Phoenix Range Test Results, September 15, 2012

[mwalsh]

I bet the 281 meter will not tell you this and its automatically set to 281 when it leaves the factory. Perhaps the cell variability falls within the error budget of the GUID meter anyways.

Indeed. I've thought for a while now that maybe 281 was the maximum that could be displayed. This is why it took so long for any or us to see drops from 281. And why the drops from 281 to where most of us are at now seemed so rapid, coinciding as they did with warmer weather for most of us. We were dropping charge from the moment our cars were first charged all the way. We simply were't seeing it.

 

[TickTock]

You've broken the code! Yes, climate control is measured through the economy meter. Hence, they aren't specifying a 10% range variation for a new car; quite the opposite.

That is why, as I said in my lengthy post, that the only explanation for the range window would be a range of capacities. Since we know this is not true, then the lower number is just completely fabricated and serves no purpose other than to let the testers show that there is no capacity loss. I suggested we just dispose of this number (as did you for the test and later in rebuttals to deniers) since it is clearly fabricated. But I am also open to input from others (particularly Nissan) who think they know why this lower number exists.

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

 

[palmermd]

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

Now this is something that I think could be on the right track. In fact this may explain the variability of the results displayed on the dashboard of the vehicles in the test. But the cars SHOULD have consumed at the exact same rate given that they were all travelling on the same road at the same time with cruise control set to the same speed using the GPS.

 

[DrInnovation]

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

Now this is something that I think could be on the right track. In fact this may explain the variability of the results displayed on the dashboard of the vehicles in the test. But the cars SHOULD have consumed at the exact same rate given that they were all travelling on the same road at the same time with cruise control set to the same speed using the GPS.

Since the vehicles were driving a constant speed range should not be a function of the economy meter. But in related potential, In reading the OP, it does nto say it was GPS checked, just 62GPS = 64 CC on the leaf. Was that speed verified at all during the test -- variations in actual speed could be significant.

While power meter reading dont impact actual efficiency, since internal resistance is a function of battery degradation/aging the degraded batteries may (should) have lower efficiency. However the correlation (by eye) there seems weak. Given other threadhs that have shown a drift in the mpkWh meter, http://www.mynissanleaf.com/viewtopic.php?f=31&t=9064&start=30" onclick="window.open(this.href);return false, it seems more likely the mpkwh meter is inaccurate in what it says, though that should not/does not effect actual range.

 

[edatoakrun]

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

Now this is something that I think could be on the right track. In fact this may explain the variability of the results displayed on the dashboard of the vehicles in the test. But the cars SHOULD have consumed at the exact same rate given that they were all travelling on the same road at the same time with cruise control set to the same speed using the GPS.

Precisely!

Which is what I was saying last night:

...And how do we know that the Phoenix test conditions resulted in all of the cars (or even any one of them) using precisely 4 m/kWh, so that the test distances observed are even valid for a direct comparison to the 76-84 miles at 4 m/kWh Nissan
estimate?...

IMO without valid and accurate m/kWh reports, the only way to calculate capacity variability in a single LEAF is by recharge (As TickTock has been attempting) or in repeated range tests under closely controlled conditions, as I have been trying to do.

If you want to try to determine comparative capacity from a range test, Such as was attempted in Phoenix, you should:

1: Normalize all variable charge capacity factors, battery temperature during the charge cycle being the largest one I am aware of.

2: Normalize all variable efficiency factors: Which I believe the test was quite well-designed to do.

However, by not using the Carwings reports to monitor driver efficiency, valuable data on how well "driving style" was normalized, was lost.

CW reports the drivers total kWh generation by regenerative braking over every trip. Regen use lowers overall drive efficiency. I suspect even the most efficient Phoenix drivers, who avoided virtually all friction braking, had to use some regen, to decelerate during the test, due to traffic conditions. Records of how this report of how efficiency varied between the different drivers, should always be noted, not only in a multiple vehicle test, but to monitor the variable driving efficiency, in repeated tests of a single LEAF.

3:Monitor recharge capacity, with the same charge capacity variables controlled as was done during the charge session prior to testing.

It is true you that by doing so, you are measuring the energy used to "refill" the battery, not the amount you used during the test, so any BMS variability will also come into play here. But, IMO, the great advantage of this approach, is that there are far fewer variables you need to control for, than in a range test. That is why I a have recently begun to try to monitor both.

If you can accurately calculate your capacity from charging data, a range test to determine capacity reduction from any standard (measured or theoretical) would seem to be unnecessary, other than to determine how LEAF efficiency varies between cars, or over time.

 

[TonyWilliams]

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

Certainly the most logical assumption so far. I doubt they are accounting for degradation, since they specify a new battery.

