I was a little suprised to see this article The Carrington Event: not something to worry about (2020) show up on Hacker News, a news/discussion forum with an educated userbase, but far from electrical engineering/power systems. It’s unfortunate that most articles about the Carrington Event or geomagnetic disturbances tend to be alarmist in nature, though human intuition is famously poor when it comes to the physics of electricity and it’s unreasonable to expect a layperson to electric power to be able to apply common sense to what order of magnitude is reasonable to expect when it comes to induced currents.

Hackaday had an article on the Carrington Event a while back, and while the article wasn’t particularly alarmist some of the comments definitely were. I see projects like Collapse OS, a Forth-based operating system designed to run on Z80 CPUs (which are implicitly presumed to be more resilient to GMD/EMP than modern designs). On account of Poe’s Law it’s unclear to what extent this is a fun thought experiment and to what extent this reflects a sincere belief in the sudden collapse of industrial civilization.

Here’s a helpful list to clarify some misconceptions about GMD:

  1. GMD is low-frequency and only couples into long conducting structures (on the order of 100’s of km) such as high-voltage transmission lines or pipelines. It won’t couple into your Ipad, Google Home, off-brand IoT devices or Smartphone - even if you are sticking it high in the air with a selfie stick to take a TikTok video.
  2. The low-frequency currents caused by GMD mainly affect transformers by causing half-cycle saturation, increasing reactive power draw and injecting harmonics into the system. This can cause problems by directly overheating transformer windings or it can cause dynamic instability of the grid. The effect of coupled currents in transmission lines on transformers tends to be highly local. Additionaly, only a small number of transformers will see significant half-cycle saturation. This is driven by the geospatial layout of the network, and those trasformers tend to be those towards the edge of the network. Last, the degree of transformer heating is strongly affected by the transformer core construction, with a particular design being most vulnerable. Your mileage may vary depending on the particular network you are studying!

If you are a citizen scientist or just curious, there are software tools that will let you make a judgement for yourself. These include

  1. MatGMD: Very simple to use if you have a Matlab license. Only supports analysis of quasi-dc currents without an additional analysis in MatPower.
  2. PowerModelsGMD.jl: Tricker to set up, but doesn’t rely on Matlab. Includes support for analyzing the impact of GMD on the flow of ac currents within a power system. I’m currently working on this package.
  3. PowerWorld Simulator Brain-dead simple to use, though it costs money. If you are a student there is a good chance your university has a license, though.