Hacker News new | past | comments | ask | show | jobs | submit login
Massachusetts gas fires: Another technological tragedy (bit-player.org)
216 points by greeneggs on Oct 20, 2018 | hide | past | favorite | 123 comments

"All of these events feel like unnecessary disasters—if we were just a little smarter, we could have avoided them—but the fires in Lawrence are particularly tormenting in this respect. With an aircraft 35,000 feet over the ocean, you can’t simply press Pause when things don’t go right. Likewise a nuclear reactor has no safe-harbor state; even after you shut down the fission chain reaction, the core of the reactor generates enough heat to destroy itself. But Columbia Gas faced no such constraints in Lawrence. "

Pretty interesting comparison. Arguably airplanes can coast down and nuclear reactors (especially Gen-IV low-pressure ones like sodium metal or molten salt) can passively cool their nuclear decay heat post shutdown without external power. But it is true that TMI and Fukushima were not chain reacting, they were both decay heat cooling problems. For those unfamiliar with this, a chain reactor immediately drops to about 7% of full power after shutdown and then exponentially decays to 1% after a day and half a percent after weeks. Turns out 1% of a gigawatt is still a lot.

On the nuclear topic, I'm annoyed that natural gas accidents that cause deaths get way lighter media treatment than nuclear accidents like TMI that cause no deaths.

Turns out 1% of a gigawatt is still a lot.

Just to be pedantic, nuclear plants generate electricity at about 33% efficiency. So to get 1000 MW of electricity, you start with about 3000 MW thermal.

Turns out 1% of three gigawatts is even more. :)

Truth. The ones I'm designing are generally like 38% efficient and 300 MWe scale so pardon my slightly unconventional sizing.

It's definitely a lot. But nothing naturally-circulating sodium metal through 4 redundant NaK-air dump heat exchangers can't handle.

HN, the place where an article using an analogy of nuclear reactors probably means that an actual nuclear reactor designer will comment on state-of-the-art nuclear reactors.

And still get pedantically corrected on a minor point! What a place this is!

Is that a positive or negative statement?

I've spent a few years in the CANDU world so it's interesting to hear opinions from elsewhere. I've grown to appreciate the lag time the design gives you.

NaK-air? Doesn't NaK catch fire in the presence of air?

That's kinda the point of a heat exchanger. The water in your car's radiator doesn't make the engine bay wet either...

Probably a giant radiator. Lots of equipment that is “air cooled” has an intermediary coolant that is actually cooled by air over a heat exchanger (radiator).

That's confusing. "Air cooled" usually means exactly that there is no "intermediary coolant". At the end of the day almost all equipment cools to the atmosphere after all. The difference between liquid cooled and air cooled cars/bikes is exactly that extra equipment of coolant lines and radiators versus just having airflow directly over the things you want to cool.

I wasn’t very clear, if the NaK is liquid it could flow through a radiator and could still be considered air cooled.

While there are some pure air cooled engines, lawn mowers, motor cycles, etc. A typical car engine is also considered air cooled since it gives off its heat to atmosphere even with assistance of a coolant radiator system.

Most marine engines, some power plants and some industrial equipment is considered liquid cooled since the coolant passes through a fluid to fluid heat exchanger and then that cooling water gets put back into the ocean/river. An example would be some ABB drives come in a LC liquid cooled or AC air cooled package. The motor drive lineups are the same with primary coolant going through all the drive bays the difference is at the end if they put an liquid to air or liquid to liquid heat exchanger.

> A typical car engine is also considered air cooled

We just have different definitions then. In car circles your typical car is definitely water cooled. I had never seen your definition but it also makes sense. The typical one is describing what fluid contacts the hot parts and yours is describing what mass the heat is transferred to.

This gets a bit blurry in (some) motorcycles.

My late 90s era Ducati is considered my most people to be "air cooled", but it has oil galleries that circle around the cylinder walls and an air-oil radiator. (Pedants might call it "oil cooled", but those are mostly the same people who "well actually" you when you call it a V Twin. "Well, actually, it's an _L_ Twin..." - just because it's a 90 degree V Twin rolled forward so the front cylinder is roughly horizontal and the read is mostly vertical...)

(And to further complicate matters, the 900cc two valve "air cooled" Ducati motors have those cylinder cooling oil galleries, but the 750cc motors also have oil coolers, but do not circulate the oil in the cylinder walls. The 600cc motors use the same oil galleries as the 750s, but do not come with the oil cooler...)

