Roadster battery pack specs?

[doug]

Could someone define for me this C terminology? Is that a measure of current as a function of charge capacity?

Iisjsmith, your units don't work out. You keep mixing energy with power.
(x)V * (y)Ah = (x*y)Wh [i.e. energy], not (x*y)W [which is power].

 

[iisjsmith]

From Battery University.com (very nice site) at http://www.batteryuniversity.com/partone-16.htm:

What is C-rate?

The charge and discharge current of a battery is measured in C-rate. Most portable batteries are rated at 1C. This means that a 1000mAh battery would provide 1000mA for one hour if discharged at 1C rate. The same battery discharged at 0.5C would provide 500mA for two hours. At 2C, the 1000mAh battery would deliver 2000mA for 30 minutes. 1C is often referred to as a one-hour discharge; a 0.5C would be a two-hour, and a 0.1C a 10-hour discharge.
The capacity of a battery is commonly measured with a battery analyzer. If the analyzer's capacity readout is displayed in percentage of the nominal rating, 100% is shown if a 1000mAh battery can provide this current for one hour. If the battery only lasts for 30 minutes before cut-off, 50% is indicated. A new battery sometimes provides more than 100% capacity.

And oops, sorry about the unit mixup. I'm gonna edit that post and add those little "h"s. They should all be Wh and kWh.

 

[iisjsmith]

Okay, I found out all the info we need. SRB posted this link (www.elecdesign.com/Articles/Index.cfm?AD=1&ArticleID=14108) on the Tesla Motors blog. It has all the specs, and I've been able to work out some numbers.

Each cell is 3.6V and 2200mAh. The cells are arranged into sheets, with each sheet being 621 cells. Each sheet is 32.4V, so if each cell is 3.6V then 32.4/3.6 = 9 cells in series. If there are 9 cells in series that means 621/9 = 69 parallel strings. So the wiring of each sheet is 9S69P. If there are 69 strings then 2.2Ah * 69 = 151.8Ah. So each sheet is 32.5V and 151.8Ah.

There are 11 sheets, but one sheet is used just for the car's internal systems and is not connected to the motor. So you have 10 sheets used to power the motor. It says "The remaining 10 sheets provide 324 V for the motor" so that means the 10 sheets are connected in series. That gives us a final amount of 324V and 151.8Ah of energy to the motor. This works out to 324V * 151.8Ah = 49183Wh or 49.2kWh. Just where Tesla Motors says we should be at ~50kWh.

I have been unable to find any Li-ion cells that are 3.6V and 2.2Ah with high discharge rates. The Sony ones I mention in my previous post are 3.75V, not 3.6V. I know that is a small difference...and these numbers are all nominal voltages anyway...but I thought I might be able to find something. The only hint Tesla Motors drops is that the cells are from a Japanese manufacturer.

Freakin sweet, man. It took forever, but finally some details are starting to drip out.

 

[malcolm]

Thanks iisjsmith.  I hadn't picked up on the 10 sheets for the motor, one for the car's internal systems.  Only problem with those 26650 VT cells is that they are 26 mm in diameter instead of the 18 mm of the 18650VTs.

Am I reading these specs correctly?  4.1V max potential discharging to 3.76V has a capacity of 1030 mAh?  So what is the capacity if you discharge from 4.1 down to 3V instead? Or 4.15 to 3V which is what Tesla say they do with their cells?  Is that extra missing 1.2 Ah capacity hiding here?  (Obviously these must be different cells designed to cope with discharge down to 3V instead of 3.76V)

 

[okashira]

Okay, I found out all the info we need.  SRB posted this link (www.elecdesign.com/Articles/Index.cfm?AD=1&ArticleID=14108) on the Tesla Motors blog.  It has all the specs, and I've been able to work out some numbers.

Each cell is 3.6V and 2200mAh.  The cells are arranged into sheets, with each sheet being 621 cells.  Each sheet is 32.4V, so if each cell is 3.6V then  32.4/3.6 = 9 cells in series.  If there are 9 cells in series that means 621/9 = 69 parallel strings.  So the wiring of each sheet is 9S69P.  If there are 69 strings then  2.2Ah * 69 = 151.8Ah.  So each sheet is 32.5V and 151.8Ah.

There are 11 sheets, but one sheet is used just for the car's internal systems and is not connected to the motor.  So you have 10 sheets used to power the motor.  It says "The remaining 10 sheets provide 324 V for the motor" so that means the 10 sheets are connected in series.   That gives us a final amount of 324V and 151.8Ah of energy to the motor.  This works out to  324V * 151.8Ah = 49183Wh or 49.2kWh.  Just where Tesla Motors says we should be at ~50kWh.

I have been unable to find any Li-ion cells that are 3.6V and 2.2Ah with high discharge rates.  The Sony ones I mention in my previous post are 3.75V, not 3.6V.  I know that is a small difference...and these numbers are all nominal voltages anyway...but I thought I might be able to find something.  The only hint Tesla Motors drops is that the cells are from a Japanese manufacturer.

