How Would a Nuclear EMP Affect the Power Grid?

📺Nebula comes free with a year of Curiosity Stream (for only $15/year!): curiositystream.com/practicalengineering 📚Looking for a great holiday gift idea? My new book is now available everywhere: practical.engineering/book

Grady, AM radio antennas are usually located nearer to the ground than FM antennas. The broadcast signal from an AM antenna can result in structures such as cranes becoming charged, resulting in high voltage, low current arcing. One of the oil tankers in our fleet reported that at a certain dock they experienced getting shocks when putting slings on the crane hook. The crew discovered if you held a small fluorescent bulb near the crane hook you could get it to light, Uncle Fester style. We found that there was an AM radio antenna nearby that was used occasionally by a radio station. When in use at the same time the ship's crane was being used the crane would become charged. In checking with commercial crane operators, this was a known phenomenon. One operator told us that they had issues when working on San Francisco Bay. It took some time, and a solution was worked out with the refinery, radio station and new safety practices for crane use on our ships. Oddly enough I probably was 30 years into my career before this problem was ever encountered. Bob

“A nuclear detonation is unwelcome in nearly every circumstance” is the understatement of a lifetime 😂 Great video Grady, really enjoyed learning about this!

I'm a retired electron pusher. I started as a microwave radio tech in the USAF (honor graduate Keesler AFB 04/1978). Over a 40+ year career, I moved from RF Tech to Engineering Tech to Design Engineer to PCB Layout Engineer to Senior EE with a major defense contractor. In fact I worked as a contractor since 1996, wearing many hats and mostly in aerospace and military design. Consequently, I have worked at times intimately with EMP and as SI (Signal integrity) oversight for an engineering department developing large integrated circuit test systems. My favorites were spacecraft and satellite design. The space environment is pretty unforgiving, and those devices typically cannot be brought back for repair. They have to function in the harsh environment of space. Your video was an excellent overview. My only comment is that the study you referenced specified a 1 megaton HEMP (high altitude EMP) detonation. That sounds a bit like hunting moose with a .22 rifle. In a tactical situation intended to disrupt, say, the United States, I would expect at least a 10 MT if not 20 MT device, and possibly more than one, e.g. one over the Appalachians at a latitude between ew York and Washington; oneover central Texas, and one over the Sierras on a latitude between San Francisco and LA. This would hit major population and technology centers while taking down all three major grids: NS along the East Coast, EW across Texas and the south-eastern US, and NS along the West Coast. As for the solar induced equivalent, we are actually overdue, statistically speaking, for another 1859 Carrington Event level solar CME (coronal mass ejection) impact. That one burned telegraph wires in two and set operators' equipment on fire, which was the most advanced electrical technology at the time. We just missed an even larger CME by less than 2 weeks in our orbit in 2012. It would have been a technology killshot. As you stated, the fast E1 pulse would couple into virtually all conductors, even the smallest, inducing spikes in the traces on integrated circuit (IC) silicon dies. We briefly discussed this in an IC Fabrication course I took around 2002. The copper traces on a printed circuit board (PCB), where components are typically mounted and interconnected, are also ready made antennas for the E1 pulse. I focused on PCB layout and signal issues for about 18 years of my career. The E2 period does indeed have many characteristics like lighting, and many lightening arrestors should handle it. The E3 pulse is another matter. In EMP and HEMP situations it is able to induce ground currents. In a solar CME induced planetary scale EMP, these could be significant, dwarfing the regular man-made HEMP. These are the currents most likely to destroy large grid regulating transformers and even generators. They can travel into electrical loops by a "sneak path", using the ground connection to enter electronics not designed with a robust power connection, and traveling "up" to the power source (normally the input power). Needless to say, a reverse current of high magnitude would be devastating to most equipment. I DO NOT have experience in large scale power distribution, "The Grid", but warnings I have read indicate it, too, could be susceptible to such high reverse currents. Add this to the directly coupling E1 pulse, and there could be significant current and high voltage spikes tnduced into the windings in both transformers and generators. These transformers and generators are not off the shelf components, as the video stated. These are custom built OVERSEAS. In a worldwide event like the solar-induced EMP, it is very unlikely these will be replaced soon... if ever. Consider the events in New Orleans in the two (2) weeks following Hurricane Katrina. Once it became apparent that services like electricity, phones, and water were not going to be immediately restored, the social structure began to crumble. When it became known that supply trucks were not going to be able to resupply the supermarkets, the normal "three days of food on the shelves without resupply" was stripped in TWO HOURS. I also have a degree in psychology, and human dynamics in events like this are one of my areas of interest. One of my best friends had family in the area in Baton Rouge, and they experienced predatory behavior escalate as well. Given a probable loss of the power system due to Grid failure, collapse of the financial system due to loss of the communication system - no internet means no banking system, loss of transportation for food resupply, and loss of refrigeration for existing food storage AND medicines, and the situation in many urban areas will probably be much worse than the few weeks following Hurricane Katrina. Don't forget that no communication, no transportation (gas pumps run on electricity even if the vehicles' computers were not fried), and totally dark cities at night means no police or other order enforcing bodies. It could become very dire very quickly. Read One Second After for a fictional depiction of a small town following an HEMP event written by an expert in EMP. This book was given to congressional members when a bill to harden the electrical grid was under consideration. My thoughts about that book were expressed exactly by the Naval Captain that wrote the epolog: I would love to have been able to read this as a work of science fiction, but I knew all too well that everything portrayed was entirely possible.

