: Panasonic Ni 18650



George S. Bower
04-03-2010, 01:07 PM
Please refer:

http://www.physorg.com/news173635366.html

This module has an energy density of 183 whr/ Kg--pretty darn good.

I have not been able to find it's C rating. Anyone care to speculate??

hermperez
04-03-2010, 02:36 PM
These cells are usually 1C, the panasonic data sheet shows up to 2C for the 2.7Ah cells. 1C would discharge the cells in 1 hour.

Since they parallel 20 of these cells then the C rating of the pack (assuming the internal wiring can take it) is 20-40C. Maximun capacity is probably at around 10C (0.5C at the cell level).

http://industrial.panasonic.com/www-data/pdf2/ACA4000/ACA4000CE240.pdf

Each module is 25.2V and 58Ah, or 1.5kwh of capacity. 12 of these modules in series would give you 300v @ 58Ah, 18kwh and could deliver 580A with no issues (if the wiring can take it).. it could power a 233hp electric motor.

Looks like a very rugged design, and it can be fixed locally.

hermperez
04-03-2010, 04:51 PM
last time I looked at these modules they had about 30% empty volume inside.. I assume for air circulation. There is probably a fan somewhere.

The neat part about these is that Panasonic promises about half the cost of present day automotive lithium batteries.. but half of what?

wtiger
04-03-2010, 06:31 PM
It looks like they took a lesson from tesla on making a battery pack. It certainly looks like it could easily mass produced. I'm curious what the price per Kw/h is or will be.

Mohsen
04-04-2010, 01:58 AM
last time I looked at these modules they had about 30% empty volume inside.. I assume for air circulation. There is probably a fan somewhere.

The neat part about these is that Panasonic promises about half the cost of present day automotive lithium batteries.. but half of what?

If my laptop is any measure - you will lose a few of the cells after about 30 deep cycles, which will make the pack useless. I hope Panasonic is smart enough to automatically open a column of cells if it has lost a cell.

hermperez
04-04-2010, 02:41 AM
Paralleling 20 cells gives you a lot of built-in redundancy.. you could lose 2 cells out of one module and never notice it. This generation of Panasonic cells are very reliable for the usual reasons: temperature and strict DoD control.

The new netbook PCs that use these cells (3 or 6 18650 cells) are getting great life due to low power and cool temps of these low powered designs.

Dont be surprised if these modules start to approach $100 per kwh capacity.

Roy
04-04-2010, 10:08 PM
battery spec. http://industrial.panasonic.com/www-cgi/jvcr13pz.cgi?E+BA+3+ACA4001+NCR18650+7+WW

There is no max C rating, but something called It, max 0.7It, anyone know what that is? Tesla goes way beyond max rating up to about 8C, must have determined it is ok.

Secret to long life is mostly do not let it discharge too much. I set my laptop to notify at 25% and shut down at 20%. Unfortunately default settings are about 10% and 5% as laptop manufacturers are more interested in advertising long use per charge.

You cannot leave failed batteries in a pack connected to the others. It must be identified and disconnected or it will short out all the other batteries in parallel and reduce max voltage in series.

Mohsen
04-05-2010, 02:53 AM
Paralleling 20 cells gives you a lot of built-in redundancy.. you could lose 2 cells out of one module and never notice it. This generation of Panasonic cells are very reliable for the usual reasons: temperature and strict DoD control.

The new netbook PCs that use these cells (3 or 6 18650 cells) are getting great life due to low power and cool temps of these low powered designs.

Dont be surprised if these modules start to approach $100 per kwh capacity.

Do you know what technology is used to switch out a column? Is it MOSFET? And what is the voltage drop across the device when it is closed? Does the discharge protection circuit sense each individual cell, or simply the column?

hermperez
04-05-2010, 07:45 AM
no switches needed for Panasonic's modules (but who knows).

These modern lipos dont short out, swell or catch fire like they used to.. the failure mode is just an early reduction in capacity. It is disastrous if one of the cells in your series string suddenly losses capacity, its voltage may actually go negative if the other cells in the string still have capacity as the pack discharges.. or even worse due to its low capacity it actually reaches full charge while the other cells are still being charged. Boom!.

So you have a series string of 7 cells (25.2VDC), ideally you want all these cells to age at the same rate.. so you try to make the string from cells of the same batch number and most important you want them to operate at the exact same temperature during their life.. this can be a problem in a laptop since you have hot spots due to the processor, space limitations for insulation and just sitting the PC on your lap.

So how do you avoid this?