Freakin sweet, man.  It took forever, but finally some details are starting to drip out.

You are completly right, my calcuation gives >8.096 amps/battery is required for 250 horsepower. that's a bitchin battery.
It looks like the Sony batteries you found fit the bill. the 3.75 is the nominal rating for a 4.2v max charge. if the Tesla uses 4.15 or 4.10max, the nominal voltage AND the nominal mAh capacity goes down, so Tesla just re-rates them.

Here's somthing neat: we need a 8.096 A/(90% effiency * 2.2 Ah) = 4.0887C discharge rate battery to produce 250 horsepower.
from previous blogs, we've learned that Tesla's motor isn't limited to 250hp, but limited only by the electrical power you can provide to it.
If tesla were to extract the full 10c from the sony battery's, that would work out to 10/4.0887* 250 = 611 horsepower!!!!!!!!

of course Tesla's electronics, inverters, and the motor would probably run into some difficulties at that power level  ;D


(edit: calculations. didn't realize Tesla only uses 10/11ths of the batteries to power the motor.)

 

[okashira]

whoops. I guess the 26650VT's are no good, they're too big as pointed out. Oh well.

 

[malcolm]

If the owner can choose to charge the main pack to 3.8, 4.1 or 4.15V is this the same for the 11th sheet?  Is this one always fully charged to 4.15 or only to 50% (3.8V) to improve working life?

 

[vfx]

Martin says the batteries are from a fortune 500 company. Don't know what year though, here's this year's: http://money.cnn.com/magazines/fortune/fortune500/full_list/

He just said in the latest battery post that there were Japanese.


Narrows the field a bit.


e

 

[vfx]

Of course I meant to say that "they" (the batteries) are Japanese.

I'm sure Martin is fully aware that everyone already knows that Japanese exist.


e

 

[iisjsmith]

At 90grams the 26650VTs are also twice the weight of a standard 18650.

The plan would be to use fewer of the 26650VTs in a future ESS, so that the weight wouldn't change. But I don't think you could get the same performance out of only ~3400 26650VTs.

The thing that really got me interested with the 26650VTs is the flat capacity curve as discharge rate increases. You are seeing 2400-2500mAh capacity from .2C to 10C. That is amazing! Plus, they have been rated up to 50A discharge!!!! Talk about a power boost.

Another Sony battery that looks interesting is the 18650G8. At 2550mAh capacity (at .2C) they have the highest volumetric energy density I have seen....545Wh/L!!! I think they are still limited in the discharge rate like most 18650 cells...but I haven't found any data to confirm this.

 

[Embassy]

I posted on the blog thread too regarding the discharge rate issue, and someone responding to my comment positively was footnoted with a "don't expect answers to every last one of your questions because we have trade secrets we'd like to keep." message. I'm very surprised to learn, though, that the battery pack segregates a tenth of its capacity for accessories. Why not just use a DC-DC inverter and work it that way? The inverter wouldn't weigh very much.

Basically, though, the big unanswered question is the discharge rate. There are basically three kinds of lithium batteries. The lithium cobalt batteries have very high energy density (2000-2600 mah in the 18650 format), the lithium manganese (also called "spinel") batteries have very high power denisty (up to 10C discharge) but lower energy density (1400-1600 mah) and the new lithium phosphate batteries have very very high power density (~30C) but reduced energy density like the lithium manganese batteries.

On the subject of C rating, a simple way I explain it to people who ask me (which is seldom) is this way: By convention, battery capacity is given in amp hours, and normalized to a one hour discharge. It's as if when we bought buckets we said "This bucket can dribble one ounce per second for an hour" rather than "This bucket holds 28 1/8 gallons." But that's how it is in the battery world - "capacity" means the amount of current the battery can sustain for an hour, not how much energy is in the battery. To calculate the energy you need to know the voltage of the cell, which is basically determined by the chemistry.

The other variable of interest is "maximum discharge rate." Basically, this is an index of how fast you are permitted to empty your bucket. This is key because very high energy density Lithium Ion (cobalt) cells can't be emptied very quickly. Lithium phosphate or manganese cells are capable of amazing discharge rates but do not have competitive energy density with many alternative cells. Complicating maters further is the fact that even when you discharge a high-rate-capable battery quickly, the capacity changes (think of this as "if you dump your bucket out too quickly you're going to spill some").

So basically, if you're dumping your bucket out in half an hour, you're pouring at 2C.

 

[iisjsmith]

To put real numbers on the C rate issue, use the numbers from the Roadster. The Roadster ESS has a capacity of 152Ah. It will provide 152A for one hour if you discharge it at a 1C rate. If you discharge the ESS twice as fast, in 30 minutes, you will discharge at 2C and it will provide 304A. You can drain the ESS in 15 minutes if you discharge at 4C, which will provide 608A.

This is why the ESS must use 4C rated cells. If the maximum power they can reach is 189kW, and the nominal voltage is 324V, then you need to provide 583A. 583/152 = 3.84 so you need to discharge at 3.84C to reach the maximum power.

