Yeah but the load peak when everyone arrives home/at work is what's dangerous for the grid. Even 10k cars at only 10 kW of charging power are a peak load of 100 MW. 100k cars are 1GW - which means the city of Munich with 700k cars has a load potential of 7GW at the minimum.

Either we get a grid that's intelligent enough to balance stuff out or the grid has to be built out with massive reserves, and that's everything but not cheap.

Today's electric cars already have the ability to delay the start of charging. It doesn't need that smart of a grid, just a smartphone app.

How does this work in practice, with a high number of cars with such apps? Say everyone arrives home from commute at roughly the same time, how is the delay factor decided/distributed?

In the most interesting fantasy, the car has an idea of how much it actually needs to charge -- Bolts and Teslas often don't use but a small fraction of their battery on a typical day -- and the grid has an idea of how much wind energy it's going to get in the middle of the night. And you'd combine the two to make sure all of the wind energy gets put into a car battery, or car batteries don't end up full because the owners don't need it tomorrow and the wind is expected to be strong tomorrow evening.

You can do the same solar energy and cars plugged in during the daytime, soaking up excess solar electric generation.

This isn't a fantasy, there's several products that do this and they've been on the market for a few years.

Microsoft provide the api for knowing in advance when energy will be the greenest: https://api.watttime.org/faq/

Some utilities will give you rebates if you use a smart charger to charge at these times (since green energy is both cheaper and time sensitive).

https://emotorwerks.com/products/juicenet-software/juicenet-...