice, I've seen it skip up to 4 counts at times (so far), so it's actually got an accuracy at any given time of only 320 watt-hours. (+/- 160)

Here's an example of the worst-case I saw:

I've set the top line on LEAFSCAN to show Gids (G), Watt-hours derived from Gids (W) which is simply Gids x 80, and finally "real" high-resolution watt-hours as calculated directly from the Battery ECU/LBC. (W2) The "3" looks funny because it just changed from a "4" and is counting reasonably fast because the Leaf is in Ready mode. (note the ~300w load)

Gids/watt hours is not interesting because it's just sitting there. You can't tell if it's charging/discharging or whatever. The real watt-hour figure is counting away reflecting the Leaf's load. (as is the SoC)

If you know your Gids at 80% charge, just multiply it by 93 to get your approximate actual usable watt-hour capacity. (roughly the last 2 gids are unusable, and most cars charge to ~95% for "full", but since this "full" varies quite a bit, but 80% is always 80%, it is presumably pretty accurate.)

The last "full" charge I did stopped at 94.591% and was reading 278 Gids and IIRC is was 22250 watt-hours. So right now I'm reading 65.3% SoC which if there was zero correction in a perfect world, I should have 15360 watt hours, but as you can see it's a little less, but is exactly 400 watt-hours off the Gids.

-Phil

 

[RegGuheert]

Thanks, Phil, for continually adding details to the picture of what goes on inside the beast! Do you ever sleep? :?:

I'm thinking that Nissan or used LEAF owners will have a little trouble dumping a low-capacity LEAF on a buyer toting a LEAFscan!

The last "full" charge I did stopped at 94.591% and was reading 278 Gids and IIRC is was 22250 watt-hours. So right now I'm reading 65.3% SoC which if there was zero correction in a perfect world, I should have 15360 watt hours, but as you can see it's a little less, but is exactly 400 watt-hours off the Gids.

So this makes me wonder why you calculate full capacity in Gids if you have access to higher resolution Wh data. Is it because Gids is all you get at the end of a charge? Or perhaps Gids is the only energy number you ever get directly from the car and the W2 number LEAFscan is showing is calculated by integrating the power used over time and subtracting that integral from the Wh computed from Gids at the end of the last charge?

 

[EVDRIVER]

I still see no value in using Carwings other than to turn on the heater once in a while.

 

[edatoakrun]

...

If you know your Gids at 80% charge, just multiply it by 93 to get your approximate actual usable watt-hour capacity. (roughly the last 2 gids are unusable, and most cars charge to ~95% for "full", but since this "full" varies quite a bit, but 80% is always 80%, it is presumably pretty accurate.)

-Phil

So, to repeat, what capacity in kWh do you (or anyone else) see between 80% and LBW?

Looks like CW reported 11.1 kWh for my battery, on this day.

It should be fairly close to your observations, of your own battery, if it was based on the same information you now are getting in real-time from your LEAFSCAN.

What is your level of uncertainty?

How variable by battery temperature?

Does LBW always occur at a constant level of battery capacity, or, perhaps, variably, as reported here:

http://www.roperld.com/science/NissanLEAFRangeCalculation.pdf" onclick="window.open(this.href);return false;

More discussion on this question here:

http://www.mynissanleaf.com/viewtopic.php?f=31&t=8420&start=10" onclick="window.open(this.href);return false;