Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable)

[Khatsalano]

Very interesting information in this thread.

But in the end, I don't care much about the actual technicals of the pack. I care about the final-derivatives: range (especially in different temperatures and conditions), performance, and durability. That's what that "85" number on the back of the car means to me ... it is just a marketing approximation.

- K

 

[TSLA Pilot]

I'm not advocating people getting up in arms and marching an army of lawyers on the lawns of Tesla. Honestly, I'm sure they'd find some loophole or other slick way out of such a lawsuit anyway.

What I do want is for Tesla to start advertising honest specifications for their products. Their products sell themselves. They don't need these lies to push them, and they should realize that.

After reading through to the current last post (#98) of this thread, I have to say this is the part that is the crux of the matter, and it affects all of us that spent big money (at least for us) to buy multiple versions of these cars (in our case, four MS's thus far):

"What I do want is for Tesla to start advertising honest specifications for their products. Their products sell themselves. They don't need these lies to push them, and they should realize that."

When I agonized over the extra $20k (IIRC) for the 85 kWh battery on our first Model S purchase back in early 2013 (or "just" the extra $10k for the 60 kWh), I presumed that the extra $20k was going to get us a full 45 kWh of capacity.

So we spent the extra $20k for the 40 kWh, plus the "extra" 5 kWh of battery.

But we, apparently, did not get the promised/advertised extra capacity.

Add in the remarkable EPA-rated range calculation (265 miles, really? On what planet?) and, well, this information makes me feel mildly cheated.

And wk057 sums it up well--and assuming his calculations are accurate, of course--we now know with 20/20 hindsight that Tesla didn't need to lie. Originally it was, perhaps, going to be a Model S with 40, 60 and 80 kWh batteries, but some Tesla marketing genius figured out that "85" sounded better and would lead to more people selecting the top, most profitable model as it would seem to provide more value/capacity for less marginal cost?

It worked on us.

As we burst through 52 week lows on the SP, it's just another depressing day in Teslaworld.

What next?

Call me "Frustrated."

 

[wk057]

The amount of hate mail I've gotten in the last 15 hours... sheesh.

I mean, I expected it this time at least, but damn. Impressive amounts of hate. It's almost as if Tesla has a free army of Internet defenders at their command. lol. I mean really, I guess some people have nothing better to do. You guys all realize I don't get paid for nor profit from my postings here in any way right?

For those of you who bought not caring about the actual specs, or are otherwise cool with Tesla having false specs published... well, good for you I guess. I wish your comments, responses, and direct communication were a little more objective rather than personal attacks, insults, and mocking... but, whatever. Not that everyone who disagrees has done this, but out of ~100 posts here in this thread and the ~20 emails/PMs I've gotten, the stats are not in favor of objectivity.

Anyway... such is the internet. People are totally different when they hide behind a pseudonym and the thin veil of anonymity that the Internet provides.

Back on topic, I see the hard drive analogy all over the place, and the ICE tank analogy. Those fail miserably in their own ways.

For hard drives, this is pretty well defined. They use decimal numbers and prefixes (you know, like most of the world is taught?) to describe capacity. If I buy a 1 gigabyte HDD, it will have a capacity of 1,000,000,000 bytes. Just like 1 gigawatt power plant will put out 1,000,000,000 watts, or 1 gigajoule is ~278kWh. There's no conspiracy there. Just because tech folks want to redefine basic prefixes when they're used for tech related stuff doesn't mean that's really what the numbers mean. *shrugs* A super annoying analogy for the battery capacity issue. So while some people will go on about base2 vs base10 units, there is no such thing even close for the battery capacity issue. Even if Tesla somehow decided they were going to use the base2 "kilo" for their 85 number (which to my knowledge no one in the world has ever attempted to fudge a number that badly) that'd still mean 83 real kWh should be there. So really, I don't know where people are trying to go with this particular analogy.

Formatted vs unformatted capacity for the HDD analogy... OK, I'll kind give you that one in regard to usable capacity. But for the Tesla battery it still fails since the Tesla pack starts with less capacity than it's actually rated at to begin with. Tesla's "formatting" loss of capacity (the anti-brick safety buffer) of 4 kWh is subtracted from the true capacity giving ~77 kWh usable, not 81.

Now for the ICE tank analogy, this is even worse. If I have a 10 gallon gasoline tank and a 12 gallon gasoline tank side by side and I pump in gasoline at 1 gallon per minute into each, and when it got full I immediately started pumping out/consuming gasoline at 1 gallon per minute... the tanks are unchanged, even though the 10 gallon tank is being filled and emptied more often than the 12 gallon tank. Why does this matter? Let's go to batteries. If I take, fully charged, an 80 kWh battery and put it next to an 85 kWh battery (real measurable capacities from full to empty), then I start discharging them each at 10kW, and immediately start charging at 10kW once it's empty, then after a year of this torture the 85 pack will have been cycled significantly less than the 80 pack. Cycles = wear = degradation. So while the ICE fuel tanks, regardless of capacity, will retain their capacity, in the battery case this doesn't hold true. The 80 pack will have lost more capacity due to degradation than the 85 pack in this scenario.