I like how calmly this guy talks about the end of civilization

As an electrician, I have been involved in a lot of interesting construction projects. My favorite and most memorable one was an entire building that was built with EMP shielding. Just like an MRI room. It was a building to house the control center of a major city’s grid control. It looked very much like your footage of the grid control center in this video. I think you could do an entire long-form video about the engineering that goes into MRI rooms. They have copper walls and EM filters and wave guides. They can be huge or fairly small. The world-renowned company ETS-Lindgren is one that has done all the shielded rooms and buildings I’ve ever seen. They’ve done secret projects but also lots of health-care projects.

VERY surprised you didn't mention 1962's Starfish Prime (1.5MT) which was over Johnston Atoll and it's subsequent EMP knocked out power and phones here in Hawaii for at least a few hours. Hawaii is nearly 900 miles away. Some LEO satellites also got knocked out because of it. It's a perfect answer to the videos question and a literal what would happen in the real world.

Additional notes about transformers: Even small substation transformers cost $MM's. Additionally, they're often custom-sized specific to the substation they are installed at and the lines they are tied to. Manufacturers make them to order and they often have multi-year order-to-deliver dates. Even repairing a transformer can take over a year.

Great Videos Thanks! As a Power Supply Designer for many years, I remember an experienced senior designer when I was fresh out of school told me: "When a transformer core saturates, it's just a piece of wood!"

Man, these Election Day ads are getting out of hand. Now I don't know which EMP plan to vote for! 🤣

You might find this interesting regarding an AM tower. Back in the early 1980s I worked at a radio station and for the AM side of the station, the tower was out behind the building. During storm there would be these pulses that would run through the system such that they would spike the needles on the monitoring needle gauges (old school). I would go outside to look at the tower and you could see and hear electrical pulses running down the guy-wires. It was amazing and scary at the same time. I never knew why it would do that (I'm not an electrical engineer or anything) but I assumed that the air was, like, "charged" from the surrounding storms and that was interacting with the "AM waves" and generating electrical current that was grounding down the guy-wires. Do you think AM towers still do that or do they ground them differently now?

A balloon, you say?

You missed something. A high voltage relay has a mechanical backup just in case the power is lost to the GE 745's (or whatever they are using). So even if a EMP hit the grid, it's still going to shutdown to protect the grid. A restart can take place with operators controlling the system. And yes, the grid does have lightening protection which reaches 1 gigawatt worth of power. Once the EMP has passed, the grid will reset and they can bring everything back up. There are many lightening protectors in the grid so you might lose some of the relays but not all of them and threw the SCADA control system will act as a controller to any part that might still be down. In short, an EMP will take the grid offline, however it will come back up within minutes of the event.

Protection and control engineer here - I design the control systems that trip the breakers using the digital relays mentioned in the video. Very informative way to explain this paper to the public, in a much more interesting manner! There are many utilities that are currently buying concrete control houses or having their control houses shielded with elaborate copper shielding systems on critical bulk electric system sites. These approaches (largely) mitigate the issue of E1 pulses affecting the control relays. It's proven much more difficult to prevent damage from the E3 events, as the protective relays we use now won't necessarily detect those conditions and trip everything offline. Furthermore, while your control relays may be shielded from most of the issues, your voltage transformers and current transformers out in the yard will not be as protected. If you can't rely on the data coming into your relays, you can't effectively trip and protect your equipment.

One critique from a EE. In your Marx Generator experiment, you can see the spark jump to the picture frame. So, the resulting damage is more likely from the high voltage entering the case and not the EMP pulse triggered by the spark.

I once used a vandegraf generator to build up a surface charge across a bunch of my students holding hands and standing on desks to insulate them, when we discharged the person-circuit, it caused the television in the room to turn off and was not able to turn back on immediately. Wasn't able to reproduce that day but we had a fun few days of discussion afterwards

I used to work for a defence contractor producing military avionics [military radios for military aircraft], and we used to test for EMP in our military radios... and I remember the test set was able to generate a pulse with a 1KV peak in a 1ns rise time. This is a lot faster than lightning, which is typically 1KV peak, with a 1 microsecond. This is the EMP spec that we used for testing our avionics systems, as we had special test rigs for testing the effects of EMP. The power supply rails, such as the 5 Volt rail, had special zener diodes that would short out the power supply rails, and was designed to detect the gamma ray flux. It would create a short glitch for the radio, but protect against surges that might occur. In fact we tested a muzzle velocity radar unit, by putting it into a reactor, and the protection circuits worked. In addition that filter connector that connect to the radio system also had special EMP filters to protect the avionics.

I actually had a VERY nearby lighting strike that fried my home router, and my brother's next-door. We had a Cat5 cable connecting our homes, running down the wall and underground. This seems to have acted like an antennae and allowed the EMP generated by the lightning strike to fry both of our home routers connected via that cable!

The EPRI study is a source I've been using for years on educating other amateur radio operators regarding hardening electronics against EMPs. However an EMP is not my fear, a Carrington Event level CME is and is much more likely.

You make some incredibly thorough, highly educational videos. I just discovered this channel and I must say, I love it!