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

[ChadFeldheimer]

So you got the increased range you wanted and you paid for but you are concerned that the battery rating difference may not be as much as advertised???

Hello!

No, I've never driven the EPA test cycle nor have I gotten anywhere near 265 miles.

If you could magically have a 87KWh battery but it had the SAME range, would that put a good taste in your mouth?? Didn't you get every bit of the additional range you paid for? This seems to be the foolish part...

EPA range estimates only have a loose relationship to real world range. To be clear, I'm not saying that the EPA range estimate is useless - that would obviously be unreasonable. But to go the other way and argue that EPA range estimate is the end-all-be-all is similarly unreasonable.

The point is that an educated consumer takes whatever datapoints are available to them and fuses them together as they best see fit to make their purchasing decision.

When I made my decision, I didn't have all the datapoints I have now. I thank all the technical contributors to this thread for that.

 

[JRP3]

the car will switch into a ultra low power mode (everything off but the bmb's and they even go down and cells continue to self discharge) and you can drain it to zero. the reserve capacity is technically usable capacity in storage, just not while driving. try driving to zero miles and park your car in the desert. you will use the "reserve" capacity fairly fast in that situation as self discharge is dependent on temp

Do NCA cells that don't have a drain on them self discharge? Tesla has said something to the effect that a car can be left for a year or so without damaging the pack, though it has to be "woken" back up by Tesla, presumably enabling the BMS and contactors to allow charging.

- - - Updated - - -

How do you feel about the fact that we have strong indications that Tesla have been very self serving in the way they been rounding up and down to the nearest 0/5 when assigning the 60/70/85/90 numbers, very likely in order to exaggerate the difference between the different packs and thus (I assume) up sell the larger ones?

I doubt anyone choosing between the 60 and 85 would have chosen the 60 if the 85 were 80. Now that the 60 and 85 are gone, which was obviously part of the long term plan, comparisons between them and the 70 and 90 aren't really relevant.

- - - Updated - - -

when car stopped 2.888 and while driving min. 2.6V !

Cell voltage under load is not the same as resting voltage. So 2.6V under load means there was some capacity left in the pack.

 

[flathillll]

Do NCA cells that don't have a drain on them self discharge? Tesla has said something to the effect that a car can be left for a year or so without damaging the pack, though it has to be "woken" back up by Tesla, presumably enabling the BMS and contactors to allow charging.

- - - Updated - - -



I doubt anyone choosing between the 60 and 85 would have chosen the 60 if the 85 were 80. Now that the 60 and 85 are gone, which was obviously part of the long term plan, comparisons between them and the 70 and 90 aren't really relevant.

all lithiums cells self discharge with zero load. if you figure the reserve is roughly 5% of the pack and you know the self discharge can vary between .25-3% month dependeding on the age and temp of cell and whether the BMB is still active

best case it could be around a year. worst case it could be a little over 2 months roughly

 

[JRP3]

Haven't seen any measurable self discharge in LiFePO4 cells stored for many months, and I'm pretty sure there have been LiCo cells stored for years that were not bricked.

 

[pupik]

Cell voltage under load is not the same as resting voltage. So 2.6V under load means there was some capacity left in the pack.

yes, I agree...
resting voltage after "game over" 3V
battery was very soft
BMS protects cells to drop below 2.5V
car current (outside pack) completely shut down 0A
inside current BMS unknown for me now( maybe WK057 ?)
so, life main pack is "???"
life for 12V bat. very small... (16A unlocked car and ?A locked car)

 

[jpet]

But how do feel about the numbers? Do you think the designations are fair or unfair. Or are you indifferent?

I understand that Tesla tries to present their cars in the best possible way by using the highest specs available.

Since this is a battery capacity centered thread, let's take a look at the specs for the Sanyo NCR18650GA cell.
This is the cell that is allegedly being used as the basis for the custom version of the Tesla cell in the "90 kWh" battery.

The NCR18650GA is being marketed as a 3500 mAh cell (see the title on the webpage).
3.5 Ah x 3.6V x 7104 = 89.510 kWh

The minimum capacity of this cell is 3350 mAh and the typical capacity is 3450 mAh.

Now let's take a look the the detailed spec sheet from Sanyo (Panasonic).
On page 7 of this document, you'll be able to find the rated capacity of 3300 mAh.

So here we are, 1 cell and 4 numbers...

