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

[Benjamin Brooks]

My Tesla Model S 85 was advertised as 85kWh. IF it's really 81kWh, that's a BIG deal.
I know Ford and Mazda got in BIG trouble for overstating the horsepower on a couple cars and had to buy the cars back after owner realized the car was not as fast as they should have been.
This is a BIG deal. It just takes a small group of people to start making noise about this and some law firm will pick it up. I'd be very happy with a new battery pack that's really AT LEAST 85kWh, which is what I paid for.

.

Yes, class action lawsuit for all 85 owners to get a free pack upgrade to 90 (which is really 85)! That ought to kill the company... :/ agree it's a big deal but I don't want to see the company fail...

 

[wk057]

My Tesla Model S 85 was advertised as 85kWh. IF it's really 81kWh, that's a BIG deal.
I know Ford and Mazda got in BIG trouble for overstating the horsepower on a couple cars and had to buy the cars back after owner realized the car was not as fast as they should have been.
This is a BIG deal. It just takes a small group of people to start making noise about this and some law firm will pick it up. I'd be very happy with a new battery pack that's really AT LEAST 85kWh, which is what I paid for.

Yes, class action lawsuit for all 85 owners to get a free pack upgrade to 90 (which is really 85)! That ought to kill the company... :/ agree it's a big deal but I don't want to see the company fail...

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.

 

[stopcrazypp]

Further, these cells are almost certainly NOT Panasonic NCR18650B cells, like many have assumed. Close, but not identical. I've tested actual Panasonic NCR18650B cells on the same equipment as the Tesla cells, and the retail Panasonic cells always perform better and have slightly lower internal-resistance. So, the NCR18650B specs are not quite relevant.

Actually I think they would be very relevant. How much capacity did you measure on the NCR18650B cells? Given you can buy these brand new, you can do a simulated cycling (same amount of cycles as the Model S cells to you have) and then measure their capacity. This will give a you a sanity check on your methodology (in terms of cell temperature, cut-off voltages, charge/discharge current, CCCV curve, etc) given there are manufacturer and many third party tests of these specific cells. Same if you get your hands on NCR18650A.

 

[Mike K]

I guess I look at this I and I ask myself what Tesla has to gain and I can't come up with a good answer. For 99.9% of the buying public the difference between 60 and 85 is entirely arbitrary. It could just as easily be 90 and 120 and it would make no difference to most. With that in mind, had Tesla called the car the Model S 80kwh would it have been met with any less enthusiasm? Absolutely not. People would have, as the internet likes to put it, given absolutely zero f*cks. Nobody would have cared.

So I just don't see what they have to gain by intentionally misleading on a metric that is absolute greek to all but a fraction of potential customers. People shopped by range and acceleration numbers.

To further demonstrate just how much of a non-issue I think it is, let's look at how well Tesla has hid this "problem". To put it bluntly: very poorly. It doesn't take a scientist to realize there's a discrepancy here. On paper the 85kwh pack is 34.5% larger than the 60kwh yet only has 24% more range. I don't mean to diminish the OP's work. I just don't think it's the "aha!" people are making it out to be. Rather, it's an extrapolation on what anyone that could do basic math already knew and decided they didn't care about. Why people suddenly care now that the elephant in the room has been acknowledged and named is a bit confusing to me.

I am not a Tesla apologist by any stretch of the imagination. I'm not one of the people on the forum extolling the virtues of no center console or trying to explain to others why fixed beam HIDs are actually better than adaptive headlights because they have less moving parts. The Model S has some obvious shortcomings. That said, I don't view this any differently than BMW calling their 3 liter inline 6 the 335/ 535 or putting that exact same engine in the 7 series and *gasp* calling it the 740i. It means nothing to me.

