I don't think tiny A123 cells should be used for car-sized packs unless you're drag-racing or are mass-producing the packs. Otherwise it's way too much effort and time and you just won't be approaching the charge or discharge rates it's capable of.
which means you won't be stressing the cells.... which means longer life and less capacity loss after time. Now there are benefits of both, but just because you're not USING the high discharge/charge rate doesn't mean that they're automatically not as good... Not putting the cells to their limits is a VERY GOOD thing. Plus, the A123 cells are higher energy and volumetric density than prismatics (from ones I've seen). Charging can be done faster on smaller cells (like a123), on prismatics its 1/2C or thereabouts. Also, if one cell goes, you can have a fuseable link to take that one out of the pack, try doing that with prismatics. Voltage sag on prismatics compared to smaller paralleled cells can be much higher.
The only benefit would be a longer lifespan than regular LiFePO4 but it'll also be more expensive just for the raw cells, I'm pretty sure. Typical ebay prices are $10/cell or $1.32 / Wh. Then you have to get a BMS and charger separately that can handle your needs.
Ebay is way overpriced, you can get them much cheaper.....read some of the links people posted... one of them was a link to another forum with someone selling a123 packs of 50 cells in a 3.3V 115Ah configuration with welded for $325.... thats $6.50 a cell... as I also stated. and it ends up being about $1.16 a Wh. hovering just above the $1 a Wh mark.
The charger/bms argument is null... you need that either way... its required reguardless of what kind of LiFePo4. BMS for the same voltage is the same.... same number of cells in series (all parallel batteries self equalize within that parallel pack, so only one BMS channel is needed).
Tesla uses old-school lithium-ion 18650s for power density but their pack is highly engineered. It's actively liquid-cooled even when the car is off (using power from the pack itself as long as it remains > 90% full) and is designed to handle a cell exploding without setting off a chain reaction.
they use li-cobalt... unstable and volitile if overcharged/undercharged. High density, but at what cost? Also a 5 year pack life...you're talking apples and oranges. This is LiFePo4, much more stable, no cooling needed in most cases, and safer than most other chemistries.
Anyway, if I were you, I'd go with a large-sized LiFePO4 that people are already using in cars and motorcycles so we know what the performance is like and the level of support the company offers. There are many bad companies or vendors.
I think the long and the short of it, you're not basing the choice for a pack on anything that matters... so what if it takes more to assemble, thats not really a requirement is it..... So what if others are using thundersky and hi-power, A123 has good peformance and support as well. And who cares if there's more cells, you can fit more into your vehicle in a smaller space and it'l weigh less.
What he needs to do is give us the peak power levels needed from the battery pack. If he can achieve that with prismatics
without stressing the cells (staying below 1-2C continuous)... then prismatics might be the best choice. If needs
continuous current above 3C, then a smaller format battery might be a better choice, because each cell is at a lower C rate, but the overall pack can discharge tons of amps, hence more power. The voltage sag is much less. Sure more effort to assemble, but its a pack that is matched to his requirements.
You really can't just blindly chose batteries, they need to match the requirements of what they're being used for, thats the ONLY thing that matters as well as cost. Labor might matter to some of the lazier people here.... but I'd rather assemble 2000+ cells and have it last years, than to assemble 40 and be crying because i have to replace it in a year.
