Of all the lithium chemistries available, we’ve found the LiFePo4 based lithium cells to offer the best combination of safety, stability, cycle life, power, and energy density. Though slightly less capable in terms of energy density per kilogram, the trade offs of much more stable and safe operation, and the very much longer life expectancy of these cells have made them a favorite.
Of the LiFePo4 cells, the A123 cells are often offered as the mark to beat. We have not found it so. We early on identified the Sky Energy large format prismatics in 100Ah and 180Ah sizes as much more convenient to build into an electric vehicle and offering a flatter discharge curve and long life when compared to all other cells on the market.
Sky Energy received a large investment from the China Aviation Missile Academy, a government entity, and became the China Aviation Lithium Battery Company (CALB). They continued production of the Sky Energy design as the SE series of cells which we have used successfully in a number of builds.
In June, 2012, CALB is introducing their new CA series of cell, notable by its newly designed grey case. Commonly referred to as the “grey cell” this introduction has been eagerly awaited since its announcement well over a year ago.
EVTV has developed a reputation for ruining the party at Lithium battery celebrations by actually performing direct first party testing on cells – often to destruction.
In this comparison, we do NOT perform any direct testing but provide a comparison of the SE and CA series cells based entirely on data provided by CALB, which we’ve found in the past to be quite reliable. Indeed, if we have a criticism, it is that CALB tends to hide their light under a bushel, seeming almost hesitant to make claims for their product.
In this fashion, they have very modestly provided us a spreadsheet of almost puzzling data that on further examination, would seem to hide a stunning advance in LiFePo4 cell quality. We are not at liberty to share the spreadsheet, but rather here present our best take on it, with excerpted graphics.
We do of course plan an entire series of episodes presenting EVTV test data to support or refute these claims. But if past experience holds true, we look forward to discovering performance quite beyond their claims, and we are genuinely excited to examine this new incremental advance in cell chemistry.
It should be noted that it is our intention to begin distribution of these cells through the EVTV online store at our web site. And so we may be viewed as something more than a disinterested party in the matter.
That said, note that we chose the battery, the battery didn’t chose us. Indeed, subsequent to our rejection of the head of their marketing groups desire to establish a U.S. marketing slogan – CALB – Cost A Less Bill, we have never successfully had a relationship with CALB for advertising or sponsorship at all. Indeed, our humorous response to this suggestion must have lost something in translation as he was pretty seriously offended by it. None intended of course.
Despite this, we’ve remained enthusiastic over their SE series and have sent many conversions their way. With the advent of this new CA series, and the ongoing difficulty our viewers have in finding reliable trusted sources to purchase lithium cells from generally, we have painfully arrived at the conclusion that we not only have to sell the batteries ourselves, but stock them in our facility at a somewhat enormous capital expense.
We intend that the immediate impact of these cells will simply be better cars, better conversions, and better results for our viewers. As they demonstrate these very desireable vehicles to their friends, neighbors, and relatives, we believe the demand for electric vehicles will be accelerated and the adoption of electric vehicles in America will within a few short years comprise over half of all vehicle sales.
In the short and the long of it, the story of DIY conversions in America since 2008 has been all about one thing – the enabling power of the batteries. We believe we enjoy SUPERIOR batteries to those used by Tesla, Nissan, and General Motors in every respect. These cells have allowed electric cars to cross the threshold of viability, moving DIY conversions from semi-interesting science projects, to actual drivable useable automobiles, in one smooth move.
The introduction of the CA series represents an incremental, but quite important advance in these cells, and consequently the cars that use them. They really do NOT offer any particular advance in energy density and thus range. But we believe they will provide greater power, longer life, more consistent capacities, better thermal characteristics, and much improved cold weather performance. This represents a kind of maturation of the chemistry, and we find these kinds of improvements really more important than range. Our cars go further than we do now. But longer life, better power output, less heat, better performance in cold weather are all extremely valuable characteristics at this point.
One feature we expect is not provided in the test data, but we intend to test it. Part of the process, and the delay, in introducing the new grey cells was some pretty serious behind the scenes advances in the scale of cell manufacture through the building of newer, larger, and more automated production lines for these cells, enabled by the financial backing received from the government.
This would appear at first glance to hold nothing for the end user/purchaser of the cells.
But one generally overlooked aspect to cell management is the fact that we don’t use cells in cars at all. We use STRINGS of cells in our cars. And our range, and our performance, and our power output can never exceed, to even the slightest degree, the abilities of the LEAST capable cell in the string. You can have 35 cells capable of 10,000 amps of power and a 1000 mile range. If you have one cell limited to 200 amps and 30 miles, your car will never travel more than 30 miles, nor accelerate at a rate greater than 200 amperes can deliver.
I would much rather have a string of cells rated at 180 Ah where each and every cell showed exactly 170Ah capacity, than to have a string where all the cells actually tested at 200Ah, except for two at 165Ah. A more consistent string is easier to charge and discharge safely and effectively and will provide better range and performance simply by being consistent.
So one of the less obvious aspects of the CA cell is CONSISTENCY – what do we actually get out of the box in terms of capacity from one cell to the next. And with these larger, more automated, more efficient assembly facilities, we would expect to see a more consistent product – leading to much improved cell strings in our cars. We intend to actually test this against a box of undisturbed SE cells we have on hand.
And now the bad news. At this point, the erosion in cell prices appears to have not only stopped, but reversed in some ways. CALB, now well financed, has been very stuck on pricing and we are seeing some firming in the other companies as well. Some second tier cells were going as low as $1 per Ah for a few months, but that appears to be over and if anything cell prices appear to be rising slightly.
