Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

This electrical transmission tower has a problem

twitter.com

Post reply on HN
Print view View original

191–200 of 479 posts

Re: This electrical transmission tower has a problem

#191

Post by c3534l »

jrockway wrote: post #15

I guess my takeaway from this is if you're a government building some sort of asset you're going to transfer to the private sector, overengineer it. Nobody is ever going to do maintenance, so your thing essentially has to last forever. Imagine that they made the hooks so they only lasted for 10 years -- we'd have 10 times as many billion-dollar fires. But some engineer in 1918 said "you know, let's make it a little b…

If they only lasted 10 years, it would have been replaced with something easier to service by now.

Re: This electrical transmission tower has a problem

#192

Post by TeMPOraL »

taneq wrote: post #170
legulere wrote: post #168

Earlier quoted context omitted.

1000 fires would actually be probably better than 10 because they help keep flammable material at bay and reduce the size of the fires.

So your saying we should introduce chaos monkeys into power grids to improve robustness?

No, GP is saying you need to start doing controlled fires on a regular basis, to reduce the accumulation of flammable material. Power grid is not related to that.

Re: This electrical transmission tower has a problem

#193

Post by harha »

onetimemanytime wrote: post #184
harha wrote: post #165

Earlier quoted context omitted.

Take driving as an analogy - there's a non-zero risk of having a crash, but it's significantly higher if you don't follow the rules (drunk, speeding, broken car). This feels like PG&E was betting it wouldn't happen to them. I think for a utility the risk should be very very close to zero, you can really do a lot of planning around it and make sure that if you're running infrastructure in a dangerous area to take suff…

>> I think for a utility the risk should be very very close to zero They can lower the risk by a lot, but are you ready to pay 65 cents (or whatever) a KW ? PG&E, with all its faults, has been put in an impossible position: one fire (that is bound to happen no matter what because wires fail) can bankrupt them since it can burn hundreds of thousand of acres, thousands of homes, untold deaths and other property destruc…

I understand the constraints and I'm not blaming them for everything, creating this dangerous situation is also a great issue, but I think this was handled very badly on their part as well.

As a utility they should have a set of rules and guidelines on how to operate safely. If it isn't possible in a situation, they should say no to the other party. With the rules they should be able to point at exactly what is missing and get the other side to fix their part (e.g. the risk or get them to pay extra for maintenance) or stop operating that bit. This looks like they intentionally looked the other way.

Re: This electrical transmission tower has a problem

#194

Post by noselasd »

skytreader wrote: post #183
tshaddox wrote: post #155

Earlier quoted context omitted.

> With proper engineering and maintenance, PG&E could probably reduce the frequency of fires significantly, but there's always going to be a non-zero risk of fire, both from power lines and other sources, and it seems basically inevitable that one of them would have started another fire at some point. I don’t understand. You’re conceding that if PG&E were a competent and responsible company they could significantly r…

I think the point is PG&E should be blamed for negligence but not for the fire[1]. They could follow all regulation to the letter and there will still be a non-negligible chance that their infra will cause a fire simply because existing regulation is not suitable for the environment their infra stands at. [1] Although I think the thread is rather vague as to the damages assigned to PG&E no?

That's usually not how laws works though. If you act irresponsibly, break laws, don't perform due dilligence and so on, you are also responsible and get blame for the consequences.

If you act in good fait, follow all regulations, act responsible, the law might see it as an accident where you don't get blame.

Re: This electrical transmission tower has a problem

#195

Post by harha »

Zanni wrote: post #185
harha wrote: post #165

Earlier quoted context omitted.

Take driving as an analogy - there's a non-zero risk of having a crash, but it's significantly higher if you don't follow the rules (drunk, speeding, broken car). This feels like PG&E was betting it wouldn't happen to them. I think for a utility the risk should be very very close to zero, you can really do a lot of planning around it and make sure that if you're running infrastructure in a dangerous area to take suff…

An even better analogy is Jenga. California's uncontrolled fuel load was going to result in massive fires at some point. Yes, PG&E could have reduced the chance of it being them, which would be great for PG&E, but would make no difference to California.