 

[TonyWilliams]

Since the vehicles were driving a constant speed range should not be a function of the economy meter. But in related potential, In reading the OP, it does nto say it was GPS checked, just 62GPS = 64 CC on the leaf. Was that speed verified at all during the test -- variations in actual speed could be significant.

They were GPS checked. Each car started with the maintence screen displayed to verify that Vgps=1000 at whatever speed was displayed (which in every case was 64mph).

 

[TonyWilliams]

There is another explanation for the range in the chart - variability in the economy meter due to instrumentation accuracies (i.e. gids). So you may be only getting 4.0 mpkWh when your meter indicates 4.5.

Now this is something that I think could be on the right track. In fact this may explain the variability of the results displayed on the dashboard of the vehicles in the test. But the cars SHOULD have consumed at the exact same rate given that they were all travelling on the same road at the same time with cruise control set to the same speed using the GPS.

Well, if the battery internal resistance were greater in one car over another, the rate of consumption would go up. I'm not convinced that's the case, and is simply loss of capacity. But,mthere could be a combination of both.

What I mention above is not a reflection of the actual dash meter; that's just a bit wacky, like the GOM.

 

[DaveinOlyWA]

so i did an very unofficial range test http://daveinolywa.blogspot.com/2012/09/leaf-instrumentation.html" onclick="window.open(this.href);return false;

and i guess my instrumentation is way off too. based on the miles/kw gauge i should have been able to go about 98 miles but looks like i could only actually go about 87 miles (i went 82.3 to 6.5% SOC @ 4.7 miles/kw)

granted, the temps were much colder than optimum so a shorter range was probably expected. but based on what i could actually get from my LEAF. the miles per k should be about 4.1-4.2 at the most instead of the 4.7

now wondering how much stock I can put into the GID meter? since it can only tell me what the car tells it.

GID reading this morning; 280. apparently my suspected 2-5% loss is now officially gone

**note** on the cars in the AZ test since this was not mentioned so can only guess that all were still using OEM tires?

 

[TonyWilliams]

now wondering how much stock I can put into the GID meter? since it can only tell me what the car tells it.

GID reading this morning; 280. apparently my suspected 2-5% loss is now officially gone

The Gid display (not really a meter, since it doesn't calculate anything; it displays the LEAF's data) is pessimistic in our 12 car test, and seemingly more accurate at the top. My car with 89% Gid wint 91% of the expected range.

Cars with more degradation were much harder to predict. I think at 280, you're just fine.

 

[opossum]

Are the conditions of an EPA range test publicly available? To put all this debate and statistical analysis to bed, it would be nice if we could pay a firm to put one or two of the capacity loss cars through the official EPA driving cycle under the same conditions and measure the range. Ideally, we would use the same city/course, same temperature and fresh OEM tires. This is probably way more hopeful than possible, but I'd love to donate to this cause!

Dave: Yes, all the Tempe 12 had OEM tires.

 

[RegGuheert]

Certainly the most logical assumption so far.

You seem to assume that a car that has not had a 100% charge in a while can come up to a fully balanced condition with a single charge plus remaining plugged in for a few hours. My experience is that is an invalid assumption and that it takes a few cycles to 100% to achieve balance. This is almost certainly one of the reasons why your LEAF did not have as much capacity as you expected when you arrived in Phoenix.

I doubt they are accounting for degradation, since they specify a new battery.

I would, if I were Nissan. That battery could be over six months old before it is purchased by a customer.

They aren't going to have their service guys run a test with targets that are likely to fail on a vehicle with "normal" characteristics.

 

[TonyWilliams]

Certainly the most logical assumption so far.

You seem to assume that a car that has not had a 100% charge in a while can come up to a fully balanced condition with a single charge plus remaining plugged in for a few hours. My experience is that is an invalid assumption and that it takes a few cycles to 100% to achieve balance. This is almost certainly one of the reasons why your LEAF did not have as much capacity as you expected when you arrived in Phoenix.

I doubt they are accounting for degradation, since they specify a new battery.

I would, if I were Nissan. That battery could be over six months old before it is purchased by a customer.

They aren't going to have their service guys run a test with targets that are likely to fail on a vehicle with "normal" characteristics.

Good points. Of course, I do know that the batteries in a big imbalance can take several 100% charges, as I've experienced the imbalance issue first hand (and got towed). After that, the car did Post 100% "top off" charges for the next three 100% charges, the first top off over 4 hours later, then each subsequently less.

I got to that imbalance with multiple partial charges and very low battery SOC.

Edit: my car doesn't go more than one week without a full charge.

 

[edatoakrun]

so i did an very unofficial range test http://daveinolywa.blogspot.com/2012/09/leaf-instrumentation.html" onclick="window.open(this.href);return false;

and i guess my instrumentation is way off too. based on the miles/kw gauge i should have been able to go about 98 miles but looks like i could only actually go about 87 miles (i went 82.3 to 6.5% SOC @ 4.7 miles/kw)

granted, the temps were much colder than optimum so a shorter range was probably expected. but based on what i could actually get from my LEAF. the miles per k should be about 4.1-4.2 at the most instead of the 4.7...