In fact, in Tesla's case I think the simplest explanation would be that they simply ended up not needing as large of a pack as they thought they would. I would imagine their initial goal with the S was to hit a mileage target, not to hit a pack size target. They know as well as us that advertising a car based on pack size is stupid. So they're not trying to reach a battery size metric; they're trying to reach a range metric. It's my assumption that during the design of the car it was determined that in order to meet their mileage metric they would need an 85kwh pack and thus called that model the Model S 85. At some point either through software or some other pleasant surprise it was determined that they could meet their mileage goals with a slightly smaller pack and here we are. I don't see what the fuss is. If someone told me tomorrow that my 85kwh was a 60kwh it wouldn't matter to me at all. It's not as if it suddenly doesn't go as far or as fast.

If you bought the car on range and/ or performance and were already satisfied, I have no idea what you have to suddenly be mad about. If you bought the car solely on advertised KWH, you still have nothing to be mad about because you would have had no clue what the motor's power draw was, how much power would be set aside as a buffer, etc.

I take nothing away from the OP's research which frankly, is incredible. I just think that the serving of conspiracy theory that's being dished out with all that data is unnecessary. It sets the tone for negative responses instead of intelligent discussion.

 

[Johan]

Well... that escalated quickly

Sorry to those who saw my Think-o-saurus meme before, in the 691 hp thread, but I feel it sums up Tesla's attitude to these advertised numbers pretty well.

 

[rns-e]

The 85 (and now the 90) is the performance battery in the Tesla line up compared to the 60 (and now 70) so it would make perfect sense to assume there is some kind of equivalent battery standard to the US 1-foot roll out 0-60 mph standard.

And this standard is now established by Tesla, as they are currently the only manufacturer that offers a true EV performance battery

Sorry - just couldn't help my self

 

[MP3Mike]

Definitely not. Charging these cells to 4.35V causes them to pretty much explode randomly. I've tried it.

So just how many cells have you exploded so far? :smile:

 

[wk057]

Actually I think they would be very relevant. How much capacity did you measure on the NCR18650B cells? Given you can buy these brand new, you can do a simulated cycling (same amount of cycles as the Model S cells to you have) and then measure their capacity. This will give a you a sanity check on your methodology (in terms of cell temperature, cut-off voltages, charge/discharge current, CCCV curve, etc) given there are manufacturer and many third party tests of these specific cells. Same if you get your hands on NCR18650A.

Yeah, I made sure I had a fair playing field. I have done dozens of cycles of both the NCR18650B's and NCR18650A's. The B is still higher capacity after 20 deep cycles than the Tesla cells that have ~5 cycles. The A always lower capacity even from the start. So the Tesla cells are somewhere in between. I still have pairs of both the Panasonic A and B variants running cycle tests along with my Tesla cells. My fastest C-rate groups are close to 400 full cycles, with NCR18650B still having a very strong lead.

- - - Updated - - -

I guess I look at this I and I ask myself what Tesla has to gain and I can't come up with a good answer. For 99.9% of the buying public the difference between 60 and 85 is entirely arbitrary. It could just as easily be 90 and 120 and it would make no difference to most. With that in mind, had Tesla called the car the Model S 80kwh would it have been met with any less enthusiasm? Absolutely not. People would have, as the internet likes to put it, given absolutely zero f*cks. Nobody would have cared.

So I just don't see what they have to gain by intentionally misleading on a metric that is absolute greek to all but a fraction of potential customers. People shopped by range and acceleration numbers.

To further demonstrate just how much of a non-issue I think it is, let's look at how well Tesla has hid this "problem". To put it bluntly: very poorly. It doesn't take a scientist to realize there's a discrepancy here. On paper the 85kwh pack is 34.5% larger than the 60kwh yet only has 24% more range. I don't mean to diminish the OP's work. I just don't think it's the "aha!" people are making it out to be. Rather, it's an extrapolation on what anyone that could do basic math already knew and decided they didn't care about. Why people suddenly care now that the elephant in the room has been acknowledged and named is a bit confusing to me.