 

[mwarren_us]

The specs for the LR18650 cells from BiPOWER USA include graphs showing up to 3C discharges. A 4C discharge, limited to around 10 seconds, should not be overly abusive to the batteries. At that point, the Roadster is at or beyond top speed, the ESS can start cooling off the batteries and the Porsche you were racing can turn tail and head home.

What physics limit a Li-Ion cell's discharge rate? While a complete 15 minute, 4C (100%) discharge would be very abusive, what damage would a 10 second, 4C (1.13%) burst do to the cells? Such a burst would represents just 1.13% or 555 Wh from a 49.2kWh pack.

A more practical way to ask this might be: "How frequently can the Roadster's ESS handle 0-60 mph hole shots before cell damage starts?"


Thanks,

Mark

 

[TEG]

Yes, I think you folks have it figured out.

Some Tesla people spoke at tech forums and confirmed that the ESS can send 600amps to the e-motor and pull ~4C from the cells.

When fully charged it puts out 411V (4.15V per cell)
Drained to the stopping point it is 297V (3V per cell)

The PEM may limit discharge rate not just to save the battery life, but also to prevent the (air cooled) motor from overheating.

 

[TEG]

Well, I have a hypothesis that seems to fit all the facts we know about .. try this on:

The application is a 4C discharge application, we know the pack is about 375 volts and we know it's 6831 cells. Clearly that's 99 series by 69 parallel. But the 10C cells you can get in the 18650 format have too little energy, and the 2C cells have too little discharge rate.

I think they're using both kinds in parallel. Specifically, I think there are 47 strings of 99 2300 mah cells that have 2C discharge (such as the Sanyo 18650F), and 22 strings of 99 1500 mah cells that have 10C discharge characteristics such as the Sanyo 18650W. That stack gives you 536 Amps at maximum, about 52 kWh of total energy, 141.9 amp hours of capacity.

That was a brilliant theory even if it turns out not to be true...

 

[iisjsmith]

We cleared this up on page 2. The ESS is not a single large "pack" with all the cells arranged in a 99S69P config. There aren't two different packs with two different discharge rates.

 

[markwj]

Resurrecting this 10+ year old thread, can we confirm this with what we know now?

Confirmed: For a 2.x pack, that is 11 sheets in series, each of 9 bricks in series, and each brick 69 cells in parallel.
Confirmed: 11 * 9 * 69 = 6,831 individual cells (11S9S69P layout).
Speculation: 99 bricks in series, 375V nominal, is around 3.79V/cell.
Speculation: 2.1Ah a cell gives around 54kWh total
Speculation: Roadster will refuse to charge below average cell voltage around 3.64V (pack voltage 360V)

Does anyone know minimum and maximum cell voltages for standard and range modes? Was the 2.1Ah/cell confirmed?

Then, for the 3.0 battery, we have new cells presumably in the same 11S9S69P arrangement. @hcsharp mentions the 3.0 cells have a slightly higher maximum voltage. Any details?

It would be good to get this all documented in one place.

 

[markwj]

One source also told me he had gotten these numbers from T:

• 11S9S69P
• Cells are 2.3Ah
• Nominal sheet voltage is 36V
• Nominal pack voltage is 396V
• 3V/cell (297V/pack) is 0% SOC
• 4.2V/cell (416V/pack) is 100% SOC in range mode
• 4.18V/cell (414V/pack) is 100% SOC in standard mode
• Balancing aims for a delta of around 20mV, and operates above 4V/cell

But that doesn't match the Gruber newsletter figure of a minimum 3.64V/cell (360V/pack) to be able to charge, and the numbers don't add up to 53kWh.

 

[petergrub]

Per the Roadster Lithium-Ion Battery Emergency Response Guide, the 1.5/2.X battery specifications are as follows.

Module
Nominal Voltage: 33.3V
Maximum Voltage: 38V
Weight: 100lbs
Dimensions: 70cm x 65cm x 8cm (27.56" x 25.59" x 3.15")

Pack
Nominal Voltage: 366.3V
Maximum Voltage: 420V
Weight: 1200lbs
Dimensions: 110cm x 82cm x 70cm (43.31" x 32.28" x 27.56")

Then, for the 3.0 battery, we have new cells presumably in the same 11S9S69P arrangement. @hcsharp mentions the 3.0 cells have a slightly higher maximum voltage. Any details?

I have a 3.0 battery pack open in the shop. I'll get a sheet out to take pictures and note any differences. The 3.0 pack is using Panasonic H cells, but I don't have a part number or datasheet yet.

 

[X.l.r.8]

if they upped the cells to the 3.4Ah cels like in the early model S then they are pretty much the same as the NCR B's.
If Okashira is the same person i am thinking he is he has probably pulled apart more tesla battery packs than most. He's also the most likely person to have replacement cells for a reasonable price. If someone threw him one of each cell (or if they own a cell analysier like most DIY battery builders do) we can easily spec out the cells.