Note the similar results, as a percentage of change in reported m/kWh, that I have seen in my range tests, where my dash m/kWh have increased from 4.9 m/kWh to 5.7 m/kWh over the last year, in a (hopefully ) more normalize range test.

http://www.mynissanleaf.com/viewtopic.php?f=31&t=9064&start=30" onclick="window.open(this.href);return false;

Unless you have previous range test results that actually show lost range, l I think it's very likely your LEAF is actually not calculating the correct kWh use, and so is miscalculating the m/kWh reports.

I've lost count of how many times I've read on this forum:

"I've lost capacity since my LEAF was new, but my increased driving efficiency means I am still getting the same range I always have. "

 

[RegGuheert]

Of course, I do know that the batteries in a big imbalance can take several 100% charges, as I've experienced the imbalance issue first hand (and got towed).

Ouch! I didn't know about that!

After that, the car did Post 100% "top off" charges for the next three 100% charges, the first top off over 4 hours later, then each subsequently less.

I'm interested to hear what your new car can get up to subsequent to the Phoenix test after a few daily 100% charges. Would you mind doing that if you haven't already and post the GID reading? It would be particularly interesting for a 100% charge done at around 80F.

Edit: my car doesn't go more than one week without a full charge.

I don't charge to 100% nearly that often, but I also don't know how long it takes to get out of balance, either.

 

[TonyWilliams]

I charged to 100% now, to go up to Orange County's NPID event, but it's 68F here now.

 

[DaveinOlyWA]

Certainly the most logical assumption so far.

You seem to assume that a car that has not had a 100% charge in a while can come up to a fully balanced condition with a single charge plus remaining plugged in for a few hours. My experience is that is an invalid assumption and that it takes a few cycles to 100% to achieve balance. This is almost certainly one of the reasons why your LEAF did not have as much capacity as you expected when you arrived in Phoenix.

I doubt they are accounting for degradation, since they specify a new battery.

I would, if I were Nissan. That battery could be over six months old before it is purchased by a customer.

They aren't going to have their service guys run a test with targets that are likely to fail on a vehicle with "normal" characteristics.

my GID count count over past 6 consecutive days

274
271
271
273
277
280

what the diff here? Fall has arrived. the last 2 days morning have been in the upper 40's to low 50's with garage temps in the upper 50's. previously garage temps would be high 60's to low 70's meaning temps play havoc on charge at a level that is lower than I expected.

but the additional charge capacity is suspect. I did an unofficially range test and per the dash only had access to about 18.5 KW which is about .7 KW less than last spring (which was colder by 10-15 º)

so guessing the 4.7 miles per KW i got should be more like 4.1 or 4.2 now i did not go to turtle but did get to 6.5% SOC per GID meter. (display or whatever) driving 82.3 miles in a large variety of driving conditions

 

[RegGuheert]

what the diff here? Fall has arrived.

Thanks for the data! Interesting stuff! It almost looks like the BMS protecting the pack by reducing the charge at higher temps. Do you happen to have termination voltages for the charges on those same days? (Yes, I know that we don't know for sure that the voltage from the GIDmeter is without temperature compensation applied, but Phil's recent statement makes me think that it may be both accurate AND absolute.)

I think we are going to start getting some interesting data as we move into fall and winter this year. Many of the cars no longer read simply 281 all the time so we will be able to see things moving around some. OTOH, we also have good evidence from Tony's test that GIDs are not a great measure of stored energy.

 

[DaveinOlyWA]

what the diff here? Fall has arrived.

Thanks for the data! Interesting stuff! It almost looks like the BMS protecting the pack by reducing the charge at higher temps. Do you happen to have termination voltages for the charges on those same days? (Yes, I know that we don't know for sure that the voltage from the GIDmeter is without temperature compensation applied, but Phil's recent statement makes me think that it may be both accurate AND absolute.)

I think we are going to start getting some interesting data as we move into fall and winter this year. Many of the cars no longer read simply 281 all the time so we will be able to see things moving around some. OTOH, we also have good evidence from Tony's test that GIDs are not a great measure of stored energy.

voltages rarely vary. i checked the SOC this morning at 6 am when garage was 56º. about to check it again. did not check voltage this morning but its always between 392 and 394 so not a whole lot of variance.

getting ready to head out, so hold on...

DEGRADATION!!

ok now its 18.4 C (65 F) 393 volts but GID count is 279...must of power cycled too many times. now i have to redo my "on the fly" charging plans

 

[surfingslovak]

In case you missed it - Transport Evolved Episode 119


Click to open