Tesla Model S 18650 Cell Test Data

[Johan]

No, the energy capacity never changes, just the energy delivered.
Tesla could be rating them for a quasi-static discharge rate.

But what does it then matter to me that the battery is 85kWh when this is just theoretical? I mean if the car battery couldn't ever under any thinkable normal circumstances deliver more than 81kWH then, according to me, it's an 81kWh car battery (when used to drive a car). They shouldn't market them as 85kWh.

 

[tga]

All batteries are like that. Lead-Acid battery are WAY worst than that. Under load, you can't even get 30% of the capacity. It all depends on internal resistance and battery chemistry.

It's Peukert Effect. All batteries Ah rating is determined at a specific current draw.

Peukert's law - Wikipedia, the free encyclopedia

 

[okashira]

But what does it then matter to me that the battery is 85kWh when this is just theoretical? I mean if the car battery couldn't ever under any thinkable normal circumstances deliver more than 81kWH then, according to me, it's an 81kWh car battery (when used to drive a car). They shouldn't market them as 85kWh.

What difference does it make to you what the capacity is? What if it was 20kWh and the car still did 285 miles; would you not buy it? Serious question.
The only way to accurately rate them would be in Amp-hours, but electric car stuff is confusing enough for most consumers.

- - - Updated - - -

It's Peukert Effect. All batteries Ah rating is determined at a specific current draw.

Peukert's law - Wikipedia, the free encyclopedia

Does not apply to Li-ion, just lead acid AFAIK

 

[Johan]

What difference does it make to you what the capacity is? What if it was 20kWh and the car still did 285 miles; would you not buy it? Serious question.
The only way to accurately rate them would be in Amp-hours, but electric car stuff is confusing enough for most consumers.

You're right it's range that matters, but we're used to thinking kWh when it comes to energy use and energy cost. Also if it had a 20kWh battery and did 285 miles I would really want to how they did that. If that was possible a LEAF with a 24kWh battery would be doing 300 or more miles

 

[Hasse]

Not trying to hijack thread, but here it seems there would be someone who could tell me approximately how much is left in the 85kWh battery (percentage or miles) when voltage of the pack is 347 volts?

 

[llavalle]

Not trying to hijack thread, but here it seems there would be someone who could tell me approximately how much is left in the 85kWh battery (percentage or miles) when voltage of the pack is 347 volts?

Hard to say for sure...I don't have the data but 347V = 3.614583V per cell (16 modules in series of 6cells in series). Check the graph. at 1 amp he has around 1500mAh at that voltage.

Assuming these are 3100mAh cells, and using Ah for measurement, this is 51.6% battery left. This is obviously flawed since we need to use Wh for accurate measurement, not Ah.

 

[stopcrazypp]

Capacity depends on how you test and rate them. Nominal voltage depends on discharge rate. I'll run one at ~100mA load to see the actual energy capacity and update the chart to show watt - hours instead of amp-hours on the x-axis.
The pack is 7104, and the cells I have have lost ~1%-2% capacity from new.

Right, you need to do a test with much lower load to get a capacity rating and you need to integrate to get Wh instead. At a 200mA load, for 3400mAh you should get around 12Wh, for 3100mAh you should get around 11Wh.

Here are the test conditions for the 3350mAh (AKA 3400mAh) NCR18650B:
http://industrial.panasonic.com/lecs/www-data/pdf2/ACA4000/ACA4000CJ425.pdf
And the real world:
http://lygte-info.dk/review/batteries2012/Panasonic NCR18650B 3400mAh (Green) UK.html

Here are the test conditions for the 3070mAh (AKA 3100mAh) NCR18650A:
http://industrial.panasonic.com/lecs/www-data/pdf2/ACI4000/ACI4000CE25.pdf
And real world:
http://lygte-info.dk/review/batteries2012/NCR18650A protected (Green) UK.html

 

[kennybobby]

Not trying to hijack thread, but here it seems there would be someone who could tell me approximately how much is left in the 85kWh battery (percentage or miles) when voltage of the pack is 347 volts?

It depends.

Is this a measured voltage or something you calculated? If measured, was it the open circuit voltage (no load), or while charging, or while running (discharging under load)? 347 Volts will have different meanings depending upon how it was measured.

If calculated, then 3.6vpc is Panasonic's average cell voltage, but they also discharge to lower levels than Tesla uses, so i would say it is about 35% left.

 

[Hasse]

It depends.

Is this a measured voltage or something you calculated? If measured, was it the open circuit voltage (no load), or while charging, or while running (discharging under load)? 347 Volts will have different meanings depending upon how it was measured.