Yes this makes sense - back to driving, it's like a road that clearly wasn't safe to operate, but the drunk driving didn't help either.

Re: This electrical transmission tower has a problem

#196

Post by zaroth »

bhawks wrote: post #154
zaroth wrote: post #93

Earlier quoted context omitted.

The average age of a PG&E tower is 68 years old.

PGEs own linemen were concerned about wear on these towers. They sent samples from the field to labs for metallurgical analysis. PGEs own scientists estimated that the components would wear out in 100 years. PGE took this information and did nothing . https://pbs.twimg.com/media/EiErOkaUMAA178t?format=png&name=... This is not how you maintain critical infrastructure. This is embracing failure and simply not caring ab…

So they staff and run a metallurgical lab in house to which linemen can send samples for analysis, and then management simply throws away the reports?

Re: This electrical transmission tower has a problem

#197

Post by chaz6 »

extropy wrote: post #190
taneq wrote: post #170

Earlier quoted context omitted.

So your saying we should introduce chaos monkeys into power grids to improve robustness?

Actually yes, since they have a huge incentive to Run to Failure. Chaos monkeys would provide a consistent source of failures to make them do their maintenance. Similar to red teams in security.

That makes sense, but can we do it in winter when it is less likely to start a fire?

Re: This electrical transmission tower has a problem

#198

Post by gorbachev »

pavlov wrote: post #62
burlesona wrote: post #41

This is interesting but not really surprising. Most folks don't realize, because we don't stop and think about it, but the US has unbelievable amounts of infrastructure in this sort of condition. When things can last 40-100 years with little to no maintenance before catastrophic failure, there is _extraordinary_ incentive to defer maintenance until catastrophic failure occurs. For every mayor / governor / utility dir…

> “Most folks don't realize, because we don't stop and think about it, but the US has unbelievable amounts of infrastructure in this sort of condition.“ With respect, it’s obvious to anyone coming to USA from elsewhere in the world. Observing America is a strange mix of “wow, these people were so early, they really invented most of this stuff” and “wow, they really haven’t updated it since”. (The American banking sys…

Great comment, completely agreed. This is one of the first things I noticed after moving to the US. Everything is run down, except lobbies of Fortune 500 corporations.

Re: This electrical transmission tower has a problem

#199

Post by techie128 »

legulere wrote: post #168
tshaddox wrote: post #155

Earlier quoted context omitted.

> With proper engineering and maintenance, PG&E could probably reduce the frequency of fires significantly, but there's always going to be a non-zero risk of fire, both from power lines and other sources, and it seems basically inevitable that one of them would have started another fire at some point. I don’t understand. You’re conceding that if PG&E were a competent and responsible company they could significantly r…

1000 fires would actually be probably better than 10 because they help keep flammable material at bay and reduce the size of the fires.

This is a ridiculous argument. Why don't we forcefully tear down all of PG&E's infra every few years. They'll be forced to rebuild it with the latest safety standards... smh

Re: This electrical transmission tower has a problem

#200

Post by chmod775 »

londons_explore wrote: post #146
nullc wrote: post #86

Earlier quoted context omitted.

> But many things can negligently cause sparks, including dozens of human and natural causes. My favourite: https://www.latimes.com/local/lanow/la-me-ln-mendocino-compl... Someone tried to stop an angry yellowjacket nest by driving a concrete stake into the hole in the ground and their hammer thew off sparks and started a 459,123 acre fire-- the largest recorded in state history at the time.

I suspect that man was lying... It is very very very hard to start a fire with a hammer, even deliberately... Yet if you had a wasps nest, you'd totally be tempted to douse it with a bit of gasoline, believing you could put it out before it spread to anything else... Except you couldn't... And when the investigators came round, you'd totally tell them about the stake but not the petrol...

> I suspect that man was lying... It is very very very hard to start a fire with a hammer, even deliberately...

Out of the millions of people who use a hammer to strike metal on metal every day, one of them will manage start a fire eventually.

Even the unlikeliest events are actually pretty likely at scale.

Sure, two thirds of them may just be lying, but unless you can prove they are (which really shouldn't be that hard, considering burning fuel leaves obvious traces), you should assume each of them are telling the truth.

Post reply on HN