Now it doesn't stop there. A real 85 pack can charge and discharge about 6% faster than an 80 kWh pack under the same conditions.

Also, I'm trying to throw notes out there without responding to anything specific, but I have to single out one post because the fact that this information is still being perpetuated today is ridiculous:

What I find even more interesting is the mismatch between EPA certified rated miles and displayed rated miles in the car. These are not the same. The S85 has 265 miles EPA rated range and the dashboard shows 265 miles. However, the EPA test includes the extra reserve below zero rated miles. That means the dashboard should display less than 265 mi rated range because not all rated range is available to the end user without going into reds. It would be more accurate if the S85 displayed 248 rated miles and it was explained to the end user that the other 17 miles was reserved to improve battery life. This issue is best explained on this page before the battery diagram.

From EPA's perspective, there is no problem because the capacity is available to the end user, even if it's not clearly displayed and the end user is not aware of it. They don't want to get involved in display methods each car manufacturer uses. The problem is, it is impossible to measure battery degradation from range. Over time the computer could reduce the reserved rated miles (maximum 17 miles) to compensate for part of the degradation and the end user wouldn't know that. For example displayed rated range might drop from 265 to 260 miles in two years. That's what the end user sees. It doesn't look so bad. However what the end user doesn't see is, the reserve might have dropped from 17 rated miles to 5. There is no way to know that.

Let me be absolutely clear here, which is going to require bold large font text:

THERE IS NO CAPACITY AVAILABLE TO THE END USER BELOW 0 MILES ON THE DASH. ZERO MEANS ZERO. PERIOD!

"But, wk! People have driven below zero! You're wrong!" I'm not. The only reason that OCCASIONALLY people have been able to drive below zero is because Tesla's algorithm for determining capacity is not perfect. Depending on many factors, the algorithm can under-estimate the amount of power available at a given time, and thus when the pack reaches 0 miles the algorithm realizes it wasn't properly calibrated when it sees a higher voltage than was expected at this point based on it's estimate of capacity. So it will let the user drive until the actual low voltage cut point, which will still leave the 4 kWh anti-brick/cell-safety buffer in the pack. There are zero guaranteed miles available below zero miles showing on the dash. There is no programmed reserve. There is no user available buffer. PLEASE, I beg everyone, PLEASE stop saying that there is. You're going to get someone stranded when they think they can push it past zero.

 

[bridaus]

Challenge: Someone post the year, make and model of a car that meets all it's written specifications.

What, my 12V battery isn't actually 12 volts at all times? I wasn't able to use all the gas in my 20 gallon gas tank? A "2 by 4" isn't actually two by four inches!!! I'm suing everyone, my whole life has been a sham, I've overpaid and been under-served my whole life!

I think the Smartphone analogy is best. I bought my current because it has a long battery life. I have no idea what the actual capacity is, even though I looked at that spec when I bought it, but it didn't matter to me. Doesn't matter now. Wouldn't matter if I bought it again.

PS: Wk's work is awesome, and as an engineer I love to see when marketing numbers are outed. But if anyone here thinks that advertised specifications are more true than not, they are kidding themselves. You can pretend that Tesla (which is in almost every way better than a typical car company) should also be more "truthful" in marketing, but they hire the same human beings as all other companies do. Get over it, you are marketed to every day.

 

[Matias]

Wk057, do you think discharging pack cells with very low, e.g. 0.01C would help to achieve 85 kWhs?

 

[dsm363]

After reading through to the current last post (#98) of this thread, I have to say this is the part that is the crux of the matter, and it affects all of us that spent big money (at least for us) to buy multiple versions of these cars (in our case, four MS's thus far):

"What I do want is for Tesla to start advertising honest specifications for their products. Their products sell themselves. They don't need these lies to push them, and they should realize that."

When I agonized over the extra $20k (IIRC) for the 85 kWh battery on our first Model S purchase back in early 2013 (or "just" the extra $10k for the 60 kWh), I presumed that the extra $20k was going to get us a full 45 kWh of capacity.

So we spent the extra $20k for the 40 kWh, plus the "extra" 5 kWh of battery.

But we, apparently, did not get the promised/advertised extra capacity.

Add in the remarkable EPA-rated range calculation (265 miles, really? On what planet?) and, well, this information makes me feel mildly cheated.

