ChadS
Last tank of gas: March 2009
There are many interesting ideas in this thread. Lots of possibilities for the future - that makes me feel good. I don't know the real answer, but it's fun to think out loud:
Like Citizen-T, I'm trying to look at this from the perspective of what problem Tesla is trying to solve. We already have free solar-powered long distance for life; that is something that gas cars can't provide. But naysayers argue that it still takes longer than gas - and that's true. That's the ONLY real disadvantage to driving electric. So the key aspect of this swapping solution is that it has to be faster than gas. And of course it can't have any disadvantages compared to gas; though it can sacrifice some of the electric advantages. So, for example, it no longer has to be free or solar-powered; people that value those can stick with Supercharging. Both would be great, but as long as it's not more costly or dirtier than gas, a solution can be used as long as it's faster than driving gas.
I assume Tesla's metric will be an LA-to-NY trip. In a comparable gas car (say Audi A8, base engine, 28mpg hwy) that would cost about $400 in gas each way. So the solution can cost up to $800 for a round trip of ~5500 miles. The A8 has a 23.8 gallon tank; adjusting for real-life driving and filling up well before empty, the gas car would probably make about 12 stops. Using Lloyd's 12 minutes each, that's 144 minutes. So that's the refueling budget for the trip. We need a solution that will take less than 144 minutes on an LA-NY-LA trip, and cost less than $800.
Swapping 85kWh packs just doesn't seem like it's going to do it. You have to swap too often; it's going to go over the time limit. Plus you need serious automatic machinery (or employees and heavy-duty tools) in many, many places to make that workable, so it's going to go over the cost budget as well unless they've figured out a way to get a lot more money out of idle packs than I can see.
When Tesla announced the pack sizes, they said they could build one bigger than 300 miles, that just seemed like a sweet spot. On a recent call when asked about a hypothetical 400-mile battery, Musk pointedly asked "what about a 500-mile battery"? Sure, one of those batteries may cost more to build (then again, maybe not if it's consumable rather than rechargable). But you have to stop less often, so less time is spent swapping. You can also save money by having fewer swapping stations; in fact given that they already say >80% of their customers are within 50 miles of a service center, they could say you only do the swaps at service centers. Question: Do employees at each service center have (or will they get) tools to allow them to do quick battery swaps? I don't know, but that's a key element to this plan. (Even if they do, this is not ideal as service centers are not open 24 hours. Service center build-out is also not complete yet, but neither are robotic swapping stations - and they want more service centers anyway, so it's already baked in. Not being robotic also means there are likely appointments involved - though other than the first one, perhaps the car makes them for you based on where you are?).
Even with a "500-mile" battery, in real life you'd want to swap about every 350 miles. That's 15 stops. Is that going to be faster than 12 12-minute gas stops? Erm. I'd sure hate to claim that; but maybe they can with some Tesla math. Maybe they'll say it's only seconds faster now; but in years to come, as the batteries get bigger, the time advantage will widen.
Can they do it for under $800? That depends a LOT of what the batteries cost to build (at least this theory doesn't have to pay for a lot of robots). My rough guess is that employee time (internal cost, not normal billing rate) would be about half of that. Looking at expected pack lifetimes...hmm, that doesn't leave a lot. Well, Tesla is the king of low battery prices. And maybe they really are planning on integrating some cheap metal-air thing inside the case to get this density. Maybe they can get more than I think I can by using the batteries for grid balancing when not in use. Maybe they figure gas prices will rise and they'll soon be able to charge more. Or maybe they're just willing to take a loss at first, knowing that battery prices will come down - and that few people will every really USE this (few people take trips this long; and many of them will like the free solar-powered Superchargers), so losses will be limited. It's more about having the ability to say it can be done, so nobody has any excuse to not consider an electric car. Early losses could be a marketing expense.
I assume they would bill the user per swap (that's what takes up employee time), and per day (need to get those big packs back for others), and possibly per mile (they are saving on wear-and-tear on their own battery, after all - but most people will be driving a LOT, so they could just assume that and bake it in to the other fees).
I dunno. There's a lot of "ifs" in there...if Tesla is able to build a 1,000-mile battery in the Model S pack form factor at a reasonable price (using who knows what technology), this may start to make sense. A 500-mile pack doesn't seem quite big enough; at least not for me to feel comfortable making the "faster than gas" claim. But it's getting really close...