Airplanes coast down fine after a (double) engine failure, but they don't coast nicely when your primary instruments aren't working. The Air France flight didn't just loose airspeed indication, they got invalid readings and all associated warnings. Unfortunately that is the common failure mode for flight instruments, they don't just show nothing, they show wrong values. Fortunately pilots are trained to know how instruments work on the inside and what the most likely wrong indications and their causes are.

The problem the author points out is that you cannot "pause" to think about what information may be correct and which instruments to follow. You have to react immediately, which is much harder than considering the same decisions sitting at a desk reading the NTSB report.

If the low speed indication/alert is correct you need to push the nose forward immediately to keep the aircraft out of a stall, if the high speed indication/alert is correct you need to pull power to flight-idle and nose up carefully but immediately to avoid overspeeding the aircraft and risk structural failure.

AFAIK the Air France flight would have been ok if they had not even touched the controls. They had a lot of altitude. Of course easy to say from the arm chair.

I'm not sure how easily a regular airline overspeeds or stalls if the controls are close to a middle position. They should be aerodynamically stable (if loaded correctly).

For a fighter jet it's a different situation.

They were initially close to a high altitude stall, the margins are not that large at that altitude, so just dropping everything isn't really an option.

The report highlighted training, which I guess is always possible to point out, but looking at the event I think they could have made more conclusions regarding how the control system performed counterintuitive by emitting stall warnings at the time they actually did the right thing, and provides little feedback when it drops into alternate law making a standard full pullback into a potentially dangerous move.

What I remember reading of the Air France flight was that both people flying understood perfectly well the situation, and had plenty of time to get out of it, but one panicked and tried to pull up even though that was exactly the wrong thing to do. The other problem was that no one else realised that one of them was trying to pull up and the plane averaged the actions of the two sticks (pull up and dive) causing effectively zero net action. (Other planes feedback each stick's action to the other so the other person would soon have realised what was going on.) They had so long to fix the problem that the captain (who had been sleeping) had time to wake up and enter the cockpit, and he finally managed to coax out the information that one of the copilots was trying to pull up, and for them to stop. By that point they were only seconds from the water and it actually was too late.

Number of deaths isn't the only metric of severity. For example, there is also the number of people who lost their homes (evacuated and not able to return for many years), which for Fukashima was apparently 150,000. [1]

Since nuclear power plants are so expensive to build, even just the loss of plant itself is in the billions.


> ... which for Fukashima was apparently 150,000

To put it in perspective, the prediction for the number of climate refugees vary, but papers seem to put the number somewhere between 150 and 300 million by 2050.[1]

That's about one Fukushima every five to eleven days, starting from now and continuing until 2050.

Think about it.

[1] https://orbi.uliege.be//bitstream/2268/136972/1/11-1188-mr7-... (Table 1, page 5) - Disclaimer: this was the first link from Googling "climate refugee prediction".

To put that into perspective, about 120,000 people are displaced every day in the middle east because of a mess ultimately created in the pursuit of oil interests.

That's one Fukoshima every day.

Apparently the US felt that hundreds of thousands dead, millions displaced, and democratic governments overthrown was an acceptable price for slightly cheaper oil. I fully expect that calculation to come out the same when climate refugees are considered.

The lengths we'll go to for cheap energy and car culture, right?

Do you have a source for that number? UNHCR lists 44,400 displacements per day worldwide in 2017 [0].

[0] http://www.unhcr.org/globaltrends2017/

I just took the first number I found in the excerpt of an article in my Google results.

Since I can't find it anymore I'll say that my number is very likely wrong and your's is right.

Sadly can't edit my post anymore though.

Not sure why you were downvoted. You're entirely right. Oil is incredibly destructive for the environment and society, yet it's accepted because that's just how things are.

Our entire civilization runs on energy, mostly from coal and oil. Until recently it was not feasible to use alternatives. Change takes time.

"Recently" was — conservatively — thirty or forty years ago.

Time which we might not have, especially given the political pushback.

> Turns out 1% of a gigawatt is still a lot.

Reactors are circa gigawatt electrical. Nukes usually have a thermodynamic efficiency around 1/3rd, so a 1GWe reactor will produce 3GWt at the core. That's the heat they have to deal with when uncooled.

> On the nuclear topic, I'm annoyed that natural gas accidents that cause deaths get way lighter media treatment than nuclear accidents like TMI that cause no deaths.

Why this annoyance? I'd prefer the media for the most part shut up about energy accidents in general, considering how damaging their outcries have been for nuclear power. A few sober reports are fine. It's probably a minor miracle the natural gas industry hasn't suffered like the nuclear one.

Heh, I guess you're right: I'd prefer both nuclear and natural gas to get sober reports. I think natural gas has a better name (it's natural! As if uranium is not), no ties to superweapons, and no The Simpsons. Regular fire is just easier to understand to most folks than radioisotopes.

