So, I'm shopping for new batteries, given that I'll need them pretty soon. Price is relevant to the engineering side too, particularly since battery costs dominate EV running costs. I got very excited when I realized that the small size Kokam lithium polymer cell is just about the right size to fit in my battery case, except that it's very thin. There's enough room (I think) to put 8 cells of the 11 A-h size in series (29.6 V nominal) times 3 in parallel, for a total of 33 Ah. That's about triple the capacity of my lead-acid battery pack, with very good power and less weight than the original. Price is the kicker: ProEV, an SCCA electric race car sponsored by Kokam, gives this formula for price in US dollars: 1.4*(cell capacity in A-h)*3.7V, which works out to about US$1400 for my ebike pack. Ouch! New lead acid batteries will cost about C$80, much much less. But they don't go as far, or for as many cycles. So how do they all compare, apples to apples?
Here's a table:
| Chemistry | Wh/USD | Wh/l | kg/l | Cycles |
|---|---|---|---|---|
| PbA (T105) | 15.5 | 106.9 | 2. | 300 |
| PbA | 4.3 | 104 | 4.4 | 300 |
| NiMH | 0.90 | 298 | 300? | |
| LiIon | 1.2 | 174 | 1.4 | 500 |
| LiPoly | 0.71 | 390 | 3.0 | 500 |
| H2 @ 150 bar | 405 | 1 | ||
| Gasoline | 14000 | 9700 | < 1 | 1 |
Short summary: lead acid is cheapest battery, even if you have to replace the pack more often, but for six times the price per km, you could go four times as far with lithium polymer.
It's not as simple as that, though. My "Watt-hours" are based on nominal capacity, where the batteries are discharged slowly, over 20 hours, a so called C/20 rate. You get less capacity if you use them harder, and how much less depends on the battery. Lithium polymer is quite good at high rates, conventional lead-acid not so good. Eg., for a 12 A-h rated SLA, I'd expect to get only 8.4 A-h out if discharging it in an hour (a 1C rate), a fairly plausible rate in ebike use. Cycle life is all fairly approximate, I've given life at 80% DOD. The Lithium ion prices seem pretty good, less than I expected, only triple lead acid. On the other hand, they aren't nearly as good as lithium polymer in terms of volumetric energy density. (The look pretty good if you care about weight, but lithium polymer packs the same energy into the same weight into less volume.) They are ThunderSky cells, with prices from importer Metricmind, made in China, who recently announced a 50% price increase. It's that development thing again. So every number in this table could be substantially off.
If you're comparing cost to gasoline, I think only lead-acid is competitive (about 50% less, according to a long-term Connecticut study done at 90's prices), though NiCd may be also, since for the extra cost you get vastly increased cycle life. I'd guess LiIon would cost more, though it really matters how long your cells last, and what price you pay for gas. Still, LiIon is good enough to build a car with a 200 mile range (Steve Green's Jester), and lithium polymer could do substantially more again. If you believe that lithium polymer prices can go lower, and their longevity increase, then you'll be excused for wondering what the fuss about fuel cells is.
Added. The PbA energy density is maybe a little high. One sources quote 40 Wh/l, which seems low. The specs on 13 Ah Hawker Genesis give 64 Wh/l at C/1, 88 Wh/l at C/20, and some NREL presentation uses 75 Wh/l for PbA, 100 for NiMH, so this is consistent. The NiMH spec I added is based on a C$4 2500 mAh 1.2V Energizer AA.
I added the Trojan T-105, which is a very cheap 6V golf cart battery, of a conventional flooded design, used in eg PV and boating. These numbers are again C20, and you have to need a 220 A-h battery.