Want to know the real capacity an owner of a 90D or P90D will be able to experience?
Let me take an educated guess: 3.3 Ah x 3.6V x 7104 = 84.396 kWh with 80.396 kWh available for driving after removing the 4 kWh energy buffer (brick protection).

My take is that Tesla's specs are OK but, as wk057 indicates, they should come with an asterisk indicating typical usable capacity when new.

 

[flathillll]

Haven't seen any measurable self discharge in LiFePO4 cells stored for many months, and I'm pretty sure there have been LiCo cells stored for years that were not bricked.

ask yourself why Roadsters have been bricked with NCA cells? all lithium cells self discharge. it dependes on temp and age. accept it

 

[stopcrazypp]

It seems to me, that you just confirmed everything what was stated in the first Wk post in this thread.

Your car total batery capacity is 81,6kWh = so your cells are ~11,48Wh/3200mAh (It fits with specification of NCR18650BE cells).
Your car total usable capacity is 77,6kWh (4 kWh energy buffer is forever locked for you)
Your car range indicator is "dumb" (in this particular case it has incorrectly calculated available range and showed you about 10km less than it was possible to drive = there is no zero mile protection).

After you have fully discharged the total usable capacity 77,6kWh the car was automatically "bricked" via BMS.

That is exactly the problem people have been having with this numbers. This is a ~1 year old car with 14048km (8729 miles, ~40-50 cycles) of wear on it. Under the same wear, NCR18650B cells would lose 5.8% of its capacity (unfortunately there is no official datasheet for NCR18650BE). That doesn't tell you what is reported for a car right off the assembly line, much less a pack or cell off the assembly line (before supposed QC related cycling and aging). If you back calculate with 5.8% loss, it would indicate a brand new pack (with cells not QC cycled by Tesla) would get 82.1 kWh + 4kWh = 86.1 kWh.

Also the "4kWh" (it should be 4kWh, not 4kW) is also never measured and not possible to measure inside a car.

 

[Auzie]

That is considered an urban legend by folks who have actually tested cells for this, e.g. jack rickard over at evtv.me, and others have repeated his experiments. The websites you linked made statements of 2-3% loss without data or proof or peer-review, so you assumptions and conclusions cannot be accepted as true.

The links I provided are to electropaedia

Woodbank Communications (WC) provides all the information on electropaedia and does all the screening of the published materials.

WC is a consultancy company specialising in batteries, battery applications and battery management systems. WC has over 50 employees, hence more than sufficient for a quality peer review prior to publishing.

By definition a consultant is a professional and an expert who provides their professional or expert advice in a particular area.

 

[stopcrazypp]

ask yourself why Roadsters have been bricked with NCA cells? all lithium cells self discharge. it dependes on temp and age. accept it

Roadster (non-3.0) uses Sanyo LCO not NCA. 2200-2400mAh (exact number disputed: most people assumed 2200mAh; Martin Eberhard said 2400mAh personally).

If I recall correctly, self discharge rate is also not constant. It varies with SOC (much more rapid at high SOC).

 

[pupik]

Anyway I just cannot technically understand that the car does not release the brake when it is not driving anymore and a 12V battery is still in condition. Everything worked in the car like display, windows, wipers but NOT the hand brake. So I was not able to push the car to the charger (0.1miles). I had to charge the car on the crossroad.
But don't worry it was just a testing crossroad near to my house :biggrin:

 

[tomas]

Mods: suggest the pupik discharge issue has its own thread so he might get some prompt help and advice.

 

[pupik]

Mods: suggest the pupik discharge issue has its own thread so he might get some prompt help and advice.

ok, I delete my question, no problem, my test with SOC 0% ended. :smile: Torture batteries ended. :smile:

 

[MP3Mike]

There are a couple of links if you are curious

Battery performance

Battery life


If the above is accepted as true, then it follows that the battery capacity specification (which is given as a single number for a new battery) will always be higher than the values that people can see in their cars.

But I suspect that the car report numbers right at, or maybe above, rated capacity on the 60s and 70s. The fact is that Tesla uses a different rating calculation between models, which misleads people on what they are actually getting.

I wish somebody with a 60, 70, and 90 would get together with someone with a CAN logger so we could get some actual numbers.

- - - Updated - - -

Do you know what a hypothetical example is? Fine replace 155 flat out with "155 until the speed is limited and then continue to hold down the accel until you run out of juice" and the same is true.

anyone even remotely familar with lithium cells knows the capacity is inversely proportional to the discharge rate. the point is when you go really fast, not only do you get less range because of the air resistance (higher load) and the reduced motor efficiency, and blah blah blah; your effective battery capacity will also be reduced given the much higher than average discharge rate

At full draw on a non-ludicrous 85 you are only drawing ~0.2A per cell. (And even in your hypothetical situation does it actually draw 1,300A continuously?)

