Tuesday, March 11, 2014

Techy Tuesday - Potshots on Transformers

By now, everyone has heard the story about what certainly appears to have been an attack on power stations in the Silicon Valley area of California last year.  This week, Lou Frenzel, currently an editor for Electronic Design magazine addressed it in an online editor's blog. 
The main components in our power grid distribution system are transformers.  They step up the generated voltage to a higher level for more efficient transport over very long lines.  Then they step the voltage back down in several stages for final consumption.  These transformers are located at the power plants and thousands of substations.  They are big and right out in the open.  Weather does not bother them, but bullets do.  They are the weak links in our power grid.

Last year someone decided to shoot out the transformers at PG&E’s Metcalf substation in Silicon Valley taking the substation out of service.  Luckily, PG&E was able to reroute the power to prevent a total blackout.  But it took months to get replacement transformers and restore service.

This is a major wake up call to the utilities.   Terrorists, or disgruntled customers, can easily take down a substation with a rifle at long range and get away with it.  The transformers are defenseless as they are not covered or protected in any way.  Furthermore, replacement transformers are hard to come by.
Lou Frenzel is a relatively sane old graybeard, and fairly well known as a technical author; he sees this as a problem the power company engineers need to address.  No calls for gun control or massive government programs.  What bothers me about this is the reaction of ED readers whom I have to see as my fellow engineers.  Yeah, I'm sure some are not, some are students, some are probably loosely affiliated with the profession, but the reactions are bothersome.  The first half dozen comments included censorship - that ED shouldn't have even mentioned this was possible; that the story is paranoid conspiracy theory (?); and that the real reason transformers are vulnerable to rifle shots and few spares are held is "electric utilities must afford despite paying their "non-profit" CEO's millions of dollars in a system allowed to break continually so they can maintain artificially large budgets despite tax-payers being broke."

When I was "but a larval engineer", a graybeard told me "you can't out think someone who isn't thinking".  It's a truism that has stuck with me ever since.   

The problems with spare transformers are (1) these transformers are expensive, so the companies and their regulators probably consider having warehouses full of spares as poor use of capital, and (2) they are enormous - and that means problems storing them - those warehouses have to be enormous (read that "expensive", too). Add to those issues that it appears there is very little standardization of these designs, and very few companies who make them.  Frenzel says there are seven companies in America that know how to make them.  And without someone shooting them, these transformers have a very long life.  With reasonable maintenance, they might last virtually forever.  There are no moving parts.  Needing to keep spares to deal with an attack like this is a major change in how utilities figure replacement costs

As a bunch of gunnies, you and I know that there's no reasonable way to prevent a determined individual with a big bore rifle from getting at them, if the transformers are found in the open as they usually are.  It's really a pretty simple operation, after all.

As a bunch of engineers, the other readers and I should be thinking about increasing the survivability of the transformers.  Hardening the targets.  Perhaps steel or Kevlar curtains.  Where possible, perhaps they could be put under ground, or recessed below where they're very visible.  How about in the bottom of an empty concrete "swimming pool"?  The idea being to obscure the visibility, making seeing the target harder, and then to absorb the energy of the shot.  Perhaps transformers that can run without cooling oil, by reducing losses. 

Obviously, like personal vests, we can't make them "bulletproof" to every level of attack, but this was too easy.  Although we all know how badly reporters get these facts, I read the shell casings found were 7.62x39 (of course they just called them AK-47 cases).  If Bubba or Abdul can take out a substation with just a .30, they need to be hardened.  If we can make an installation that would survive impact from .50 BMG, that's got to be as good as they're likely to get, short of an RPG. 

Compared to hardening against EMP, hardening against bullets is a cinch.


  1. From what I read about that deal in CA a while back, it all says 7.62 by 39.

    That didn't sound right at the time, not a metal puncher. 7.62x51, 30.06, maybe.

    Press could confuse/obfuscate anything in the 30s, I reckon. But all the articles said AK stuff.

    Bad news all around, and this on Drudge today:


  2. Aside from the obvious possibility that a bullet could sever an internal conductor it seems the real risk from rifle fire at transformers is that a bulet causes the oil to drain out and the transformer then overheats until it fails. It seems that for a few dollars (compared to the cost of one of these beasts) you could install a failsafe device to shut them down short of catastrophic failure and to simultaneously sound an alarm to alert both the maintenance people and the police. Plugging a hole and refilling the transformer with oil would be cheap by comparison with the replacement costs. Also these things can be rebuilt, they aren't really complicated devices. Perhaps teams could be trained and equipment positioned so that when this happens again they could be quickly repaied and brought back online.

  3. Siting could be a good first start, though, admittedly, most of them are already in place. You mentioned recessing them.

    You could also set them on hilltops with concrete screen walls - no way to hit that with a rifle bullet.

    What about lots of cedars or the local equivalent for deflecting bullets? You'd pacify the greenies at the same time.

    Of course, once you effectively remove the possibilities for bullets, you have to worry about up close attacks. Maybe you have site security at that point.