1. monitor voltage for each cell in the series string.. if one cell reaches the limits STOP charging or discharging the whole string.
2. balance the cells.. each cell is monitored and discharged until all the cells have the same voltage.
3. design the pack so that all the cells are thermally matched.
4. leave safety 10% slack at both the top and low end of the charge state.. never charge to 100% or discharge to 0%.
5. hope the pack dies an early death so they can sell you a new one :)


This is how Panasonic does it (I think).. each of the 7 "sub-modules" in series is made from 20 cells in parallel. These 20 cells can be in thermal contact with each other in an aluminum heatsink (probably overkill), and electrically shorted case to case.. always all at the same temperature. Each sub-module will still have the 3.6VDC of an individual cell but now has 20 times the capacity, 58Ah instead of 2.9Ah of an individual cell. The sub-module is still discharged at the 1C rating (or higher for short pulses, under 10 secs) but that now is 58Amps.

What happens if two cells have reduced capacity in the sub-module?.. not much, the sub-module has as a whole reduced its Ah capacity proportionally.. but ALL the cells are forced to be at the same voltage level since they are literally shorted together in parallel. Very safe and you still have to monitor the voltage of that sub-module and keep it in the 80% safety window.

Instead of monitoring (or balancing) 140 cells all you now have to do is keep track of the voltage state of 7 sub-modules.. much simpler wiring.

Both lead-acid and lipos can be paralled safely, even if the individual batteries are of wildly different capacities.. the stronger battery will proportionally deliver more of the current in the team.. but the voltage will be constant for the paralled batteries since they are electrically strapped together.

Due to manufacturing defects or contamination the separator film can fail and allow the fully charged cell to instantly short out.. it gets very hot very quickly and generates gas. If it is a soft pouch prismatic cell (the pouch is made of mylar, polyester) it swells up like a balloon.. certain chemistries will vent hydrogen and ignite.. if it is encased in a steel cylinder such as 18650 types then an end cap will pop off and ooze out electrolyte, or vent hydrogen and ignite.

Modern 18650 Panasonics are not so dramatic.. they briefly get hot until the cell is discharged. The cell is now shorted, it will channel the remaining 19 cells and melt violently.. but not so, each individual cell has an internal fuse that will pop if it overheats.. thus the short is broken, and that particular cell is electrically isolated from the other 19 cells. You may not even notice one cell failed and that particular submodule has now a capacity of just 55.1Ah.. the battery management system may not even notice it either, but perhaps it saw a brief temp spike on one cell. Maybe.

Tesla takes advantage of the limited energy storage of each cell, and designs the pack in such a way that an overheating shorted cell will not damage its neighboring cells.. this would not work if the cells were larger (lets say D size instead of AA size) and stored more energy.

These shorting events are very rare in modern high quality 18650 type cells.. but they can be spectacular :)

This modularity is much easier to do with a bunch of small cells, like both Panasonic and Tesla uses.

hermperez
04-05-2010, 07:56 AM
I dont think the cells in the Panasonic modules touch each other, there is lots of empty space inside the modules probably allowing for airspace around each cell and an internal fan to redistribute/equalize the heat.. probably no external/exchanging venting of the air. Lipos operated in the 80% DoD regime generate very little heat. Very efficient.

hermperez
04-05-2010, 08:22 AM
if panasonic is smart, each 25VDC 58Ah module will have a built-in bypass relay (probably solid state.. or failsafe mechanical with a loud click) so that it can be taken out of the array of modules in the car without affecting the other modules. Probably a built-in fan and CANbus communications.

Lets say you have a Leaf style BEV using 12 modules scattered all over the car.. 12 modules in series will do 300VDC at 58A.. at the 1C rate it would provide 17.4kw of power or 23hp. This will allow the car to cruise at 75mph for exactly 1 hour, covering 75 miles of distance (at 100% DoD, probably 60 miles to be safe).

If one module failed it could be bypassed out of the series array, now you have 11 modules providing a lower voltage (277VDC) but at the same current of 58A. The car would probably go into a crawl-home mode and lower its top speed.

You could do many combination of modules, in parallel/series arrangements as long as the modules used pass-thru bypass relays.

George S. Bower
04-05-2010, 11:46 AM
The Tesla Roadster runs cooling tubes inside it's pack, with liquid coolant running in the tubes. I am not sure if they also have a fan in the pack to maintain constant temps at the cell level.

The Tesla S will have the capability to switch out sub packs under the car. The switching of packs would be much simpler if there was no liquid coolant within each sub pack.

Anyone care to speculate as to whether or not the S might rely on air cooling ONLY in it's pack ??

George S. Bower
04-05-2010, 05:00 PM
These cells are usually 1C, the panasonic data sheet shows up to 2C for the 2.7Ah cells. 1C would discharge the cells in 1 hour.

