Phoenix Range Test Results, September 15, 2012

[mkjayakumar]

Nope only once. And then i gave the GID meter to someone else to check their car capacity.

 

[Stanton]

Did you measure the gids several times over several days? Remember when I borrowed that same meter, my first measurement was sub 270, but over the next couple of days each reading crept higher until I got 274 or so I believe.

I did, and it stayed around 260 @100% for me. Mine were done in late May, after which I handed the meter off to Jay (who obviously took his reading at a later date).

 

[TonyWilliams]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

I will receive $965.49.

Nissan has also now bought back 4 of the 12 cars tested.

 

[DaveEV]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

Wow - I did not see Nissan being your mystery benefactor for this test! Good on them!

 

[jspearman]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

I will receive $965.49.

Nissan has also now bought back 4 of the 12 cars tested.

Wow! Congratulations. Glad to hear Nissan doing something right in all this. Interesting that 4/12 cars have been bought back, which means 2 besides Scott and Azdre/Opossum. Did the other buybacks also go through the lemon law process?

 

[Stoaty]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

I will receive $965.49.

Nissan has also now bought back 4 of the 12 cars tested.

Now that's what I call a goodwill gesture! Way to step up, Nissan!

 

[Volusiano]

I hope this token of good will gesture is a good start but it doesn't end there. I truly hope Nissan will continue with a general good will remedy to all early adopters who bought the car but suffer early capacity loss. I truly think a pro-rated battery swap plan will go a long way in restoring good will with early adopters.

 

[mwalsh]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

Well done Nissan!

 

[surfingslovak]

I hope this token of good will gesture is a good start but it doesn't end there. I truly hope Nissan will continue with a general good will remedy to all early adopters who bought the car but suffer early capacity loss. I truly think a pro-rated battery swap plan will go a long way in restoring good will with early adopters.

+1

My expenses are being reimbursed as well, and I wanted to take the opportunity to thank Tony, Jeff Kuhlman and Chelsea for their efforts. I must admit that I was pleasantly surprised by Nissan's reaction to the Phoenix range test. That said, some of the affected owners might feel differently about their situation, and hopefully this will be addressed before long as well. Glad to hear that four Leafs have been repurchased.

 

[DaveinOlyWA]

I am happy to report that Nissan has agreed to a reimbursement of $2,593.67 for expenses related to this test.

I will receive $965.49.

Nissan has also now bought back 4 of the 12 cars tested.

Wow! Congratulations. Glad to hear Nissan doing something right in all this. Interesting that 4/12 cars have been bought back, which means 2 besides Scott and Azdre/Opossum. Did the other buybacks also go through the lemon law process?

being the one of the contenders for "Nissan fanboy of the year" i have to say i am disappointed. Nissan may have done right by alleviating any financial burden for those involved, but to me its more of a surrender. I wanted to see Nissan come out and start pulling packs to analyze them, replacing packs, etc. to find some answers

payoffs dont provide answers.

 

[Stoaty]

being the one of the contenders for "Nissan fanboy of the year" i have to say i am disappointed. Nissan may have done right by alleviating any financial burden for those involved, but to me its more of a surrender. I wanted to see Nissan come out and start pulling packs to analyze them, replacing packs, etc. to find some answers

I think they didn't see the need to do that, they probably already know the answer: Arizona and the Leaf are not a good fit.

 

[TonyWilliams]

being the one of the contenders for "Nissan fanboy of the year" i have to say i am disappointed. Nissan may have done right by alleviating any financial burden for those involved, but to me its more of a surrender. I wanted to see Nissan come out and start pulling packs to analyze them, replacing packs, etc. to find some answers

I think they didn't see the need to do that, they probably already know the answer: Arizona and the Leaf are not a good fit.

Phoenix buyers and leasees might have reached the same conclusion, had they known that the battery would be at End Of Life in 4 years with just normal 12,000 mile driving.

Unfortunately, Nissan didn't think it very important, particularly when they stick the consumer with battery capacity issues. Bravo!!

 

[walterbays]

Phoenix buyers and leasees might have reached the same conclusion, had they known that the battery would be at End Of Life in 4 years with just normal 12,000 mile driving.

