My predictions, fwiw. I think there are two timeframes where it would make sense for Tesla to update the current Model S battery packs: when introducing the Model X and when introducing the Model 3. The Model 3 is a slam dunk for a refresh driver. By then, battery tech will have improved enough that it makes sense to switch to different cells and even different chemistry. They'll need to do that for the 3 to meet their cost, weight, size and capacity goals. If they need a smaller (dimension-wise) pack for the 3, they'll have two packs - one for the 3 and one for the S/X - and each pack will have different capacity points to choose from. The 3-pack will ship first, followed by a redesign/refresh of the S/X packs unless Tesla has grown to the point where they can handle refreshing both at once. Possible but I'm not going hold my breath. The cheapest way to bring out a new roadster would be to use the 3-skateboard and pack for a new Roadster. The problem there is that the Roadster needs to be shorter than the 3 so you either need a shorter skateboard with smaller pack for the Roadster or let the Roadster and 3 both use the smaller pack which leaves range on the table for the 3. Using the smaller pack makes sense for the 3 only if the larger pack is so expensive that it makes the 3 too expensive. The second possibility is a refresh timed to ship with the X. If they do that, that pack should work with the S. I'd love for that to happen but have no idea if it will. It depends on the improvement they can get in range/cost vs the cost of moving to the new battery (qualification, production line changes, etc.). I'm guessing there isn't enough range improvement to justify the cost but that's just my guess. The engineers may have better things to work on.