So I doubt that it makes a significant difference in the amount of power you get out of the cells.

 

[AWDtsla]

At full draw on a non-ludicrous 85 you are only drawing ~0.2A per cell. (And even in your hypothetical situation does it actually draw 1,300A continuously?)

So I doubt that it makes a significant difference in the amount of power you get out of the cells.

Negative. You can't divide the total amperage by the cell count. It's something closer to 17A.

 

[stopcrazypp]

Negative. You can't divide the total amperage by the cell count. It's something closer to 17A.

Right, the pack is in 96 parallel 74 series configuration. You divide total amperage by 74 to get individual cell numbers.

To get down to ~0.2A per cell you need to be doing EPA UDDS (which averages 19.6 mph and Model S takes more than 16 hours to discharge its usable capacity given it can go 320+ miles on it).

The EPA highway cycle is ~0.5A per cell and in a test of the NCR18650B, that meant a 2.4% reduction in capacity.
http://www.teslamotorsclub.com/show...able)/page62?p=1368818&viewfull=1#post1368818

If you go top speed, the drop should be drastically more.

 

[JRP3]

ask yourself why Roadsters have been bricked with NCA cells? all lithium cells self discharge. it dependes on temp and age. accept it

As Stop pointed out the Roadster used older LiCo, and I don't know if the Roadster disconnects all loads at some point including the BMS, so the Roadster example proves nothing.

Some further data from Professor Jeff Dahn, paraphrase: "I have cells from 1993 stored at 20% SOC, put them on in 2013, like new"

https://youtu.be/9qi03QawZEk?t=1h7m55s

 

[islandbayy]

Just a update. I have cycled the cells WK057 has sent me a few times. Interesting results. I have numbered them 1 and 2 respectively to keep track. I will not go into depth until I have done more series of tests, and then I will have part one of my testing video series, but so far, I am averaging about 3,000mAh/cell, no matter what rate of charge and discharge.

I'm using my automated Lithium Cell tester right now, It has a 200mAh limitation, options are 200, 300, 500, 700 and 1000 discharge rates, and the same, as well as 2000 mA charge rate. So I am doing tests at all rates, on both charge and discharge. Then I will apply a 50mAh load, and test it's discharge capacity in that manor. As you can imagine, the 200 setting takes a very long time for both charge and discharge. But I will report back. I am testing with a ambient temp of 70*F. The Tesla cells arn't even getting warm at or below 500 mAh charge or discharge (not noticeably via touch), but discharged at 1000, they do start to noticeably warm up.
Also, for fun, I tossed my crap Ultrafire knock off cell (Re-labeled old laptop cell), that does not hold anywhere near it's rated capacity. First discharge of it at 1,000 and I ended up needing to put it in the Freezer as it was getting dangerously hot. Likewise at the initial first test I did, at 300mA rate, the ultrafire got very warm.

In terms of thermal build up, the Tesla cells are doing great.

 

[JRP3]

ask yourself why Roadsters have been bricked with NCA cells? all lithium cells self discharge. it dependes on temp and age. accept it

I remembered I had a Tenergy LiCo 18500 cell which has to be at least 2 years old, (probably older), that I bought new and never used. I just put a voltmeter on it, two different ones in fact to double check the reading, and it's sitting at 3.85V. This was sitting at moderate room temperatures for most of it's life, (55-75F), not in a fridge or freezer.

 

[bhzmark]

Cool to see crowd sourced analysis of this. Thanks to those doing the work.

My main takeaway is that this is complex and true apples to apples comparisons are tough. I don't see any wrong doing by tesla, but I didn't see any on the HP issue either.

Another variable is that the acceleration of the 70 is less than the 85 and 90 and I wonder if that isn't a basis to rate the car somewhat more efficient at extracting more energy from the same cells and partially explain the difference.

Even if that doesn't explain it, rounding up a little on the 85 and down on the 70 doesn't seem a major offense. I made my decision based on rated range and acceleration differences. I don't care at all how they got to those performance metrics.

The marginal cost weighed against marginal range and acceleration benefits is how I made my decision. The methodology or resulting calculations for the battery (or hp) specs are irrelevant to me. I just care about the real world results.