I am not a Tesla apologist by any stretch of the imagination. I'm not one of the people on the forum extolling the virtues of no center console or trying to explain to others why fixed beam HIDs are actually better than adaptive headlights because they have less moving parts. The Model S has some obvious shortcomings. That said, I don't view this any differently than BMW calling their 3 liter inline 6 the 335/ 535 or putting that exact same engine in the 7 series and *gasp* calling it the 740i. It means nothing to me.

In fact, in Tesla's case I think the simplest explanation would be that initially they thought they would need x amount of power to reach their mileage goals and then through the course of design and implementation, determined they needed less and simply decided not to re-name the model. I would imagine their initial goal with the S was to hit a mileage number, not to hit a pack size number. They know as well as us that advertising a car based on pack size is stupid. So they're not trying to reach a battery size metric; they're trying to reach a range metric. It's my assumption that during the design of the car it was determined that in order to meet their mileage metric they would need an 85kwh pack and thus called that model the Model S 85. At some point either through software or some other pleasant surprise it was determined that they could meet their mileage goals with a slightly smaller pack and here we are. I don't see what the fuss is. If someone told me tomorrow that my 85kwh was a 60kwh it wouldn't matter to me at all. It's not as if it suddenly doesn't go as far or suddenly doesn't go as fast.

If you bought the car on range and/ or performance and were already satisfied, I have no idea what you have to suddenly be mad about. If you bought the car solely on advertised KWH, you still have nothing to be mad about because you would have had no clue what the motor's power draw was, how much power would be set aside as a buffer, etc.

I take nothing away from the OP's research which frankly, is incredible. I just think that the serving of conspiracy theory that's being dished out with all that data is unnecessary. It sets the tone for negative responses instead of intelligent discussion.

My whole point is that it is an advertised spec that is false. It's not just the model number, which could be whatever. It's a technical specification. As you and others have said, it really doesn't make a difference to most people what the capacity is as long as the range is there. This doesn't help the case at all! It just makes it even *more* crazy that they misquote the spec knowing there is no reason to do so. The same as has been with other specs I've mentioned elsewhere. It's just stupid of them to overstate these things when they really just don't have to.

So just how many cells have you exploded so far? :smile:

After I blew up the first one by accident, I started doing my most stressful tests inside a little aluminum blast chamber I made for the purpose. lol. To answer your question, probably about 25 or so by now.

 

[szh]

This is quite sad... I never get more than 300km (about 190miles) from a 90% charge...

Good to know we've been lied to again.. honestly I like the car, but alot of these things annoy me

 

[Troy]

Another research from November 2014 found similar results to WK's. It wasn't a secret that the 85 kWh rated battery pack has about 80 kWh actual capacity. Details can be found on page 12 in this PDF file. In the section that is hidden, it shows the cell voltage too. You can read these if you copy and paste somewhere. I have mentioned this source elsewhere recently but wanted to add it here too because it is relevant.

Screenshot:

 

[stopcrazypp]

Yeah, I made sure I had a fair playing field. I have done dozens of cycles of both the NCR18650B's and NCR18650A's. The B is still higher capacity after 20 deep cycles than the Tesla cells that have ~5 cycles. The A always lower capacity even from the start. So the Tesla cells are somewhere in between. I still have pairs of both the Panasonic A and B variants running cycle tests along with my Tesla cells. My fastest C-rate groups are close to 400 full cycles, with NCR18650B still having a very strong lead.

What I mean is how much Wh did you measure compared to the Panasonic claimed 3400/3350 mAh (12.24Wh/12.060Wh) name plate rating for the NCR18650B? And how much Wh compared to the Panasonic claimed 3100/3070 mAh (11.16Wh/11.052Wh) name plate rating for the NCR18650A?

