/* ]]> */
Jul 132012
 

Our national grid relies on 2100 of these mammoth – and in many respects, irreplaceable – transformers.

We regularly worry in our articles about a failure of our nation’s electricity grid – the criss-crossing network of power lines that connect the various power generating facilities around the country with the various power consuming facilities – most particularly, the major switching substations that route the highest voltage connections around the country.

Think of the power grid a bit like a transportation network.  We have super-highways, regular freeways, highways, arterials, surface roads, minor roads, cul-de-sacs and so on.  For example, to drive from home to work, you first leave your driveway, maybe go down a residential street, then to a more heavily trafficked street, then to a major arterial, then onto a freeway, then through an interchange and onto another freeway, then off, via various surface streets, and ending up in the parking garage underneath your office.

It is the same thing with the movement of power across the country.  Power originates in a generating station, then travels to a switching station where it then joins a ‘super highway’; it travels across the country, and perhaps goes through some interchanges as it changes ‘freeways’, then starts to feed down through surface streets and their intersections, until ending up coming in to your own household.

The key points of vulnerability to the power network are not the thousands of miles of power line.  It is the ‘interchanges’ – the switching stations.  The power is useless and meaningless in the power lines – it only has value if it can pass through all the ‘interchanges’ and ‘intersections’ and complete its journey at your light switch and light.

Our Power Grid is a Mismatch of Incompatible Components

Unlike our national interstate system (and also unlike the internet), there aren’t a huge number of different routes power can travel to the people who need it.  And not every different path is fully compatible with every other different path.

There are 2100 major high voltage transformers (consider them as freeway interchanges) and in total, the nation’s power grid is operated not by a single authority or even by a coalition of half a dozen major players (as is the case with the internet, for example) but instead by an assortment of some 5,000 different entities, most of whom are competing with each other.

Furthermore, these 2,100 transformers aren’t all the same and interchangeable.  An industry rule of thumb says that for every 13 transformers, you’ll encounter ten different designs.

Unsurprisingly, all these different pieces fit together somewhat clumsily.  For example, this article talks some more about the vulnerability of the power grid to solar storms.

Repairing a Damaged Grid is Difficult

A retail chain, some years ago, had a famous and very successful slogan – ‘It is the putting right that counts’.  The key concern, with our power grid vulnerability, really is not so much the vulnerability itself (although that is of course a concern too) but rather ‘the putting right’ – restoring electrical service to the nation if/when it is disrupted.  If power can be restored in a matter of hours, then it is hardly life changing.  But if a grid failure could lead to many years without any power at all, then clearly it becomes a matter of highest national strategic importance.

Unfortunately, for anything other than very minor disruptions, restoring the grid becomes a huge and lengthy problem.  The main reason for this?  The US no longer makes high voltage transformers itself.

These days, if we want a new high voltage transformer, we have to order it from an Asian (ie Chinese) manufacturer and wait for it to be built then shipped to us.  Due to their size and weight, they can’t be airfreighted.  A new transformer can weigh up to 200 tons, and they are too large to be trucked to their ultimate destination – they have to travel on special flat-bed rail wagons (and these rail wagons are in short supply, too).

The need to ship by rail adds another dimension to the problem of replacing transformers – as our nation’s rail network shrinks and shrivels, many places that formerly had rail lines leading directly to them have lost their track, leaving different remaining distances for the transformers to somehow be transported from the nearest railhead to the switching power station where it is needed.

Because transformers normally last for about 50 years, and because in much of the developed world, there’s only modest ongoing growth in power consumption, there’s not a lot of manufacturing capacity.  Only 2% of transformers need to be replaced each year, and usually these replacements are planned well in advance.  Most power companies and most manufacturers don’t keep an inventory of spare transformers – a problem made worse by the lack of standardization of transformers.

It is generally accepted that a new order for a transformer will take around 3 years for it to be made and shipped.  If there was a rush on transformer replacements (eg after a solar storm damaged many) then the first 2% of transformers could be made in 3 years, the next 2% would have to wait another year, and so on and so on.  It could take as much as a decade to replace a major series of transformer failures.

And this decade guesstimate assumes that the Chinese manufacturers dedicate all their capacity to our country’s needs, and also assumes they urgently expand their production capabilities.  Can we really rely on other countries such as China – countries that don’t necessarily have our best interests closely at heart and inseparably aligned with ours – to help us when we’re at our most vulnerable?

This article details some more about transformer issues.

Storm Related Outages Are Different

Maybe you’ve had a power outage yourself – perhaps after a windstorm, or perhaps due to some inexplicable thing that you never really were told exactly what it was.  Maybe it was just for a few minutes, maybe it was for a week or longer, and maybe the outage was limited to only a half dozen houses, or maybe it extended over a half dozen states.

