Hacker News Reader

Comment by londons_explore | Hacker News Reader

I disagree.

A large solar storm is unlikely to damage many transformers. Transformers have low impedance, yet solar storms have the biggest impact on high impedance systems (like copper telephone cables, especially when 'on hook').

The only time a power network is high impedance is when it isn't connected (ie. when part of the power network is switched off). Trivial operational changes could be made to ensure that offline parts of a power network are grounded during a solar storm.

Reserve transformers might be a good plan for other reasons - eg. fast recovery after a war on US soil - transformers are usually early targets in war, and hard to defend.

View on HN

Even a Carrington-level event, which refers to a solar coronal mass ejection (CME) observed in 1859, is indeed a concern for modern electrical infrastructure. During such an event, geomagnetically induced currents (GICs) could flow through electrical grids, potentially overheating transformers and causing them to fail. These GICs are particularly harmful to transformers because they can lead to half-cycle saturation, increasing the reactive power demand and causing internal hotspots.

While transformers themselves may have low impedance, it’s the connectivity and expanse of the grid system that makes them vulnerable during geomagnetic disturbances. The longer the conductive path, the more susceptible the system is to GICs, regardless of the impedance characteristics of individual components like transformers.

An X500-level solar flare would be many times more powerful than the Carrington event and could wreak havoc on our electrical infrastructure. Given our increasing reliance on electricity for everything from communications to transportation, the societal impact would be significant.

Hence, the idea of a Strategic Transformer Reserve isn’t just about preparing for solar storms; it’s a multi-hazard approach that also considers other threats like cyber-attacks, terrorism, or even conventional warfare.

The key question remains: do the potential catastrophic consequences justify the financial and logistical costs of maintaining a transformer reserve? From a risk-assessment standpoint, considering the dire impacts of a severe solar event or other threats, investment in a transformer reserve could be seen as a rational and scientifically justified precaution.

Particularly when the lead time to manufacture such equipment is in the order of years. A grid outage of that duration would be biblically catastrophic: without electricity, we have no water, no gasoline, no communications. We’d survive, but the suffering is not something it seems wise to consider an acceptable cost. Especially at a time of enhanced geopolitical competition, the effect of a storm whose severest effects would likely be localized to the sun-facing hemisphere, seems important not to underestimate.

I think it would be great if, instead of countries merely having arms races, they had preparedness races. I guess for us ape-brained humans, tribalism feels more compelling than a threat you can’t bite the face off.

> The key question remains: do the potential catastrophic consequences justify the financial and logistical costs of maintaining a transformer reserve?

isn't this something that can be trivially answered with game theory? you just need the cost and the probability.

> geomagnetically induced currents (GICs) could flow through electrical grids, potentially overheating transformers and causing them to fail.

yes, but such currents are typically measured in milliamps or, during the biggest storms, amps. Most utility scale transformers would be needing DC currents of the order of thousands or tens of thousands of amperes before failure.

Even then, overheat sensors would detect such a condition, and operationally the circuit could be de-energised and connected to earth to protect it.

Also, where there are perhaps thousands of transformers connected to a particular distribution grid, the load from the GIC's is shared amongst them, further reducing impact.

I dunno. The American Geophysical Union seemed to think it was a BFD in one of their conferences back in the 2000s. Had a whole presentation about how certain parts of the US grid would lose large transformers that would likely take months to years to replace as there are few spares and they have to be wound with miles and miles of copper by hand. Said that large swaths of the US would be without grid power for very long periods.

Said that third world countries would fare pretty well though.

It sounds like they need some type of automatic or robotic winding system... Even without some catastrophe, they surely have to make new transformers from time to time, and winding miles of copper by hand sounds ridiculously inefficient for a manufacturing process.

Hand winding allows for a high degree of precision and quality control that may be difficult to achieve with automated machinery, particularly for specialized or custom designs. Skilled technicians can closely monitor the process and make adjustments on the fly to ensure that the winding is as precise as possible. This is crucial for large transformers, where even minor imperfections can lead to inefficiencies, increased heat, or failure.

Transformers for electrical grids can vary greatly in their specifications depending on a variety of factors, such as location, usage, and existing grid architecture. Customization is often necessary, and hand winding allows for this level of customization to meet specific criteria, including the number of windings, the type of core used, and other design elements.

Copper wire is both flexible and delicate. It needs to be handled carefully to avoid nicks, kinks, or other imperfections that can compromise the transformer’s performance. Human technicians can adapt to the nuances of the material more effectively than machinery in some cases, ensuring that the wire is handled with care throughout the winding process.

Geomagnetically Induced Currents (GICs) are often measured in amps rather than milliamps, and their values can vary significantly depending on the specifics of the geomagnetic storm and the electrical grid. During the 1989 geomagnetic storm that led to the collapse of the Hydro-Québec power system, GICs of hundreds of amperes were measured. While transformers are designed to handle thousands of amperes of alternating current, they are generally not designed to handle direct current like GICs, which can lead to transformer saturation and increased reactive power demand.

Now, let’s talk about resonances. The electric grid can have resonant frequencies due to the combination of inductive, capacitive, and resistive elements. Resonances can cause the system to magnify the effects of incoming disturbances, much like how a tuning fork resonates at its natural frequency. When a geomagnetic storm induces currents in the Earth, those currents generate a magnetic field that interacts with the magnetic fields of transformers and transmission lines. If the frequency of these geomagnetic disturbances happens to match or come close to a resonant frequency of the electrical grid, the amplitude of the induced currents can be significantly magnified.

To be specific, resonances can occur in multiple parts of the system:

1. Transformer windings have their own resonant frequencies, at which the impedance becomes high, leading to larger voltage across the windings for the same amount of GICs.

2. Long transmission lines can have characteristic impedances that interact with the impedances of transformers and other elements to create resonant circuits, thus amplifying the GICs in localized regions.

3. Harmonic resonances can occur when nonlinear elements like transformers generate harmonics that coincide with resonant frequencies in the grid, thereby magnifying the effective GICs.

So, even though individual transformers may be able to handle small DC currents without immediate failure, the presence of resonances and the cumulative effects across a large, interconnected grid make GICs a non-trivial concern.

IIRC, the main argument for spare transformers is that the lead time on them back in 2017 was 2+ years, and we have no domestic sources. I can't imagine the covid era supply chain issues have improved things. If North America is facing the sun during a major event, many countries would be fighting for replacements, and it could easily take a decade to get everything back online.

>we have no domestic sources.

Where are they made? This sounds a bit like a national security failure. A country as large and world-dominating as the US should be able to manufacture its own electrical distribution infrastructure.

The 1989 solar storm did cause damage to some Quebec transformers:

https://hackaday.com/2017/04/10/lights-out-in-quebec-the-198...

> The ground connections of Hydro-Québec’s transmission system of high-tension lines and transformers eventually started conducting some of these earth currents, and at 2:43 AM, protective circuit breakers tripped at the Chibougamau substation in central Québec. This caused an imbalance on a 750 kV transmission line, which tripped breakers 150 km away.

> Within one minute, the cascading failures ...

So the actual problem was that all their power grid was turned off... ie. disconnected... ie. high impedance... ie. the most vulnerable state.

Operational changes could be made so that all 'tripped breakers' lead to at least one or other side of the breaker being grounded rather than left open circuit, which would solve the problem. That only needs to be the case during a solar storm, and electrical power lines are frequently grounded during maintenance for safety anyway.

transformers are usually early targets in war, and hard to defend early targets in war, and hard to defend.

Not if they're more than meets the eye, like, say, robots in disguise.

Reply on news.ycombinator.com