If calculated, then 3.6vpc is Panasonic's average cell voltage, but they also discharge to lower levels than Tesla uses, so i would say it is about 35% left.

I measured with a multimeter on the pack contactors with no load on the battery (did this on a salvage battery, not on the car I am driving)

 

[kennybobby]

...re: Peukert's law...
Does not apply to Li-ion, just lead acid AFAIK

Why do you say that--your own test data in the first graph at 1A and 3A clearly show a capacity difference. Your 10 Amp test would have also but for the fact that the unit shut down. The A-Hr capacity shown on the X-axis is a calculated value based upon a measurement of current and time--the 10 A data likely has an error in one of those since it shows an impossible increase in capacity at a higher discharge rate.

 

[kennybobby]

I measured with a multimeter on the pack contactors with no load on the battery (did this on a salvage battery, not on the car I am driving)

Okay that's a good OCV data point, 347 V is 3.615 volts per cell, so it looks in the middle for Panasonic cells, but toward the low end for Tesla's purposes.

During supercharging folks have shown screenshots with various miles available: 0 at 341V, 35 at 358V, and 41 at 360V.

If you have a 24vdc power supply you could charge up the individual modules one at a time (assuming you don't have a 400 V lithium battery charger).

The WhizKid found some of his salvage modules were bleeding off current thru the BMU board resistors. It is only ~100 mA but could drain a full pack in 90 days.

 

[Hasse]

Great, thanks for the input, guys! Will try to patch the salvage car together first and see if I get it to charge. Will update in a while.

Now back to original programming

 

[Johan]

Great, thanks for the input, guys! Will try to patch the salvage car together first and see if I get it to charge. Will update in a while.

Now back to original programming

Out of curiosity Hasse: I'm I understanding correctly that 1) You've got a Model S in Åland and 2) You also managed to get Your hands on an extra pack? (I do hope you didn't get the 85kWh model for Åland. How big is the island from coast to coast?)

 

[Hasse]

Out of curiosity Hasse: I'm I understanding correctly that 1) You've got a Model S in Åland and 2) You also managed to get Your hands on an extra pack? (I do hope you didn't get the 85kWh model for Åland. How big is the island from coast to coast?)

I got the 85kWh, perfect for day trips to Stockholm and Helsinki. Also for longer trips, drove to Italy last summer. For driving on Åland (roughly 50 km in diameter), a 20 kWh pack probably would be enough.

Yes, did also get my hands on an extra pack (actually an entire salvage car).

 

[scaesare]

Great testing data okashira, thanks for sharing!

Do you know the weight of an individual cell?

Also, do you mind sharing your method for liberating the cells from the module assemblies (glue, etc...)?

Thanks.

 

[Johan]

@okashira and others with proper insight and understanding of Tesla's batteries:

Did any test a cell from a 60kWh pack? If so, was if different from a cell in an 85kWh pack and how?

The reason I'm asking is the new 70kWh pack. Did they achieve that by switching to the cells that are in the 85 pack, but jeeping the same number as in the old 60? Or did they add cells? Did they increase the voltage interval that defines 0-100 SOC? (Okashira's testing suggest the current lower cutoff may be very conservative). Or could it be the next generation of NCA cells (the ones they'll manufacture in the Gigafactory)?

Exciting times...

 

[okashira]

60 and 85 kwh are the same cells

They just use 14 modules instead of 16

And they put in dummy cells to fill in empty spaces

The 70kWh pack will be the same cells with 14 modules with less empty spaces 98% sure on this. Nothing special.

 

[Johan]

60 and 85 kwh are the same cells

They just use 14 modules instead of 16

And they put in dummy cells to fill in empty spaces

The 70kWh pack will be the same cells with 14 modules with less empty spaces 98% sure on this. Nothing special.

Thanks for your expert opinion.

 

[snooper77]

60 and 85 kwh are the same cells

They just use 14 modules instead of 16

And they put in dummy cells to fill in empty spaces

The 70kWh pack will be the same cells with 14 modules with less empty spaces 98% sure on this. Nothing special.

How come 16 modules / 14 modules is not equal to 85 kWh / 60 kWh then?

 

[scaesare]

How come 16 modules / 14 modules is not equal to 85 kWh / 60 kWh then?

As okashira mentioned not only do they omit two entire modules.. they also omit some cells in each module... about 10 per parallel group, I believe. That makes it a 64p84a arrangement for a total of something like 5,376 cells in a 60. They can fill in some of those as well.