Outages are nothing new, indeed, on average, half a million power customers have some type of outage every day.

But – and here’s the catch.  These outages are very different to the ones we are considering.  They are typically due to power poles being blown over, or trees falling on the power lines, or, at worst, a very minor substation transformer blowing.

Fixing these outages simply requires a crew to re-run the power lines, or to truck in another transformer, and maybe to shift some loads in some parts of the grid.

These outages – even when extending over several states – are not due to one, or ten, or a hundred or more of the 2,100 major super-transformers failing, and so are easy to respond to and resolve.

But if we do lose a number of the super-transformers all in close succession, we have nothing to replace them with.  We can’t restore power until we get new super-transformers, some years later.

Not Just Solar Related Dangers – Hackers Too

In addition to the random acts of the sun’s solar storms, we also have to consider more directed attacks on our power grid – manmade attacks.

The easiest way to disrupt the power grid is of course simply to physically blow up transformers.  With only 2,100 key transformers in total, and only a small percentage of those needing to be disabled to impact on many millions of people, and little or no effective security protecting the super-transformers, it is far from unthinkable that terrorists might attempt a low-tech old-fashioned bombing campaign to destroy a region’s power network.

But that is, indeed, a low-tech and old-fashioned approach, and not without difficulty and risk to the terrorists.  A much easier approach is to hack into the control systems – the computers that control the operation of the transformers and the flow of power across the network.

While some commentators say ‘it is not possible to do this’ and promise us that the control computers are secure, they are, alas, talking nonsense.  It serves their purposes to downplay the extent of the risk and the vulnerabilities that are already being exploited, but when you can get people to talk more frankly, for example as reported in this Wall St Journal article, the truth is scary.  Not only are our power control computer networks vulnerable, but they have already been hacked into and compromised.

This is unsurprising.  It seems there is no computer on the planet which is not now connected to the internet, and if we and the Israelis can hack into Iran’s nuclear research and development computers and take them over, causing the computers to run amok and destroy the centrifuges they are controlling, surely other nations can do the same to us.  We’re not the only nation with precocious teenage hackers by any means.

Although the April 2009 Wall St Journal article we linked to immediately above reported – as all such articles do – on how steps are being taken to improve the security of the power grid, here’s a December 2011 article in the Christian Science Monitor headlined ‘Power Grid grows more vulnerable to attack’.

The article quotes an MIT study which suggests that the electrical utilities are creating new vulnerabilities faster than they are patching old ones.  The good news is the cost of improving the grid’s cyber-security is low – about $4 billion.  The bad news – the utilities feel that the possibility of being attacked is too low to worry about, and not worth spending $4 billion to protect against.

The MIT report disagrees and views cyber-attacks on the grid as inevitable.  It isn’t a case of if, it is a case of when.

An interesting related thought – the Wall St Journal article mentioned that some of the cyber-attacks have come from China.  What happens if the Chinese destroy our transformers, then refuse to sell us replacement ones?

More Risks – EMP

We explain what EMP type attacks are, here.

In the specific context of power grids, they have two vulnerabilities in the event of an EMP attack.  The first is the E1 pulse, which could destroy many or all of the control computers that manage the electricity grid.  If the controlling computers go down, so too does the grid.

The second vulnerability is the E3 component, which would be received through the power lines acting as gigantic antennas, and then directed into the transformers and destroying them.

As we discuss in the next section, our grid has become more vulnerable to solar storms; and the mechanism which creates a vulnerability from solar storms is identical to the E3 component effects of an EMP.

How Severe a Problem Are the Grid’s Vulnerabilities?

Opinions differ as to the extent of the vulnerabilities that relate to our power grid.

At one extreme are reports such as this article in Time, which says ‘because we’ve never had a total disruption before, there’s no danger of one in the future’.  That’s brilliant logic, isn’t it, and sadly consistent with much of the non-prepper mindset.

The article goes on to say ‘Don’t worry, all essential services have backup power supplies’.  We don’t find that very reassuring.  Just a week ago, Amazon’s web services had a power related outage.  What happened to their backup power supplies?  We’ve no idea, but we do know that Amazon’s terms of service specifically exempt them from liability in the case of power supply failure.

We also know that the state of the art ultra-sophisticated super-hardened colocation facility where our primary webserver is located has also suffered power failures in the past too, even though they have more in the way of backup systems and redundancies than any two normal computing centers would have.

