I'd like to state a few comments on your post.
In general I agree with your points. The horrible "efficiency" of ICEs and the splendid performance of electric machines are definitely on our side. But most people don't get that. If people complain about fuel prices and you tell them that their beloved cars throw ~ 75% of the valuable fuel out of the tailpipe, they won't believe you. This has to be communicated and marketed somehow.
As far as I can tell, the estimate of 33 kWh is reasonable. This is a massive bunch of energy.
I think that the batteries are a strong limit. They will cost very much. Hence we have to choose them wisely.
In my opinion, the figure of 160 Wh/kg is not conservative but rather optimistic.
The figure certainly holds for laptop batteries. But there is a reason why no laptop manufacturer gives more than 1 year or even 6 months of warranty on their batteries. Those lithium polymer batteries manage around 500 … 700 charge/discharge cycles under good conditions before their capacity drops very rapidly. This is OK for a laptop. My laptop tells me that it has 50 minutes worth of battery charge left. Five seconds later, it is gone because the battery is already empty.
Using batteries in an EV cannot be filed under "good conditions" but rather under "abuse". And replacing a 33 kWh battery every other year, because you cannot reach your workplace anymore, won't be affordable.
There are batteries that have been engineered specifically for automotive use. First of all, they are much safer than the laptop-type batteries (imagine your car burning like a sony laptop). Secondly, they can be charged very quickly (like drinking a coffee while your car is recharged).
Unfortunately these are expensive and they have a lower energy density, which means more weight for the batteries.
At the moment I know about three manufacturers of these batteries:
- Thundersky [1], a Chinese company. Overseas orders are handled by Everspring [2]. Their new TS-LFP series batteries are claimed to sustain more than 1000 cycles with 80% of the initial capacity. Some older products of this company turned out to be not worth their money. I haven't found any reviews of their new battery series yet. But at least Everspring has a price list. Without shipping, the bare cells needed to assemble a 33 kWh battery would cost alone 20000 USD. There is a discount for volume orders though.
- A 123 Systems [3]. They have a huge contract running in order to supply the batteries for GM's upcoming plugin hybrid vehicle (and for Mercedes, Think!, Renault, … ). They might be able to cut down the battery prices by massive mass production. But I don't have any figures of their current prices. The batteries are claimed to sustain more than 3500 cycles with 80 % of the initial capacity ( 150 km per charge -> 525000 km of lifetime ). I have used a small evaluation kit from them and I also heard and read from others that this company's products hold their promises.
- AltairNano [4]. They have a handy 35 kWh module with integrated battery management in production. It is claimed to have a lifetime of 10000 - 15000 cycles, which means that the battery would probably outlive the car. The battery module currently undergoes a test in a fleet of leased EV pickups. No independent tests and no prices yet.
The downside of these batteries is their lower energy density of about 100 Wh/kg for all three manufacturers. This makes for roughly 330 kg for a 33 kWh battery.
A possible advantage is the high power density of around 5 kW/kg, which would make it possible to use smaller batteries (e.g. 2 kWh -> 100 kW) for testing / evaluation purposes without the need to risk a lot of money just for testing and without losing the full power of the drivetrain. Scaling the battery to personal needs / available budget is possible.
So the choice of the right battery will be a complex tradeoff between costs, lifetime, weight, range, reliability and safety. I think we should therefore test different battery types and design a scalable battery module.
I'd like to make another point - about the engine itself. The nominal power of an ICE is normally given in terms of maximum power that is only valid for one single engine speed. The nominal power of an electric motor is normally given in terms of continuous power over a wide band of motor speeds. And if properly cooled, electric motors can be significantly overloaded for short periods of time without taking damage. A 40 or 50 kW electric motor might be sufficient to "feel" like the original engine. A 40 kW electric engine from Brusa [5] brings in another 53 kg of weight.
[1] http://www.thunder-sky.com/home_en.asp
[2] http://www.everspring.net/txt/product-battery-pricing.htm
[3] http://www.a123systems.com/#/applications/phev/pchart2/
[4] http://www.altairnano.com/markets_energy_systems.html#innovative
[5] http://www.brusa.biz/e_datasheets.htm