The 85kWh might be from the same rounding up that Panasonic does in their cell marketing, just carried over into the pack (meaning Tesla just took Panasonic's number and multiplied by 7104 cells and then rounded to nearest kWh). The rounding up on the NCR18650B cells for example is already 1.5%. That added to the 5 cycle 3.2% loss on the NCR18650A is 3.2%+1.5% = 4.7% which is pretty close to the discrepancy measured.

 

[SteveG3]

based on an Elon Musk comment (I believe about a year and a half ago), since the Model S was introduced in 2012, the vehicle has dropped a few hundred pounds. what's more, there has been a series of different battery packs.

isn't it possible that,

-the early 2012 Model S were built to use XY kWh of a battery pack that really was 85 kWh, as Tesla determined that XY kWh usable got them to a range that would hit an EPA number (265 miles) they selected for the biggest battery.
- as battery chemistry and/or pack design, management was tweaked, and weight came off the car from continual changes across the design of the car,
1. lower weight meant that a usable capacity lower than XY kWh was sufficient to deliver the same range as the original battery pack that actually had 85 kWh
2. newer battery chemistries came with improvements that allowed Tesla to increase the percentage of the battery pack available to use
- over time, these changes would mean that Tesla went from an 85 kWh battery pack to various small changes up through the one you tested and measure as 81 kWh. they didn't change the Model S 85 kWh offering to say, an 83.6 kWh Model S, then an 82.9 kWh Models S, etc... but rather knew they were delivering the same range as the original 85 kWh battery.

just to say, having spent all of ten minutes thinking about it (and not even having the knowledge to consider possible errors in your testing or calculations), there may be other explanations to this. not saying I disagree with your posting this, and raising questions.

update: re point 2 above... more broadly, a variety of potential changes to the battery pack, whether a newer cell chemistry, newer battery management, cooling, or other, that allowed Tesla to feel confident making a larger percentage of the pack available for use in propelling the car. what letter battery pack are we up to now since the original "A" pack? clearly there have been changes, and there may well have been changes allowing more of the battery to be safely used.

 

[MP3Mike]

If you bought the car solely on advertised KWH, you still have nothing to be mad about because you would have had no clue what the motor's power draw was, how much power would be set aside as a buffer, etc.

Just how do you figure that? What is set aside doesn't, nor does the motor's power draw, matter if it isn't there to begin with.

Look at it this way: Tesla charges $3,000 to upgrade from a "85 kWh" battery to a "90 kWh" battery. (The page actually says "Upgrade your Model S to a 90 kWh battery for maximum range. Available as an option on all cars equipped with both all-wheel drive and 85 kWh batteries, this upgrade increases range by about 6%.")

So if in reality the "90 kWh" battery is really only a "85 kWh" battery, like the "85 kWh" battery appears to really only be a 81 kWh battery, then you are paying $3,000 more to get what they offered you to start with. (They quoted hard specs and didn't deliver them, and from what it appears they knew that they weren't delivering what they promised/sold.)

As much as I don't like it I can only see three options for Tesla:
* Refund people who bought an 85 $3,000.
* Upgrade people who bought an 85 to a real 85 kWh battery. (Put a "90 kWh" battery in.)
* Buy the car back from people who bought an 85.

If the 90 kWh pack isn't really close to 90 kWh, then I'm not sure what option they have right now for people who paid for the upgrade other than a refund/buy-back. (There currently isn't a bigger battery to put in.)

I would really like to see the same data for the 70s and 90s, then we will really know where Tesla is on this whole mess.

Sure maybe they can wiggle out of this some way, but it doesn't seem like it will be easy for them. (And if they do that would mean that the 60 kWh pack should actually be named something different.)

 

[wk057]

What I mean is how much Wh did you measure compared to the Panasonic claimed 3400/3350 mAh (12.24Wh/12.060Wh) name plate rating for the NCR18650B? And how much Wh compared to the Panasonic claimed 3100/3070 mAh (11.16Wh/11.052Wh) name plate rating for the NCR18650A?