We wonder further what happens when the backup diesel generators run out of diesel.  If there’s a regional outage of power, there’ll be no diesel being refined, shipped, or pumped.

And, anyway, while it might be a reassuring thought, to some people, that hospitals and internet services can survive for a month or two, what about us?  There’s no backup power supply for regular consumers.  How long can we personally survive, how long can businesses survive, without power?

As well as unrealistically optimistic articles like the Time story above, we also have more soberly realistic articles such as this in Scientific American, which talks about how if a solar storm which occurred in 1921, causing only minimal damage then, was to re-occur now, the result would be a loss of 300 of the super-transformers and  130 million people being without power for years.

Part of the reason we are more vulnerable to such natural impacts is due to the changing nature of our power grid.  We have more and longer runs of power lines now than we did before – in the last 50 years the total length of power line in the country has increased ten-fold, and the average length of each highest capacity line has grown four-fold.  This four-fold increase in length makes it a better ‘antenna’ to receive the electro-magnetic interference from the sun, and for this interference to then overload and burn out the transformers.

The 2011 Scientific American article also says that NASA now has vital early warning capabilities.  We suggest that is an over-optimistic statement – as this article of ours, written a year later in July 2012 points out, NASA and NOAA are still unable to consistently predict and agree upon solar impacts.  In other words, even the more realistic articles are still showing themselves as being overly optimistic.

Summary

The security of our nation’s power grid is a bit like the security of our front door.  Hopefully you’ve never had burglars break into your home.  And you lock your door.  But you know in your heart of hearts that the lock doesn’t really give you true security.

A determined burglar will pick the lock or kick the door out of its frame, and be inside in less time than it takes to read this paragraph.  And a runaway vehicle that crashes into your front door at 60 mph is going through it, lock or not.

That sums up the ‘security’ of our power grid.  A determined hacker/terrorist, or a severe natural event, could destroy it in a flash.  Much or all of the country could suddenly find itself with no power, and the restoration of power could take 5 – 10 years to complete.

We’re not going to guess as to if a grid failure will be due to malicious deliberate attacks by our enemies, or by the awesome natural power of the sun, or through some other random act of chance.

But we do view the risk of a catastrophic long-term widespread failure of our power grid as severe, and creating either a long-term Level 2 or possibly even a full Level 3 situation.  Your response to such a threat has to involve abandoning the city you probably live in now and moving to a safe and sustainable rural retreat.

[suffusion-the-author]

David Spero[suffusion-the-author display='description']

  8 Responses to “Why Our Electricity Grid is So Vulnerable”

Comments (2) Pingbacks (6)
  1. Even if we would start now it would take twenty years to replace and standardize our system.

    A start would be to look and see how many are the same or similar. Look to see which areas are not critical, and if they have working transformers, take them from there and move them to areas of critical need.

    The areas that provide power are most critical. the refineries are primary and the docks that import oil and food. Coal production for power production and home heating are other critical areas. prioritizing where the power goes first is critical.

    Air travel other than emergency should be grounded. Cars that use gasoline should be limited to ten miles a week, diesel limited to fifteen miles for personal cars and trucks.

    This leaves people the ability to get to the market and doctor. Air conditioning will not be allowed. manufacturing other than food production is stopped lawns are not to be cut. This would be a start to limit power consumption.

    With knowing all the transformers that are interchangeable, larger areas and the most critical areas can be restored quickly. Areas that could survive with less power consumption could remove some transformers and put them in areas of need. Areas with uncompilable transformers could be engineered.

    Next the switch gear needs to be standardized and the parts replaceable and rebuildable.

    All this is easy to say on paper but the actions needed are extensive and difficult. The costs could never be absorbed by all the power companies. Because this is part of national security our government as much as I hate to admit it needs to get involved and step up.

    Grampa

  2. If the whole grid goes down from the super transformers being ruined, it’s not even as simple as ordering them, shipping them, and installing them for things to get back to normal.

    The dollar will immediately collapse, so what money will we use? But even more important than that is that the chaos of a collapsed society will make it very hard to get any grid back to normal. Cities will be burnt to the ground. Many of the people that survive will move out of homes. A mutual sense of security will be lost, so getting qualified power plant workers back to their job will be impossible. Regional war lords will rise up and potentially cause problems. Getting the transformers to the interior of the country would be too hard.

    The best thing will be to set up military ports and start electrifying there and gradually work towards the interior as things settle. I’m guessing most of this will be done by our military which is currently overseas (or out at sea). And all this is only if the outage was North America alone. If it was a CME/sun storm that knocks out power grids worldwide, then just forget about it–dark ages part 2.

Leave a Reply

/* ]]> */