Many, but not all.

If the battery costs $5,000 to exchange after 4 years that's an additional 10 cents per mile operating cost to add to something like 2 cents per mile electricity cost. If gas averages $4/gallon in Arizona over the next 4 years that 12 cents per mile is like driving an ICE that gets 33 MPG. Someone could still choose the car under those circumstances if range wasn't an issue for their driving needs, and if they preferred the smooth quiet clean power of the Leaf, and if the government rebates made the purchase cost of the Leaf comparable to an ICE car that could begin to approach the Leaf in quiet and torque.

But of course if someone were counting on 2 cents per mile operating costs to offset an initial purchase price much higher than a conventional car they considered comparable, then battery replacement would ruin their plans.

Even for me in a mild climate I'm guessing my battery cost will be 4-8 cents a mile on top of 2 cents a mile for electricity. The great uncertainty is because Nissan hasn't disclosed expected battery life nor expected battery replacement cost. I still consider it a good deal. But Nissan should have told customers that electricity would be the small portion of their operating costs, and disclosed what they know about battery life and cost.

 

[Stoaty]

Phoenix buyers and leasees might have reached the same conclusion, had they known that the battery would be at End Of Life in 4 years with just normal 12,000 mile driving.

One small correction: based on Nissan's own data and my battery aging model developed from their data, 12500 miles per year in Phoenix has a predicted End of Life of 5 years.

 

[RegGuheert]

I am glad to hear that Nissan is reimbursing expenses for the test! That is an outstanding gesture of good will!

One small correction: based on Nissan's own data and my battery aging model developed from their data, 12500 miles per year in Phoenix has a predicted End of Life of 5 years.

So you are saying that a LEAF with 66% more mileage and 3X the calendar life as Scott's car will have less degradation? That doesn't add up to me. Perhaps I am missing something...

 

[Stoaty]

I am glad to hear that Nissan is reimbursing expenses for the test! That is an outstanding gesture of good will!

One small correction: based on Nissan's own data and my battery aging model developed from their data, 12500 miles per year in Phoenix has a predicted End of Life of 5 years.

So you are saying that a LEAF with 66% more mileage and 3X the calendar life as Scott's car will have less degradation? That doesn't add up to me. Perhaps I am missing something...

No, I am saying that based on the data Nissan provided this will be the case. One outlier does not a model break. Perhaps he kept his car at 100% SOC most of the time. Maybe he parked in the baking sun day after day. Perhaps he drove like a maniac and had a very low miles per kwh and thus cycled his battery a lot more. Perhaps he had his car painted in a 150 degree oven. Or maybe we just don't know the reason. As I said in another thread, my model predicts TickTock's capacity to within 1%. I hope to add other factors to the model if I can have a reasonable idea that the numbers will be meaningful.

 

[RegGuheert]

No, I am saying that based on the data Nissan provided this will be the case.

I have provided several criticisms of the model in other threads. Here is a succinct summary:
1) I do not see evidence of a reduction in the rate of calendar losses over time in the published literature or in the data we have compiled here.
2) There is a severe dearth of calendar aging data available for EV batteries, particularly for this chemistry.
3) Most (all?) cycling tests do not account for the fact that real-life cycles amount to an increasing DOD with each cycle.
4) Extrapolation of battery life is much less predictive than interpolation from measured data. This is particularly true if the extrapolation predicts a beneficial inflection that does not appear in the literature. In fact, much of the literature predicts a break in the opposite direction.
5) We have data from several posters here that are far worse than what is predicted by the model. Given the small population of LEAFs in Phoenix, these cars cannot be far outliers. In addition, the grouping of the data in Phoenix makes me believe that sigma is not large enough to include all of these cars within the first two sigma. (Plus, an extremely large sigma around the predictions indicates that some LEAFs will last over ten years in Phoenix, which I do not think is a credible prediction.
6) Nissan's LA4 testing does not fit the superhighway driving that many drivers do in Phoenix.

One outlier does not a model break.

Agreed. But with only a couple of hundred LEAFs in AZ over 12 months old, there have been 4 buy-backs. That indicates to me these cars are not far outliers, but are within about 2-sigma of the mean.