The 85kWh might be from the same rounding up that Panasonic does in their cell marketing just carried over into the pack (meaning Tesla just took Panasonic's number and multiplied by 7104 cells and them rounded to nearest kWh). The rounding up on the NCR18650B cells for example is already 1.5%. That added to the 5 cycle 3.2% loss on the NCR18650A is 3.2%+1.5% = 4.7% which is pretty close to the discrepancy measured.

Actually, the NCR18650B new cell testing and comparison with the spec was part of my personal confirmation that my custom equipment was working as expected. My Ah measurements matched pretty much exactly the specs (3.35Ah). However, Wh is not necessarily just nominal voltage * Ah. My Wh numbers came out slightly below the expected 12.06 Wh at 11.96 Wh, new.

Also, my NCR18650A cells didn't actually lose that much capacity in the first few cycles. I don't have that data in front of me at the moment, but I recall only 1 or 2% loss in the first 15 cycles.

In any case, my tests of the Tesla cells are very consistent and show steady degradation. I think if I continue long enough the NCR18650B and the Tesla cells will actually meet on capacity since the Tesla cells are degrading a hair less per cycle than the Panasonic cells. But, again, I don't want to dive too much into my test results until they're finalized.

 

[stopcrazypp]

This whole issue is complicated by the fact there is no pack out there being sold by EV makers that can access the rated kWh capacity.

The Leaf pack for example can access 21.3 kWh out of a "24kWh" pack or 88.75% of the nameplate rating of the pack.

77kWh usable out of a "85kWh" pack is 90.59% of the nameplate rating of the pack, so you still get 1.84% more usable capacity vs the nameplate rating than a Leaf for example. The situation is very similar if you compare to other cars.

 

[Mike K]

My whole point is that it is an advertised spec that is false. It's not just the model number, which could be whatever. It's a technical specification. As you and others have said, it really doesn't make a difference to most people what the capacity is as long as the range is there. This doesn't help the case at all! It just makes it even *more* crazy that they misquote the spec knowing there is no reason to do so. The same as has been with other specs I've mentioned elsewhere. It's just stupid of them to overstate these things when they really just don't have to.

I totally get your point. I just don't agree with the conclusion that they are intentionally misleading. Ask yourself what they have to gain by calling the battery 85kwh over 80kwh, especially in 2012 when this was most certainly all going down. Nothing, absolutely nothing. They had literally nothing to gain. It's the difference between my car having 872 unicorn power and 1000. It meant absolutely nothing to 99% of the people out there and to the 1% that it did mean something to, it was still a single variable among many variables which means that nobody, even someone such as yourself, could have determined vehicle range based on pack size alone. Knowing just the battery pack size and nothing else is akin to knowing the size of a car's fuel tank without knowing what engine is in the car.

So there is literally no benefit to them to mislead with respect to the battery size. The much more obvious explanation is that they initially determined they needed an 85kwh pack to reach their mileage goals and then through either software changes, motor changes, battery technology, etc then determined that they could reduce the battery size while still meeting their mileage target and simply decided not to start referring to a different number as to introduce confusion followed by the necessary explanation.

 

[wk057]

This whole issue is complicated by the fact there is no pack out there being sold by EV makers that can access the rated kWh capacity.

The Leaf pack for example can access 21.3 kWh out of a "24kWh" pack or 88.75% of the nameplate rating of the pack.

77kWh usable out of a "85kWh" pack is 90.59% of the nameplate rating of the pack, so you still get 1.84% more usable capacity vs the nameplate rating than a Leaf for example. The situation is very similar if you compare to other cars.

I'd be curious as to what the Leaf cells actually test to. Keep in mind I'm not arguing usable capacity, really. I'm arguing actual capacity. Of course there is going to have to be some protection reserve with lithium cells, so some % set aside for that makes sense. But in the case of 85 vs 81, that's not the case. It's not like those extra 4 kWh are set aside and that's how we get to 85. Those other 4 kWh don't actually exist.

