For Model S, the inverter is now built into a small cylindrical unit directly attached to the motor. Do you still think there will be recharging circuitry in there?
Yes. They have switched to water cooling, that would make it possible to build it much more compactly.
I wonder if the Model S will switch to a completely separate box for charging compared to controlling the motor.
Yes, there is component redundancy for regen and recharging from line current, but there may be good reasons to keep them separate.
That may be, but I would be very interested in knowing what those reasons are. One credible alternative would be that both PEM and charger is sourced from subcontractors, and no-one has thought of coordinating the efforts. Or that the project is in such a hurry that there is no time to do it right. Or there might be technical reasons that I'm unaware of.
I've long since stopped laughing when reading the comic "Dilbert", though sometimes I snigger a little. I am Dilbert, or rather I used to be. I have seen so many disasters caused by bosses stubbornly refusing to admit mistakes, demonstrations of decisiveness by making rash decisions based on faulty assumptions, failures to listen to science-based advice from the guys who actually create the product and just plain boneheadedness. Dilbert is more or less a documentary. I once had a CEO who insisted that we could not buy computers that did not waste our time because we didn't have enough money. Then he convinces the board of directors that it would be a bright idea to buy the smoking ruins of a failed competitor for USD 30 million, borrowed at 16% interest rate, without even getting hold of their products and having our engineers evaluate them beforehand. Their products were utterly useless, they had nothing of value. I saw it coming, so I got another job just before the company went belly up.
And there are so many spectacular failures in the past - e.g. the horrible ATM compromise of 48 byte packet size which is not a power of two, so the packets don't fit in the cache lines of the routing hardware. The list goes
on and on.
So when something comes along that I think seems wrong or shortsighted, I will assume that it is, particularly if I try to dig up more information and I can't find anything substantial.
Currently, the Americans insist that their proposal is the best, so do the Japanese and the EU. Neither the Japanese nor the Americans seem to take any notice at all that the European grid is different, they keep insisting that their own solution is perfect, seemingly unaware that the physical grid is wired differently. I've seen similar situations before.