Perhaps he kept his car at 100% SOC most of the time. Maybe he parked in the baking sun day after day. Perhaps he drove like a maniac and had a very low miles per kwh and thus cycled his battery a lot more. Perhaps he had his car painted in a 150 degree oven. Or maybe we just don't know the reason.

Sure, we can speculate like that, but we also know of extremely pampered LEAFs in Phoenix that lost a bar after 18 months.

As I said in another thread, my model predicts TickTock's capacity to within 1%.

Just as one outlier does not a model break, one fitting data point does not a model make. This is particularly true since we have no idea which part of the distribution TickTick's LEAF occupies.

I hope to add other factors to the model if I can have a reasonable idea that the numbers will be meaningful.

I appreciate all your efforts in building a model for LEAF degradation! I hope that you don't take my criticisms of the model's predictions as an indication that your efforts are not appreciated! Keep up the good work!

 

[TonyWilliams]

Phoenix buyers and leasees might have reached the same conclusion, had they known that the battery would be at End Of Life in 4 years with just normal 12,000 mile driving.

One small correction: based on Nissan's own data and my battery aging model developed from their data, 12500 miles per year in Phoenix has a predicted End of Life of 5 years.

Here's my data, based on revised Nissan data:

76% capacity in 5 years with 7500 miles/annum = 4.8% per year decrease, but we already know that about 10% (+/-5%) is what this chemistry will do after as few as a few dozen cycles (and my April 2012 build car has already done), so:

100% - 10% (first year loss) = 90% - 76% (remaining in 5 "Nissan-LEAF-Years(TM)") = 14% / 4 (more years) = 3.5% per the remaining four years to hit 76% in 5 NLYs. I assume that at 70% is End Of Life shortly before 7 years, and where the degradation falls like a rock. So, the following is pretty much all Nissan's data (Let me know if there's anything in here you disagree with):

X = Nissan's Current (October 2012) Revised 
Phoenix Battery Degradation Projections

             New Car    Year 1  Year 2   Year 3   Year 4   Year 5   Year 6   Year 7
              0 miles    7500     15000   22500   30000   37500   45000    52750
              0 km       12070    24140   36210   48280  60350    72420    84490

100%         X

95%

90%                    X

86.5%                   #          X

83%                                          X
                                                                                    
79.5%%                             #              X

76%                                            #             X

72.5%                                                                  X
                                                          #
69%                                                                              X

# = Normal drivers at 12,500 miles/annum in Phoenix
             New Car  Year 1  Year 2   Year 3  Year 4  Year 5   Year 6  Year 7
              0 miles  12500   25000  37500  50000  62500    75000    87500
              0 km     20116   40233  60350  80467  100584  120701  140817

X = Nissan-LEAF-Year(TM) drivers in Phoenix, according to Nissan revised data (Oct 2012)
# = Normal drivers at 12,500 miles/annum

Edit: I give up on this table format !!!!

 

[Stoaty]

We have data from several posters here that are far worse than what is predicted by the model.

I don't know whether that is true until I complete my model and test other cases--and that is the next item on my list. Note that the 5 year EOL prediction was for city type driving (high miles per kwh). If you plug in lower values, my model will give a worse outcome. So perhaps I should have said 5 years EOL driving 12500 miles per year with average 4.68 miles/kwh from the dash. If someone gets 3.5 miles/kwh it will worsen the degradation significantly for the same amount of miles in my model. I hope to have an easy way to get actual numbers from the inputs by the weekend.

Bar losses do not count, low Gid readings do not count. Only range test or bench testing are acceptable data to me. We know there are a large number of bar losers, but we don't know how much range most of those cars have actually lost.

PS Criticism OK. I am doing the best I can with the information I have. Once I add the miles/kwh I think you will find that my model is a reasonable predictor for most of the cases we have (but I won't know for sure until I set this part up).

 

[TonyWilliams]

PS Criticism OK. I am doing the best I can with the information I have. Once I add the miles/kwh I think you will find that my model is a reasonable predictor for most of the cases we have (but I won't know for sure until I set this part up).

The good news is that I think we have all the parameters that affect EOL. Now, it's just tuning (which you appear to be getting mastered).