Is that right? Why is there an energy loss when cells are connected in series compared to cells connected in parallel? (Seriously - I'm asking - I'm curious and I want to know!)
Lets say you have two power sources. Both are 50v and 100amps.
If you put them in series you get 100v @ 100 amps or 10000 watts
If you put them in parrallel you get 50v @ 200amps or 10000 watts
Now if you give them each a motor which requires their respective voltages which do you think would last longer? A 100amp battery pack or a 200amp battery pack?
oobflyer is basically right. Arranging cells in series instead of parallel would increase the pack voltage, decrease its amp-hours, and not change the capacity.
***
Burton:
Let's use Zero's existing modules. ZF2.8 module = ~100V 25Ah (2.5 kWh).
You could put two modules in series for ~200V 25Ah (5 kWh).
You could put two modules in parallel for ~100V 50Ah (5 kWh). (This is Zero's ZF5.7 pack)
Let's say you hook up a motor and controller to the pack and draw power out at 1 kW. Both packs will supply the power for approximately 5 hours; the series pack is supplying 200V 5A to the motor controller, and the parallel pack is supplying 100V 10A to the motor controller.
If instead of drawing a fixed power, you draw a fixed amps - say 10A from both packs - then the 200V pack will supply twice the power (200V 10A = 2 kW, vs 100V 10A = 1 kW) for half as long (25 Ah / 10A = 2.5 hours, vs 50 Ah / 10A = 5 hours).
***
oobflyer:
The reason why Zero probably configures their packs with modules in parallel for higher amp-hours, lower voltage instead of in series basically comes down to component availability and design cost.
Component availabilityA higher-voltage battery means that you need a higher-voltage charger, motor controller, and motor.
The motor is pretty easy to make - and Zero designed their own motor for the 2013+ bikes.
The motor controller is supplied by Sevcon, and Sevcon's low-cost line of Gen 4 Size 2/4/6 motor controllers are only available up to about 150V, and (I recall hearing) even that was only available recently. Zero needs access to part samples in order to build prototype bikes; if the part samples don't exist then they can't use them. Maybe Zero will build their own motor controller someday; but an OEM-quality motor controller is
very expensive to design.
Sevcon does make a high-voltage motor controller in the Gen 4 Size 8 and Size 10. They're very expensive, multiple thousands of dollars. Harder to build a bike to a price point. Other manufacturers make OEM-quality high-voltage controllers, but again they're very expensive.
The charger isn't too difficult to design for higher voltage - the 2013 Zeros just used Meanwell LED supplies in series. (2 series, 2 parallel for the S and DS). They could double the voltage by using 4 LED supplies in series. However, the same thing about component availability applies; Zero's either gotta use whatever they can get samples of or they've got to build their own. $$$
Design costBasically comes down to safety. Making things safer takes more time, which means more money.
Luke aka liveforphysics is an electrical engineer that works for Zero, and he's very active in the electric bike builder community. He's written several rants (
rant 1,
rant 2) about the dangers of high-voltage (700V) systems.
As you mentioned, it's the current that does things (including killing you). Voltage accomplishes nothing useful other than enabling current to flow. The more voltage, the more things that didn't let current flow before start flowing current. This includes things like almost invisibly scuffed motor winding enamel (that may never have arc'd catastrophically at 350v), the humidity collecting from thermal cycles under your insulation boots over lugs, the tiny solder flux residue impurities that didn't get entirely cleaned off before the conformal coating was applied, the silicon jacketed balance tap lead that was bent sharply once, internally tearing the jacketing leaving a thin place that may never have arc'd and exploded at 350v, etc.
In this case he was talking about a 700V battery; but it's still true that a 300V battery and powertrain requires more design safety than a 100V battery. From the battery perspective it simplifies things if they stack their modules in series instead of parallel, since they can design and validate a 100V module .. but the overall battery is more expensive (for 300V with 3 100V modules, you have 3 BMS instead of 1, or you have 1 BMS with 3x the number of taps) and any type of failure is more expensive (both in terms of human life and in terms of PR damage).