The Grid by Gretchen Bakke is a great deep dive into a system we think very little about. I came away from this book with both a renewed appreciation for our certainty in turning on light switches, while simultaneously being surprised at the hodgepodge of infrastructure that powers our daily lives.
Too much power
Before reading this book, I thought our main problems with electricity and the grid were not having enough power for all of our activities. The Grid paints a different story.
Electricity is different than other materials, in that it must be consumed the instant it is produced. We don't have great methods of storing it indefinitely (more on that below).
What that means is that when there are surges in power, parts of the grid will actually turn off in order to avoid damaging components.
In both Hawaii and Vermont, there are restrictions actually forbidding new residentaial solar installations from sending power back into the grid. In the pacific northwest, wind turbine operators are paid to disable their turbines.
A big problem with too much power is "variable generation". Certain power sources (wind, solar) generate power only at particular times. Nuclear, and to a lesser extent hydroelectric power, don't have this problem.
A big problem with variable generation is matching generation with peak demand. In California, the peak generation happens from 12pm-4pm, when most of the solar power is generating the maximum output. But peak demand typically happens around dusk, when offices still need to be powered at the same time that residents are turning their lights on.
While we often think about these sources being insufficient to supply enough power, the bigger problem can be when they suddenly turn on and cause a power surge.
Sometimes, when variable generation doesn't generate enough power, utilities will turn on giant coal-powered generators to supplant the generation they need.
AC vs DC
I hadn't really understood the differences between AC and DC in a deep way before this book.
Direct-current power was originally favorted by Edison. DC power pushes electrons, well, directly from source to sink. The electrons are constantly moving in the same direction. DC power can only work over a short distance (<10 miles), so it can typically only be used for last-mile power generation. Your laptop, for instance, uses DC power that comes off an AC adapter.
Alternating current works differently, by pushing some electrons forward, and then reversing the flow of electricity at a regular interval. AC current can travel for many hundreds of miles, but requires transformers to scale the voltage and current up and down.
One other learning here: long-distance power lines are dramatically cheaper than burying cable, primarily due to heat regulation. Burying cables underground requires a large amount of insulation, and it's pretty difficult to fully dissipate all the heat.
Centralized vs Microgrids
The U.S. operates on an extremely centralized grid... but that wasn't always the case (nor does it have to be)! Apparently early grids ran in an extremely de-centralized manner, where each plant might burn its own coal, or run its own wind turbines for power directly on the premises.
The decision to use big global grids in the U.S. was relatively arbitrary. It was mostly driven by Samuel Insull, a businessman who was originally an apprentice to Thomas Edison.
Insull laid the framework for consolidation, as he realized that he could supply the best possible rates when users were using electricity at all times of day. He invented the peak/off-peak rates, and discounts for commercial usage that still exist today.
That said, there's no reason that we couldn't exist with a more distributed, local-first grid. Since that point in time, the government has continued to push for consolidation thanks to entrenched legislation and price-fixing.
I hadn't realized this, but the introduction of "smart meters" on the grid are a very big deal! They allow power companies to selectively re-route power at times of peak demand, and to selectively turn off power in areas where it's less essential.
There's been a lot of public backlash related to these (the book shares a story of a woman in texas who pulls a shotgun on the worker who comes to upgrade her meter), mostly because the public utilites have such low NPS.
That said, I think there's an interesting opening here for some sort of startup to build great software and hardware, that consumers use and is sold to the utilities. It helps make the utilities more money, and could provide consumers with a higher level of control around their energy choices.
Grid-scale energy storage
This was also discussed in How to Avoid a Climate Disaster, but one key point called out in the book is that there is no grid-scale energy storage.
What that means is that we have to use electricity pretty much the instant it is produced. In many locations, the grid will automatically shut itself off in response to too much additional power.
The book talks about a few interesting alternatives here. In Alaska, apparently there's a single megawatt battery that can power the city for 30 minutes (roughly the time for the diesel generators to engage and begin power generation). In a handful of mountain cities, there are artificial lakes which pump water into a reservoir at times of peak generation.
One other idea I hadn't considered... electric vehicles present a solution for grid-scale energy storage. You could imagine a world where electric vehicles smartly charge and discharge at different times of day, depending on their average length of commute and the energy needs of the grid.