  4. PG&E is already spending money to improve security, but obscuring fences aren't bulletproof: http://www.ktvu.com/news/news/local/pge-spend-87m-security-protect-large-substations-a/ndMrk/

  5. I see so many things like this and I'm baffled as to why they haven't happened fifty times already.

  6. According to this article by box-o-truth, it only takes 5 1/2 inches of sand to stop everything up to and including a .308. Sand could be a cost effective shield for the transformers or anything else you need to protect.


  7. Can't remember if I found this years back in the NRA Range Construction Manual or elsewhere, but along with Anon 12:35, above, 3.5 inches of fine, dry, settled pea gravel stops everything shoulder fired (this was in the days before widespread 338 Lapua and 50 BMG availability). IIRC, it was pointed out that using wood 2X4s to build plywood "boxes" to hold the pea gravel was not recommended because the 2X4 was the weak point, so substitute C-shaped galvanized metal studs so the space inside the C holds pea gravel as well.

    I'd think it wouldn't take much to design and build "decorator walls" out of bullet resistant panels of some sort to reduce the probability of bullet strikes. Add impact sensors connected to an alarm/notification system, and you might be <10% susceptible. Cost/benefit, etc.

  8. anyone own one of those 70 year old sks rifles? the standard bullet in 7.63 x 30mm has a steel core. from wiki:
    The 57-N-231 conventional steel-core bullet is designed to engage personnel and weapon systems. The bullet has a steel core. The tip has no distinguishing colour. It can penetrate a 6 mm (0.2 in) thick St3 steel plate at 300 m (328 yd) and 6Zh85T body armour at 30 m (33 yd).

  9. There's just a great group of comments here.

    I was reluctant to say it must have been AK-47s because of this well known joke about how everything is an AK-47 to most reporters. But several of you confirm it, so I'll go with that. That puts me with Bill Anderson in the first comment. I think that the 7.62x39, while respectable compared to some, is not a great choice for the job. Steel core, maybe better. I didn't see anything about them using that.

    I like the ideas about siting the substations differently, but that's probably really expensive compared to putting up sand or pea gravel berms - or other defensive measures like Anon0724 said about pea gravel with sensors to trigger alarms.

    And once these targets are hardened, it moves to another soft target. In perpetuity.

  10. Anything that will poke a hole and cause the oil to leak out will do the job.

    7.62x39 is an interesting choice because with heavy bullets it can be loaded subsonic. I don't know if that was the case here, but it is an option for someone who wanted to poke holes in transformers, use an oil can suppressor, and minimize the acoustic signature to the action cycling plus a "pop."

    Or it could have been someone smart with a bolt action 308 Win who policed up all their own brass and left some 7.62x39 empties on the ground to confuse forensic investigators. Who knows?

  11. There's no perfect solution, and it's always subject to cost/benefit analysis.

    I'm confident the electrical power industry never considered people with rifles much of a threat; sure, Harry Hunter may miss a deer and hit a transformer, but that's an isolated case. Tommy Terrorist emptying a magazine or three into a cluster of transformers wasn't on their SWOT analysis list.

    Just as one cannot totally prevent home burglaries, but merely reduce the probability by making your house a substantially less attractive target than your neighbor's, power companies need to perform an end-to-end SWOT to see where the new weak points will be once they harden transformer clusters.

    The grid will always be a target; I ran across something recently indicating that if the 8th Air Force had relentlessly targeted electrical generation and distribution facilities - which are fairly easy to identify and somewhat hard to hide - they would have brought Germany to is knees sooner. One doesn't have to destroy the distributed manufacturing of ball bearings if there's no electricity to make them, no matter where the manufacturing equipment is.

    That leads to suggesting that our electrical system is a national strategic resource that deserves more attention in that regard.

  12. I'm puzzled as to why this event from nearly a year ago is suddenly big news - it wasn't then. Cui bono? It seems utilities kw charges are based in part on capitol investment. Perhaps a gee whiz solution has been developed to address this created problem. You know, like shoe sniffers and whatnot. While commie 7.62 steel core is perfectly capable of punching through 1/4" mild steel with energy to spare a serious actor would likely choose a more robust round, and would not need 100 rounds to make a statement - depending of course upon what the statement was intended to convey?

    And the utterly superfluous acts of fence cutting and SCADA system disabling - what message does that send about the actors involved?

    Hardening seems an exercise in futility - put all the berms and baffles in place you desire, Joe Schmo could drop chunks of chain from the air onto conductors, from a position of comfort and ease some distance away. Or whatever. And then there are feeders...

    The problem is centralized production and distribution.


  13. Itor - I wondered the same thing, but I think that when you look at all the "our power grid is so vulnerable" stories, some group is angling for some federal money.

    My point in the story was that I was actually embarrassed for my profession at the comments. The censorship aspect, the class-envy crap about paying CEOs their millions, nonsensical crap like that, which I detest. My secondary point is that as engineers, we should be focused on fixing problems, not that other crap.

    Of course these are soft targets. We are awash in soft targets. The political game seems to be finding you have some reasonable claim to being an important soft target and hooking up to the infinite money hose coming out of DC, where they drip billions of dollars like a toddler dripping turds out of his Pampers.