I totally get your point. I just don't agree with the conclusion that they are intentionally misleading. Ask yourself what they have to gain by calling the battery 85kwh over 80kwh, especially in 2012 when this was most certainly all going down. Nothing, absolutely nothing. It's the difference between my car having 872 unicorn power and 1000. It meant absolutely nothing to 99% of the people out there and to the 1% that it did mean something to, it was still a single variable among many variables. Knowing just the battery pack size is and nothing else is akin to knowing the size of a car's fuel tank without knowing what engine is in the car.

So there is literally no benefit to them to mislead with respect to the battery size. The much more obvious explanation is that they initially determined they needed an 85kwh pack to reach their mileage goals and then through either software changes, motor changes, battery technology, etc then determined that they could reduce the battery size and simply decided not to change the name of the model as that might introduce confusion.

Oh no, don't get me wrong. I don't disagree. I think we're kind of on the same page. It makes no sense for them to be misleading on this. But to me, that's the problem. They have nothing to lose by specifying an accurate spec, and nothing to gain (that I can think of) for specifying the inflated one. So, the fact that they still chose to go with the latter is what amazes and irritates me.

Also, the analogies to an ICE fuel tank don't really work. The capacity of a battery implies much more than the capacity of a fuel tank.

 

[stopcrazypp]

Actually, the NCR18650B new cell testing and comparison with the spec was part of my personal confirmation that my custom equipment was working as expected. My Ah measurements matched pretty much exactly the specs (3.35Ah). However, Wh is not necessarily just nominal voltage * Ah. My Wh numbers came out slightly below the expected 12.06 Wh at 11.96 Wh, new.

This is exactly the issue I suspected. I am 100% aware that taking "nominal voltage * Ah" does not give the "true capacity" (to get that you have to integrate the area of the curve). However, that is how nameplate capacity is determined typically.

Also the test below measured 12.142 Wh from a NCR18650B. Your methodology underrates the cell 1.52% compared to that test (and the below test actually doesn't give the full capacity since cut-off is at 2.8V instead of 2.5V and the cell is a protected cell so has additional energy loss)
http://lygte-info.dk/review/batteries2012/Panasonic NCR18650B 3400mAh (Green) UK.html

Compared to nameplate method (12.24Wh vs 11.96Wh) the difference is 2.34%.

 

[wk057]

This is exactly the issue I suspected. I am 100% aware that taking "nominal voltage * Ah" does not give the "true capacity" (to get that you have to integrate the area of the curve). However, that is how nameplate capacity is determined typically.

Also the test below measured 12.142 Wh from a NCR18650B. Your methodology underrates the cell 1.52% compared to that test.
http://lygte-info.dk/review/batteries2012/Panasonic NCR18650B 3400mAh (Green) UK.html

Yeah, I didn't do a 0.2A discharge test. My slowest testing was at 340mA (supposedly 1/10C, which is a super light load for these), which is where I got my 11.96Wh number. But anyway, even with a 1.5% margin for error my data still conclusively shows that the Tesla cells can not add up to 85 kWh.

Also, Tesla's cell's IR is significantly higher than the NCR18650B's I tested, further proving that they are not the same cells.

 

[Mike K]

Oh no, don't get me wrong. I don't disagree. I think we're kind of on the same page. It makes no sense for them to be misleading on this. But to me, that's the problem. They have nothing to lose by specifying an accurate spec, and nothing to gain (that I can think of) for specifying the inflated one. So, the fact that they still chose to go with the latter is what amazes and irritates me.

That's fair and I don't disagree entirely though it could be for continuity at this point. If the car performs as advertised I don't think it's really something to get hung up on and changing the number and this point seems to serve to confuse many in an effort to appease a few.

Also, the analogies to an ICE fuel tank don't really work. The capacity of a battery implies much more than the capacity of a fuel tank.

The point was more that it would have been impossible to form any conclusion of performance or efficiency based on one metric alone.