And wk057 sums it up well--and assuming his calculations are accurate, of course--we now know with 20/20 hindsight that Tesla didn't need to lie. Originally it was, perhaps, going to be a Model S with 40, 60 and 80 kWh batteries, but some Tesla marketing genius figured out that "85" sounded better and would lead to more people selecting the top, most profitable model as it would seem to provide more value/capacity for less marginal cost?

It worked on us.

As we burst through 52 week lows on the SP, it's just another depressing day in Teslaworld.

What next?

Call me "Frustrated."

265 miles EPA range is what car gets. You may not always get that but that is a problem with the test cycle and where you live or how you drive.

 

[wk057]

wk057, when you return, please respond to what I posted above a few hours after your original post.

I see no evidence to support pack capacity decrease as production has progressed. The earliest cells I've gotten my hands on were from a Signature S with a VIN of S00xxx. The newest were from an S85 VIN P5xxxx. They tested pretty much identically. The cell counts also were the same in those modules.

Wk057, do you think discharging pack cells with very low, e.g. 0.01C would help to achieve 85 kWhs?

My slowest discharge test of a Tesla cell was at 50mA... or roughly 1/64C. There was no significant increase in capacity compared to my 1/10C testing.

 

[dsm363]

Challenge: Someone post the year, make and model of a car that meets all it's written specifications.

What, my 12V battery isn't actually 12 volts at all times? I wasn't able to use all the gas in my 20 gallon gas tank? A "2 by 4" isn't actually two by four inches!!! I'm suing everyone, my whole life has been a sham, I've overpaid and been under-served my whole life!

I think the Smartphone analogy is best. I bought my current because it has a long battery life. I have no idea what the actual capacity is, even though I looked at that spec when I bought it, but it didn't matter to me. Doesn't matter now. Wouldn't matter if I bought it again.

PS: Wk's work is awesome, and as an engineer I love to see when marketing numbers are outed. But if anyone here thinks that advertised specifications are more true than not, they are kidding themselves. You can pretend that Tesla (which is in almost every way better than a typical car company) should also be more "truthful" in marketing, but they hire the same human beings as all other companies do. Get over it, you are marketed to every day.

Well said. Tesla is run by humans and they will resort to many of the tried and true marketing techniques used for decades. Sure we can push them to be better but to be shocked that everything isn't perfect is odd to say the least.

 

[stopcrazypp]

Regarding NCR18650B - Sure, those cells may have lower resistance, but what's their cycle life? Heat tolerance? Charging rate tolerance? I find it difficult to believe that a stock, off the shelf cell is better all around than Tesla's proprietary cells. I mean why expend all the effort if in the end you could have bought a better product directly from the same supplier? There has to be something more to it.

Digging up relevant info buried in this fast growing thread:

One thing I completely forgot (and is commonly forgotten). What is the calendar life on that 85kWh cells that wk057 tested? While everyone focuses on cycle life, it is easy to forget calendar degradation is not zero. However, it is very hard to find a calendar life model (and it is not commonly tested as most of the time the cells tested are brand new ones).

 

[bridaus]

One other thing worth noting: Battery degradation and charge rate are NOT specs Tesla publishes, therefore in a true discussion about the "gas tank" analogy not working, you have to recognize that in neither case are there specifications that cover wear and tear on the energy storage device. (Unless I missed that spec). Don't even start thinking about the energy content of your gas summer vs winter and ethanol vs none. eek.

 

[JohnSnowNW]

The amount of hate mail I've gotten in the last 15 hours... sheesh.

I mean, I expected it this time at least, but damn. Impressive amounts of hate. It's almost as if Tesla has a free army of Internet defenders at their command. lol. I mean really, I guess some people have nothing better to do. You guys all realize I don't get paid for nor profit from my postings here in any way right?

Yes, of course you expected this. You set this thread up to illicit this type of response. As to your defenders comment, your motivation in posting this thread was in large part due to your souring on Tesla...as you've mentioned in innumerable posts over the last several months. Lets try and keep the "more objective than thou" statements out of it.

Continue on with the technical discussion, but I don't see the point in making inflammatory statements, and then complaining that you're being attacked...or that you can't have an objective discussion.

Also, until there is definitive proof, from an objective entity, can we dispense with treating this reduction in capacity as fact?

To be clear, I haven't sent any hate mail...don't really see the point in people doing so.

 

[sierra]

That's an interesting graph... at the lower temp range (moderate average temps in many locations), it implies that storing batteries at a higher SoC (~100%) is better for their health...

That's counter to the commonly accepted wisdom around here.

No, definitely not. It implies that storing batteries at a lower temperature is better for their health.

Back to my question: How old are these batteries, Wk057?

 

[wk057]

Back to my question: How old are these batteries, Wk057?