Since they parallel 20 of these cells then the C rating of the pack (assuming the internal wiring can take it) is 20-40C. Maximun capacity is probably at around 10C (0.5C at the cell level).

http://industrial.panasonic.com/www-data/pdf2/ACA4000/ACA4000CE240.pdf

Each module is 25.2V and 58Ah, or 1.5kwh of capacity. 12 of these modules in series would give you 300v @ 58Ah, 18kwh and could deliver 580A with no issues (if the wiring can take it).. it could power a 233hp electric motor.

Looks like a very rugged design, and it can be fixed locally.

the above quoted from #2.

I don't think the last paragraph is correct. I believe it should read as follows:

"Each module is 25.2V and 58Ah, or 1.5kwh of capacity. 12 of these modules in series would give you 300v @ 58Ah, 18kwh and could deliver 58 amps with no issues. This equates to an 18 Kw max power rate. (1C power just like the original cell)."

Since you put the 12 modules in series they are still only good for 58 amps

No matter how you series or parallel 1C cells (based on power), you still end up with a 1C pack.

If the 18650 cells reference in the article really are only 1C, they will not work in an EV that requires a 5C power rating.

prowler
04-05-2010, 07:23 PM
this would not work if the [Tesla] cells were larger (lets say D size instead of AA size) and stored more energy.
let's not spread misinformation that Tesla uses AA size cells.

hermperez
04-06-2010, 12:01 AM
Tesla does not use AA sized cells, they use 18650 format cells.. slightly bigger than AA. The Prius uses "D" sized nimh batteries. or probably close to that size.

"I don't think the last paragraph is correct. I believe it should read as follows:

"Each module is 25.2V and 58Ah, or 1.5kwh of capacity. 12 of these modules in series would give you 300v @ 58Ah, 18kwh and could deliver 58 amps with no issues. This equates to an 18 Kw max power rate. (1C power just like the original cell)."

Since you put the 12 modules in series they are still only good for 58 amps"

You are exactly right.. but these cells are usually pushed harder than 1C, and for momentary pulses (< 10 secs) perhaps as high as 10C.

http://en.wikipedia.org/wiki/Tesla_Roadster

" The ESS contains 6,831 lithium ion cells arranged into 11 "sheets" connected in series; each sheet contains 9 "bricks" connected in series; each "brick" contains 69 cells connected in parallel (11S 9S 69P)."

"A fully charged ESS stores approximately 53 kWh of electrical energy at a nominal 375 volts and weighs 992 lb (450 kg)."

In Panasonic terms the Roadster uses 11 modules connected in series.. each module has 9 submodules connected in series and each submodule is made of 69 cells in parallel. I assume they are 2.25Ah cells CGR18650CG.. Panasonic rates these at 2.15Ah at the 1C discharge rate.

http://www.panasonic.com/industrial/includes/pdf/Panasonic_LiIon_CGR18650CG.pdf

Submodule = 3.6VDC at 148Ah (69 cells in parallel)
Module = 32.4VDC at 148Ah (9 submodules in series)
Total Pack= 356VDC at 148Ah (11 modules in series)

Total capacity = 52.8kwh

at the 1C discharge rate this pack can put out 52.8kw or 71hp for exactly 1 hour. We know the Roadster can do about 103 mph at this power level, thus traveling 103 miles before the battery is depleted. The speed is electronically limited to 125mph (consuming 83kw) and the motor is rated at 288hp. For the motor to achieve this power level it would have to discharge the batteries at the 4C rating.

So we know Tesla is using a 4C discharge rating on the 18650 cells.

George S. Bower
04-06-2010, 11:18 AM
Tesla does not use AA sized cells, they use 18650 format cells.. slightly bigger than AA. The Prius uses "D" sized nimh batteries. or probably close to that size.

"I don't think the last paragraph is correct. I believe it should read as follows:

"Each module is 25.2V and 58Ah, or 1.5kwh of capacity. 12 of these modules in series would give you 300v @ 58Ah, 18kwh and could deliver 58 amps with no issues. This equates to an 18 Kw max power rate. (1C power just like the original cell)."

Since you put the 12 modules in series they are still only good for 58 amps"

You are exactly right.. but these cells are usually pushed harder than 1C, and for momentary pulses (< 10 secs) perhaps as high as 10C.

http://en.wikipedia.org/wiki/Tesla_Roadster

" The ESS contains 6,831 lithium ion cells arranged into 11 "sheets" connected in series; each sheet contains 9 "bricks" connected in series; each "brick" contains 69 cells connected in parallel (11S 9S 69P)."

"A fully charged ESS stores approximately 53 kWh of electrical energy at a nominal 375 volts and weighs 992 lb (450 kg)."

