US unprepared for an electrical grid collapse, but it costs only $300M • Hacker News Board

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#21

Post by asperous » Sun, May 03, 2020, 10:18 PM UTC

makomk wrote: post #20

btilly wrote: post #16

Earlier quoted context omitted.

It is true that the solar flare will induce current in any exposed wire. However the following factors increase the current. The length of the wire, the thicker the wire, how poorly insulated it is, and the stronger the magnetic surges caused by the flare. Long distance transmission lines are very long, fairly thick, and very exposed. The result is that the surges that they experience are several orders of magnitude…

I can see a couple of rather obvious problem with this. The electrical grid is, well, connected - the long, high-voltage sections are connected to the lower voltage sections which are connected to the even lower voltage sections. Those connections are through transformers, but if the surge is big enough to destroy even the largest transformers what's to stop it from blowing through all of the transformers and frying…

Surge protectors

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#22

Post by btilly » Sun, May 03, 2020, 10:19 PM UTC

How much should we worry about this?

http://www.lloyds.com/~/media/lloyds/reports/emerging%20risk... was prepared by Lloyds of London after the 2012 detection of a large solar flare by a satellite in interplanetary space. (That one missed the Earth.)

Their estimate was $600 billion - $2 trillion dollars of damage from an event that happens roughly once ever 150 years on average. Therefore the amortized cost of this risk per year is $4 - $13 billion.

If a substantial fraction of the risk can be mitigated over the next decade for $300 million, it would be cheap at 10x the price. In fact Berkshire Hathaway probably has enough exposure to this risk that it makes financial sense for them to not debate over who pays and to just create the stockpile to reduce potential future insurance claims.

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#23

Post by petrocrat » Sun, May 03, 2020, 10:19 PM UTC

abeppu wrote: post #10

Maybe someone who knows a bunch about this stuff can enlighten me. I was kind of under the impression that a large solar flare would damage _all sorts_ of electronic equipment, by inducing currents wires all over the place. - If we had extra transformers on hand, how would you protect them from also being damaged? - If a solar flare is powerful enough to damage these transformers, would it also have damaged a large p…

Transformers are usually housed in metal boxes. Any wires inside metal sheeting of whatever geometry would be protected because any current induced would be induced in the metal container which dissipates the power of the pulse. To learn more read about Faraday cages.

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#24

Post by Noumenon72 » Sun, May 03, 2020, 10:20 PM UTC

intopieces wrote: post #9

xkapastel wrote: post #3

> In testimony before a Congressional Committee, it has been asserted that a prolonged collapse of this nation’s electrical grid—through starvation, disease, and societal collapse—could result in the death of up to 90% of the American population. This could happen at any time due to a solar flare. The coronavirus has made it pretty clear that America is completely unprepared for any sort of major deviation from "norm…

> The coronavirus has made it pretty clear that America is completely unprepared for any sort of major deviation from "normal". This is the nature of the US system of competitive federalism. States only coordinate with the federal government when either (a) it's politically expedient, because the ruling party in the same for both at that time, or (b) the federal government forces this issue, at which time lawsuits en…

I don't understand how you have identified this as the problem. Sure, states haven't coordinated on lockdowns, testing, or mask purchasing. But it's not because they're struggling against the federal government, it's because there's no official policy to coordinate with.

Seems to me like our main difficulty deviating from normal is the fact that a 20% decrease in demand kills off 10% of businesses within weeks and causes 20% of people to instantly get behind on their rent. There's no slack in the system.

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#25

Post by dvt » Sun, May 03, 2020, 10:20 PM UTC

Being "prepared for black swans" is a bad idea. Of course, you want your system as a whole to be as antifragile as possible, but one should never prepare for black swans, but rather buy insurance (so to speak): hedge, diversify, minimize impact. For example, we can imagine a world where some other pandemic is being spread. One that, instead of attacking the lungs, attacks the kidneys. All of a sudden we'd need dialysis machines -- not ventilators. See how preparing for some specific black swan is misguided (as we can be hit by any other, equally unlikely, black swan event)?

> In testimony before a Congressional Committee, it has been asserted that a prolonged collapse of this nation’s electrical grid—through starvation, disease, and societal collapse—could result in the death of up to 90% of the American population.

Okay, this sounds like an interesting assertion. Can we back this up? Is this just alarmism? Where's the data?

> Yes, $300 million dollars for a stockpile of 30 HV transformers is far too expensive to prevent 90% of the public dying and the rest reduced to cannibalism.

I see. So now we're assuming some off-the-charts assertion, which is dubious at best[1], is absolutely true. The cannibalism is thrown in for extra flair. Gotta' get them clicks somehow.

> Update: I forget to mention that I’m far more worried about accidental nuclear war, bioterrorism and AI run amok than I am about solar flares.

Anyone that's seriously worried about "AI run amok" (whatever that means) doesn't understand the first thing about AI.

[1] https://www.popularmechanics.com/military/weapons/a25883/nor...

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#26

Post by realtalk_sp » Sun, May 03, 2020, 10:23 PM UTC

btilly wrote: post #22

How much should we worry about this? http://www.lloyds.com/~/media/lloyds/reports/emerging%20risk... was prepared by Lloyds of London after the 2012 detection of a large solar flare by a satellite in interplanetary space. (That one missed the Earth.) Their estimate was $600 billion - $2 trillion dollars of damage from an event that happens roughly once ever 150 years on average. Therefore the amortized cost of this r…

Sadly it might cost Berkshire even less to just reinsure against the risk.

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#27

Post by elliekelly » Sun, May 03, 2020, 10:24 PM UTC

dvt wrote: post #25

Being "prepared for black swans" is a bad idea. Of course, you want your system as a whole to be as antifragile as possible, but one should never prepare for black swans, but rather buy insurance (so to speak): hedge, diversify, minimize impact. For example, we can imagine a world where some other pandemic is being spread. One that, instead of attacking the lungs, attacks the kidneys. All of a sudden we'd need dialys…

Investing a nominal amount in strategic backup equipment that you might never need is hedging.

An event that has a low probability of occurring but that (if it occurs) will be highly catastrophic is exactly the scenario that calls for hedging. Most critical systems have already diversified by building in redundancies (generators) and an insurance check wouldn’t be very useful for keeping the rest of us alive.

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#28

Post by FpUser » Sun, May 03, 2020, 10:29 PM UTC

csecdaemon wrote: post #2

While I don't disagree that transformers... let alone key transformer.. are critical. By disabling 9 transformers you cannot take down “THE GRID”. You could certainly cause havoc and perhaps take down sections of a grid, perhaps a large one such as new york city. But these would be localized. While ensuring that critical components are local and available is important much of this is being done in the energy sector.…

There is no singular "grid". It just does not work that way

This did not prevent "Northeast blackout of 2003" when bunch of states in the US and province of Ontario in Canada went without power for 3 days. I remember it quite clearly. Was not big fun. Luckily my friends and I with families went camping on the lake so did not really suffer other than some little disaster in a fridge ;)

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#29

Post by phonon » Sun, May 03, 2020, 10:35 PM UTC

realtalk_sp wrote: post #26

btilly wrote: post #22

How much should we worry about this? http://www.lloyds.com/~/media/lloyds/reports/emerging%20risk... was prepared by Lloyds of London after the 2012 detection of a large solar flare by a satellite in interplanetary space. (That one missed the Earth.) Their estimate was $600 billion - $2 trillion dollars of damage from an event that happens roughly once ever 150 years on average. Therefore the amortized cost of this r…

Sadly it might cost Berkshire even less to just reinsure against the risk.

Berkshire is the 5th largest reinsurer in the world[1], so not likely to happen...

[1]https://www.insurancebusinessmag.com/us/news/breaking-news/t...

Re: US unprepared for an electrical grid collapse, but it costs only $300M

#30

Post by lisper » Sun, May 03, 2020, 10:37 PM UTC

Noumenon72 wrote: post #24

intopieces wrote: post #9

Earlier quoted context omitted.

> The coronavirus has made it pretty clear that America is completely unprepared for any sort of major deviation from "normal". This is the nature of the US system of competitive federalism. States only coordinate with the federal government when either (a) it's politically expedient, because the ruling party in the same for both at that time, or (b) the federal government forces this issue, at which time lawsuits en…

I don't understand how you have identified this as the problem. Sure, states haven't coordinated on lockdowns, testing, or mask purchasing. But it's not because they're struggling against the federal government, it's because there's no official policy to coordinate with. Seems to me like our main difficulty deviating from normal is the fact that a 20% decrease in demand kills off 10% of businesses within weeks and ca…

> it's not because they're struggling against the federal government

Actually, it is:

https://thehill.com/homenews/state-watch/495519-maryland-gov...

“The National Guard and the State Police are both guarding these tests at an undisclosed location [because] the federal government seems to be interrupting supplies that are being sent elsewhere in the nation, and so I wanted to make sure that we received what we ordered,” [Illinois Gov. J.B.] Pritzker told reporters at an April 15 press conference.