One thing I completely forgot (and is commonly forgotten). What is the calendar life on that 85kWh cells that wk057 tested? While everyone focuses on cycle life, it is easy to forget calendar degradation is not zero. However, it is very hard to find a calendar life model (and it is not commonly tested as most of the time the cells tested are brand new ones).

The cells I received with my estimate (based on mileage) of ~5 cycles on them were at roughly 55% true SoC when the vehicle was totaled. It was less than 2 months from that point to the time they were in my hands. Can't get too much better than that with these cells, since insurance, auctions, and such take time. However, cells I tested from older vehicles and ones that I know had been sitting for many months prior to reaching me showed no additional degradation from their dormant time. Further, I recently tested a couple of cells that I set aside over a year ago at various SoC and the cells stored at room temp at 50% SoC for over a year showed absolutely no measurable degradation.

- - - Updated - - -

The rest of your post isn't really on topic, but:

Also, until there is definitive proof, from an objective entity, can we dispense with treating this reduction in capacity as fact?

I'm pretty sure Tesla's own diagnostic interface reporting both the usable and total capacities of the packs that match exactly with the tests I'd previously performed is more than enough to state it as fact.

 

[JohnSnowNW]

- - - Updated - - -

The rest of your post is useless, but:

*sigh...exactly my point.

 

[wk057]

*sigh...exactly my point.

I had already edited that line to be more friendly prior to seeing you'd quoted me. My general stance on it stands, however: nothing you said in the rest of the post was on topic or useful to the conversation.

Any actual rebuttal to the on-topic point of yours that I shot down?

 

[SteveG3]

I see no evidence to support pack capacity decrease as production has progressed. The earliest cells I've gotten my hands on were from a Signature S with a VIN of S00xxx. The newest were from an S85 VIN P5xxxx. They tested pretty much identically. The cell counts also were the same in those modules.

wk057,

we know that Elon Musk said the Model S was several hundred pounds less in the summer of 2014 than it was in 2012, and there's no reason to think that those were the final improvements in dropping weight. range has not changed. if that isn't evidence of pack capacity having been decreased since the first vehicles were produced, what is your explanation for several hundred pounds in reduced weight not improving range?


Could you list out the details of the approx VIN that corresponds to where the cells came from, whether this was via a complete pack you acquired, a module, etc..

 

[wk057]

wk057,

we know that Elon Musk said the Model S was several hundred pounds less in the summer of 2014 than it was in 2012, and there's no reason to think that those were the final improvements in dropping weight. range has not changed. it that isn't evidence of pack capacity having been decreased since the first vehicles were produced, what is your explanation for several hundred pounds in reduced weight not improving range?

Could you list out the details of the approx VIN that corresponds to where the cells came from, whether this was via a complete pack you acquired, a module, etc..

No idea where they cut weight. Honestly, not particularly sure that's even relevant. If they never shipped a true 85 kWh pack, that's relevant, however.

I already noted some details on the cells I've tested here and here

 

[Skotty]

Personally, I don't really care on this issue, with the exception that it I do care that others care about it. I know roughly what the car is supposed to do, it does it, so I'm happy. The rest is just numbers fudging, like all those old 5.0 Mustangs that were really 4.9L.

It could be they had an initial design that was 85, but through design iterations, it ended up more like 81, and they didn't want to have to change all the naming and documentation. I can let that slide, as I don't think it's a big enough discrepancy to get upset about, but obviously not everyone feels that way. It does make me curious as to the real capacity of the new 90 packs.

Marketing may have stuck their grubby nubs on this at some point. Marketing it evil. I took a marketing class in college, and knew it wasn't for me, because I'm not someone who intentionally tries to deceive people for personal gain.

 

[scaesare]

No, definitely not. It implies that storing batteries at a lower temperature is better for their health.

Unless I am misreading, the "Recoverable Capacity %" for a cell stored @ 20°C with a 4.2v charge after 12 months is ~91%, whereas the same cell at the same temp stored with a 4.1v charge is only 81%.

It obviously also implies that the 60°C temps are worse, but at least on those curves the higher voltage is harder on the battery, as expected.

I almost wonder if the labels for the 20° curves were inadvertently swapped?

 

[JohnSnowNW]

I edited that line to be more friendly prior to seeing you'd quoted me. My stance on it stands, however, nothing you said in the rest of the post was on topic.

Any actual rebuttal to the on-topic point of yours that I shot down?

What am I supposed to rebut? I don't know enough about the topic to make a rebuttal...maybe you can actually say definitively that Tesla is cheating everyone out of 4kWh, and did so purposefully.

All I know is that someone who is far from objective is claiming to understand the motivations behind the results of testing he has done, and considers them nefarious.