Like Citizen-T, I'm trying to look at this from the perspective of what problem Tesla is trying to solve. We already have free solar-powered long distance for life; that is something that gas cars can't provide. But naysayers argue that it still takes longer than gas - and that's true. That's the ONLY real disadvantage to driving electric. So the key aspect of this swapping solution is that it has to be faster than gas. And of course it can't have any disadvantages compared to gas; though it can sacrifice some of the electric advantages. So, for example, it no longer has to be free or solar-powered; people that value those can stick with Supercharging. Both would be great, but as long as it's not more costly or dirtier than gas, a solution can be used as long as it's faster than driving gas.
I assume Tesla's metric will be an LA-to-NY trip. In a comparable gas car (say Audi A8, base engine, 28mpg hwy) that would cost about $400 in gas each way. So the solution can cost up to $800 for a round trip of ~5500 miles. The A8 has a 23.8 gallon tank; adjusting for real-life driving and filling up well before empty, the gas car would probably make about 12 stops. Using Lloyd's 12 minutes each, that's 144 minutes. So that's the refueling budget for the trip. We need a solution that will take less than 144 minutes on an LA-NY-LA trip, and cost less than $800.
Swapping 85kWh packs just doesn't seem like it's going to do it. You have to swap too often; it's going to go over the time limit. Plus you need serious automatic machinery (or employees and heavy-duty tools) in many, many places to make that workable, so it's going to go over the cost budget as well unless they've figured out a way to get a lot more money out of idle packs than I can see.
When Tesla announced the pack sizes, they said they could build one bigger than 300 miles, that just seemed like a sweet spot. On a recent call when asked about a hypothetical 400-mile battery, Musk pointedly asked "what about a 500-mile battery"? Sure, one of those batteries may cost more to build (then again, maybe not if it's consumable rather than rechargable). But you have to stop less often, so less time is spent swapping. You can also save money by having fewer swapping stations; in fact given that they already say >80% of their customers are within 50 miles of a service center, they could say you only do the swaps at service centers. Question: Do employees at each service center have (or will they get) tools to allow them to do quick battery swaps? I don't know, but that's a key element to this plan. (Even if they do, this is not ideal as service centers are not open 24 hours. Service center build-out is also not complete yet, but neither are robotic swapping stations - and they want more service centers anyway, so it's already baked in. Not being robotic also means there are likely appointments involved - though other than the first one, perhaps the car makes them for you based on where you are?).
Even with a "500-mile" battery, in real life you'd want to swap about every 350 miles. That's 15 stops. Is that going to be faster than 12 12-minute gas stops? Erm. I'd sure hate to claim that; but maybe they can with some Tesla math. Maybe they'll say it's only seconds faster now; but in years to come, as the batteries get bigger, the time advantage will widen.
Can they do it for under $800? That depends a LOT of what the batteries cost to build (at least this theory doesn't have to pay for a lot of robots). My rough guess is that employee time (internal cost, not normal billing rate) would be about half of that. Looking at expected pack lifetimes...hmm, that doesn't leave a lot. Well, Tesla is the king of low battery prices. And maybe they really are planning on integrating some cheap metal-air thing inside the case to get this density. Maybe they can get more than I think I can by using the batteries for grid balancing when not in use. Maybe they figure gas prices will rise and they'll soon be able to charge more. Or maybe they're just willing to take a loss at first, knowing that battery prices will come down - and that few people will every really USE this (few people take trips this long; and many of them will like the free solar-powered Superchargers), so losses will be limited. It's more about having the ability to say it can be done, so nobody has any excuse to not consider an electric car. Early losses could be a marketing expense.
I assume they would bill the user per swap (that's what takes up employee time), and per day (need to get those big packs back for others), and possibly per mile (they are saving on wear-and-tear on their own battery, after all - but most people will be driving a LOT, so they could just assume that and bake it in to the other fees).
I dunno. There's a lot of "ifs" in there...if Tesla is able to build a 1,000-mile battery in the Model S pack form factor at a reasonable price (using who knows what technology), this may start to make sense. A 500-mile pack doesn't seem quite big enough; at least not for me to feel comfortable making the "faster than gas" claim. But it's getting really close...
Last edited:
