US risks national blackout from small-scale attack | Hacker News Reader

US risks national blackout from small-scale attack

online.wsj.com

23 pointsanigbrowl23 comments

Blackouts cause people a lot of trouble. But building physical protection against attacks like the Metcalf transformer shooting would be expensive, and people already complain their electricity bills are too high. Given the choice between a 10% higher bill and a 1% greater annual chance of a blackout, which would most people choose?

The article focuses on solutions involving physical security. Perhaps it would be more cost-effective to bolster deterrents? For example, life in prison for anyone convicted of sabotaging a major facility of any kind (in the Metcalf case, this could include both the shooting and the cutting of telecom fiber beforehand).

There also seems to be less focus on improving the resilience of the system as a whole--given that the threats are being modeled to determine minimum amounts of sabotage required, why not emphasize increasing that minimum, rather than reinforcing the few critical points of failure? Computing has moved in the direction of resilient systems made of less reliable components, and so could the grid.

As an aside, a blackout in summertime can be pretty interesting. The one in 2003 brought people together in new ways--I experienced a sense of community which was quite unusual, "old fashioned" you might say.

With regards to your first paragraph, we have plenty of money to spend on defense. The USA is expected to spend around 1.2 trillion dollars on defense in 2014 [0]. The problem, is that most of that spending is overseas "defense", and very little of it is used to secure critical domestic infrastructure. Like our energy grids, our roads and highways, etc.

---

[0]: https://en.wikipedia.org/wiki/Military_budget_of_the_United_...

In the US, isn't electricity supplied and managed by corporations rather than the government? This complicates things somewhat, as it exposes the system to market forces ("I want a choice of providers and I will choose the cheapest") and may reduce the government's ability to "intrude" by adding safety measures.

Perhaps more critically, since the US grid seems to be operated by a number of competing companies, it may be less likely that the system itself will be made resilient, because there is limited cooperation.

If the people of the US decided to spend money on hardening power delivery, would the money come from tax dollars, or electricity bills?

It is supplied by corporations, but it is a HIGHLY regulated industry, and almost all corporations have a government granted local monopoly. Very little choice in where you get your electricity from.

This is a really great point. While I don't mind increasing security at the major failure points in the short term, it makes a lot more sense to make the system as a whole more resilient and get rid of the small failure points.

But building physical protection against attacks like the Metcalf transformer shooting would be expensive

Walls are not expensive. I don't think that anyone engaged in a systematic attack on the distribution network is going to be particularly deterred by the threat of prison, given the potential strategic payoff.

A large-scale blackout plus an attack on the only factory that makes those large transformers has the potential to cripple Us industrial capacity for weeks, if not months.

Your analysis is overly scenario dependent. If someone is determined to take out a transformer they'd breach the wall or go over it. Or perhaps, staying outside the wall, cover it in a cloud of carbon fibers, we have warheads that release those instead of using a pure explosive charge to take out electrical installations like these.

Certainly, but the object is not to make them absolutely attack proof but to make attacks more difficult and expensive, and thus easier to deter, detect and interrupt.

About the limited number of manufacturers of transformers: Would it be possible and inexpensive to eliminate this problem by keeping a government stockpile of spares somewhere protected with limited access? Do we have planes big enough that in a crisis we could deploy them in hours, or would it involve trucking them to their destination over the course of days? In the latter case, surely the military has secure facilities well distributed over the country and it should be possible to pre-position them.

Or are these transformers so large that they basically have to be built on-site? Or so expensive that keeping ~10 spares would cost too much (say ~$200 million or more)?

Any electricity experts want to chime in?

They're expensive, and I gather perhaps customized, e.g. at least sized to demand, even if upstream voltages might be somewhat standardized.

We in Joplin, MO lost a substation of perhaps this class in the 2011 EF5 tornado, and my father saw them delivering the replacement not too long ago, it was big. (Note the delay was due to the fact that there was no need to supply lots of power to an area that was wiped off the face of the earth prior to enough rebuilding.)

http://www.tjpottertrucking.com/trucking-oversized-loads/hea... ... if a plane can carry that, then transport by air is possible. The Antonov An-225 Mriya could probably carry a large transformer.

IEEE (or the manufacturers?) have commissioning procedures. Typically the larger transformers are manufactured elsewhere, then reassembled on-site and commissioned (figure a week to a month).

Your ballpark estimate on the cost is close enough.

One point to remember is that 'if it bleeds it leads', news stories are often sensationalized to achieve more viewers/readers. Typically for power applications, the sizing of equipment is based on peak demand such that the equipment is only fully utilized in a short time period (people get upset when power isn't available or a fire is started). While this report is probably correct in essence, without knowing the exact scenario that was modeled it is difficult to determine a solution.

BTW if you're not a subscriber, then you can circumvent the paywall by searching for the article title on Google and visiting it from there.

You should probably resubmit this without the paywall. It is frontpage worthy, but if I've learned anything, HN does not like paywalls at all. :P

This link works http://online.wsj.com/news/articles/SB1000142405270230402010...

It doesn't.

http://bit.ly/1obz7o0

http://lmgtfy.com/?q=U.S.+Risks+National+Blackout+From+Small...

should work though :D

Your link does not work for me. What works is what your parent suggested: searching for the title and clicking on the result.

Ahaha, oops.

The link is the same as the one Google takes me to, but apparently they check the referrer. If you come from Google: no paywall.

I always just open WSJ and NYT articles in a new incognito or private browsing window.

That, or slap the affiliate ID for Twitter on the URL.

While I can't read the article (paywall, and I'm not going to break the law by circumventing it), it would take many small scale attacks to cause a "national blackout". We have 3 grids, East, West and most of Texas (which is big enough to have a stable grid and avoids a lot of hassle by not crossing state lines).

And these grids aren't like you probably envision, a good description I came across is that they're a number of ponds with small streams between them. So it would take some cleverness to affect more than one "pond", it all depends on how the grid, or a subset of it is set up. E.g. taking out a lot of California, since at least as I gather it imports so much power, would be easier than many other areas.

Or take San Francisco (please! :-). I remember after one nasty blackout then Mayor Willie Brown, who's a bit more grounded, you might say ^_^, than many area politicians, saying, well, maybe we shouldn't shut down the last power plant in the city....

The disturbing takeaway from the article is that it would in fact only take 9 attacks at strategic points to bring down the entire grid, per the FERC report - hence the relative urgency on display in asking utilities to submit plans for improvements by June, and the unavailability of the report to the public (as far as I am aware, after spending a few minutes searching for it).

BTW, "American Blackout" was a recent National Geographic program about a hypothetical nationwide blackout http://channel.nationalgeographic.com/channel/american-black...

There's a British version too. Probably worth watching one or the other, if only to prompt you to make some basic disaster preparations.

Reply on news.ycombinator.com