In Panasonic terms the Roadster uses 11 modules connected in series.. each module has 9 submodules connected in series and each submodule is made of 69 cells in parallel. I assume they are 2.25Ah cells CGR18650CG.. Panasonic rates these at 2.15Ah at the 1C discharge rate.

http://www.panasonic.com/industrial/includes/pdf/Panasonic_LiIon_CGR18650CG.pdf

Submodule = 3.6VDC at 148Ah (69 cells in parallel)
Module = 32.4VDC at 148Ah (9 submodules in series)
Total Pack= 356VDC at 148Ah (11 modules in series)

Total capacity = 52.8kwh

at the 1C discharge rate this pack can put out 52.8kw or 71hp for exactly 1 hour. We know the Roadster can do about 103 mph at this power level, thus traveling 103 miles before the battery is depleted. The speed is electronically limited to 125mph (consuming 83kw) and the motor is rated at 288hp. For the motor to achieve this power level it would have to discharge the batteries at the 4C rating.

So we know Tesla is using a 4C discharge rating on the 18650 cells.

Perhaps Tesla picked their range target on the roadster because they only had access to 4C cells so they kept increasing the size of the pack until they got down to the 4C rating.

hermperez
04-06-2010, 11:58 AM
"Perhaps Tesla picked their range target on the roadster because they only had access to 4C cells so they kept increasing the size of the pack until they got down to the 4C rating."

Very astute observation.. many advantages to a large pack, one of them is that the need for liquid cooling goes away.. I wonder what studies they did on longevity, after all you never hear of Roadster battery failures.. Prowler?

I think that the Roadster combination of everyday 160 mile range and super car performance is very impressive.. range anxiety really goes away at that point. I hope the Leaf follows the lead and increases the range a little bit.

We know the LG cells in the Volt pack are limited to about 120kw or 160hp. The pack is 16kwh so that is a 7.5C discharge rating (the Volt itself uses a 7C rating). These are purpose designed automotive cells.

The Leaf 24kwh pack powers an 80kw motor, thus a 3.33C rating and no liquid cooling.

The question is what battery warranty will Nissan give?.. you know that there will be users that will suck out that 100 miles of range everyday and then crawl home.. total depletion of the battery every day!.. how long will that battery last?

prowler
04-06-2010, 06:10 PM
many advantages to a large pack, one of them is that the need for liquid cooling goes away.. I wonder what studies they did on longevity, after all you never hear of Roadster battery failures.. Prowler?
Not sure what you're saying - the Roadster pack is the largest one out there, and according to Tesla, it DEFINITELY requires liquid cooling (until the battery chemistry changes significantly).

Of the 3 components reported on the temperature display:

1) liquid cooled battery
2) ram-air cooled PCM (power control module)
3) fan-cooled motor

1) I never heard of a battery failure due to overheating (the two I heard of were both due to pinched power cables during assembly). The temperature readout typically stays a bar (of 7) below the others.

2) The PCM heats up first in aggressive driving (fast, pass, slow, pass, etc. - street racing)

3) The motor heats up quickest, and turns down the power, in track racing.

Tesla said that the next generation would have a liquid-cooled motor to enable full-time racing, but I haven't seen if it was definite in the S (wouldn't be the first time they changed direction before production, has anyone looked at the specs recently? it may be there)

The PCM shelf was "improved" in the second year (half the size, lower cost, supposedly "better") - I don't know if this is enough to balance it with the other components.

Does this answer the spirit of your question?

-SPARKZZ

prowler
04-06-2010, 07:16 PM
Tesla does not use AA sized cells, they use 18650 format cells.. slightly bigger than AA.
Just for fun, I looked up the size of the AA battery - about 14 x 48 mm vs the 18 x 65 for the 18650. This makes the 18650 more than twice the volume of an AA, which means . . . .

in order to replace 6,831 form factor 18650 batteries with AA's . . . .

it would take this many AA batteries:

Pi * 9 * 9 * 65
----------------- X 6831 = 15291
Pi * 7 * 7 * 48

OVER FIFTEEN-THOUSAND AA BATTERIES - WOW!!! (and I thought 6831 was a lot).

-SPARKZZ

hermperez
04-06-2010, 08:02 PM
did you know A123 makes LiFe cells in the 18650 size?.. only 1.1Ah capacity but can be discharged at up to 30A continuous.. they would make a hell of short range pack for the Roadster and no need for liquid cooling. You could make a pack that is 13 times smaller/lighter at the same power output, but the range would be much less since the total capacity would be 1.9kwh. You would need 520 cells at only 20kg. This would be the 1/4 mile drag racing pack.