While we don’t see this trend holding long term, in the immediate future we seem to have formed a “bumpy bottom” in LiFePo4 cell prices. We would love to see a dramatic decrease to half the price in the near future. But it appears to be wishful thinking at this point. In increasing capacity to meet demand, the expenses incurred appear to have put battery manufacturers in a kind of lithium vise. They face ever increasing over capacity and competition, with an inability to deliver cells at any price lower. And so we see their marketing efforts turn hopefully and expectantly to large purchasers for Wind farm and grid power applications. It remains to be seen if this works out for them, but the DIY EV market appears to not offer sufficient clout to drive discounting. We face more likely abandonment than price decline.
Again, we intend further direct testing in the future. But this should provide a peek at what is claimed for these cells on introduction.
HIGH RATE DISCHARGE PERFORMANCE
All LiFePo4 cells have a stated capacity usually in ampere-hours (AH). We currently carry the CA180FI and CA100FI cells that feature a stated capacity of 180 Ah and 100Ah respectively.
This amp hour rating is measured at a defined rate of discharge of 0.3C. The 100Ah cell would provide 100 ampere hours at a rate of 30 amps. The 180 Ah cell would provide 180 Ah at 54 amps.
At higher discharge rates, the capacity available is decreased. In the SE series of cells, you could expect at a discharge rate of 3C or 300 amps for the 100 amp cell, a capacity of 94% or 94Ah. In the CA series, this is much improved at 97.7% capacity at 3C. Your 100Ah cell will still have nearly 98 Ah at 3C compared to 0.3C.
The charts below compare the CA60FI and the SE60AH cells for high rate discharge performance.
All LiFePo4 cells exhibit a rise in temperature when discharging. The internal process of intercalating lithium ions has certain inefficiencies that show up as heat. At low discharge rates this is of course minor but as the discharge rate increases, so does the heat gain.
At 3C, the SE cell series shows an increase in temperature of 22.6C.
This is actually pretty good. The cells have proven to need no particular cooling at all. They are quite hardy up to about 65C where one of the electrolytic solvents begins to deteriorate.
So at ambient temperatures of 30C for example, an increase of even $25C leaves us at the 55C level. And they cool rather quickly under lower discharge rates. So even under heavy acceleration, we’ve found cooling of these cells just not necessary. Indeed, they perform better up around 45-50C.
But thermal gain is a sign of an efficiency loss. For the CA series, this is dramatically decreased to a temperature increase of 8.6C. Internally, the CA series is just much more efficient than the SE series. And this points to even more durability and likely longer life under high accelerations and power demands.
One of the more important factors in LiFePo4 cells is their ability to deliver power (current) on demand. We usually relate the number of amperes a cell can deliver with the corresponding decrease (sag) in voltage. More properly, this is measured as POWER DENSITY in Watts per kilogram at various levels of state of charge (SOC)
This chart compares the power density of the SE40AHA, the SE60AHA and the CA60FI cells.
At a 50% state of charge, the SE40AHA shows a power density of 890 W/kg while the SE60AHA shows 779 W/kg. The CA60FI shows 1322 W/kg – a 70% increase over the SE series.
We were able, for example, to discharge the SE180Ah cells 1000Amps with about a 22% voltage sag when testing the Speedster Redux with a Soliton1 controller. This would imply that the CA series cells could deliver 1700 amps in the same situation.
The spec sheets on the new CA series cells have thus far been spotty, conflicting, and incomplete. The SE cells we used to rate at 8C at EVTV. We would conservatively claim 12C on the CA series from this data, and we will attempt some sort of testing as soon as practicable. This is a little difficult on these large format cells. Even the 100 Ah cell would thn require a 1200 amp load to prove this.
COLD TEMPERATURE CAPACITY
One of the disadvantages of LiFePo4 cells compared to other lithium chemistries is a rather sorry performance in cold weather. While not nearly as debilitating as it was in the old lead acid battery era, the decrease in capacity of LiFePo4 cells is very much a factor.
At a temperature of -20C, and a discharge rate of 0.3C, the SE series of cells will provide 71.9% of capacity. Your 100Ah cell will provide a discouraging 72 amp-hours at that temperature. This is of course even worse at higher discharge rates.
The CA series cells provide a dramatic improvement in cold weather performance with 87.49 % of the original capacity. Your 100 Ah cell now provides a little better than 87 Ah at -20C or -4 degrees Fahrenheit at a 0.3C discharge rate.
One of the most important advantages of LiFePo4 cells compared to other lithium chemistries and certainly with Pb chemistry cells is their very long life. We measure this life in the number of expected charge/discharge cycles to 80% discharge and until the cell exhibits 80% of its original capacity..
This is projected by doing several hundred charge/discharge cycles at 1C and to 100% depth of discharge and extrapolating the data to the point where the cell exhibits 80% of its original capacity. This calculation is also improved by assuming an 80% depth of discharge instead of the tested 100%.
At 290 cycles to 100% at 1C, the SE series cells show about 85% of original capacity. The CA series has upped that to about 91%. This represents a huge increase in cycle life for the CA series cell.
Assuming that this had extrapolated to a 2000 cycle life at 80% DOD, the new cells would imply 3300 cycles. For 3000 cycles at 70% DOD, the CA cells should see 5000 cycles
The result appears to be a bit of a huge leap in battery performance at a very minimal increase in price. So while prices have not really come down, we’re getting more battery per ducat in a number of very interesting ways.
After a period of relatively static lack of movement in battery cells over the past year, we are enormously excited by this new introduction. After seeing this data, we are attempting to triple our investment in stock on hand on the assumption that these cells will be very much in demand for the foreseeable future.
We intend to ship all cells with our braided cell strap with Nordlock washers