With natural gas incidents, the entire surrounding area doesn't become unusable for hundreds of years...

No, but over the past 50 years, we’ve had two severe nuclear incidents resulting in any kind of long-term environmental damage (Fukuhima and Chernobyl), and Fukushima will be remediated within a decade.

In the mean time, we’ve had Exxon Valdez, Deepwater Horizon, Andover, San Carlos, CA, plus oil trains burning down entire towns, coal mine explosions, two Gulf Wars, and sea level rise due to global warming.

Pick your poison.

Right. And the evacuation now in most of fukushima is based on old and likely very conservative dose-health relations (linear, no threshold). There is a lot of talk these days of getting a better understanding of the real risk of low dose radiation. Some people evacuated Fukushima and went to places with higher natural background and ironically got more dose

I find this to be an interesting comparison: https://en.wikipedia.org/wiki/List_of_pipeline_accidents

The list of accidents is so large that the US has to create separate wikipedia pages for every 25 year span.

> The list of accidents is so large that the US has to create separate wikipedia pages for every 25 year span.

Actually it has a separate list page for each year since 2000. And the 25-year lists before ("List of pipeline accidents in the United States (1975–1999)" etc.) are huge lists.

We also have burning coal mines that slowly eat whole landscapes like in Centralia https://en.wikipedia.org/wiki/Centralia_mine_fire

> Fukushima will be remediated within a decade.

Do you want to put some money Fukushima being fully decomissioned by 2028?

I think you are only considering the damage done on land, but you severely underestimate the damage done (being done) to the ocean, which is difficult to assess though all studies I have seen so far tend to agree that anything living there is having a rough time.

I wonder how much land area has been rendered unusable from pollution by oil, coal and gas compared to nuclear. I suspect even by this measure nuclear would prove far preferable.

The difference is many tens of thousands of smaller sites affected by fossil fuel pollution and accidents as compared to a couple of large incidents.

Other kinds of consequences like cost of mitigation, value of lost property (and lost value of property what wasn't destroyed by itself - you can easily see thousands of homes near, but not in evacuation zone deprecating severely), costs of health checks for hundreds of thousands of people etc. - are worse in nuclear accidents.

Fans of the nuclear industry doesn't like to columnize the costs of 'externalities'.

For one, they rarely mention the amount of US government subsidies ($hundreds of billions) poured into getting the early plants up and running. For another, the unwillingness of insurers to take an interest in those operations.

When Rocky Flats spilled a little Pu dust into the Denver suburbs, they just raised the 'safe' exposure by a factor of 10. Easy peasy.

> are worse in nuclear accidents

Do you have numbers for this comparison?

Regionally this was covered extensively with pictures of devastated houses and graphic detail about the fatality.

There are modern controllers which will detect this situation:

"When Open Loop Detection is enabled L.dE, the controller will look for the power output to be at 100%. Once there, the control will then begin to monitor the Open Loop Detect Deviation L.dd as it relates to the value entered for the Open Loop Detect Time L.dt. If the specified time period expires and the deviation does not occur, an Open Loop Error will be triggered. Once the Open Loop Error condition exists the control mode will go off and an Open Loop message will be display. If the process value goes in the opposite direction, a Reversed Loop message is display. The sensor is likely wired in reverse polarity."

If this crew was replacing an old cast iron gas main, it probably predated such controllers. That's not necessarily a bad thing. A mechanical controller can work for many decades. It's hard to get that level of uptime from microprocessors.

The real problem here is "maintenance induced failure". Especially on a system which is not fully shut down.

[1] https://www.watlow.com/-/media/documents/user-manuals/pm-pid...

The Chemical Safety Board produces some pretty fascinating videos that reconstruct the chains of events that led to large industrial accidents in chemical plants. The videos normally conclude with recommendations for voluntary actions industry can take to prevent similar incidents in future. They basically summarize the results of detailed investigations by CSB agents.

The videos are well produced and, apart from those accidents that involve injury or death, they are quite entertaining if you happen to be in a forensic frame of mind.

Here’s an example of one of their videos: https://youtu.be/_icf-5uoZbc

I like their videos. Unfortunately, this incident will be investigated by the Transportation Safety Board.

The report will be thorough, but we may not get a great video out of it :(

The TSB does do videos, but you're right that they're not as polished as the CSB's.

These are produced like one of those Air Crash Investigation type shows.

Why are these actions voluntary? Shouldn’t there be regulations that at least ensure we don’t make the same mistake twice?

In some cases it allows the agency to operate freely regardless of the political environment and in others, there are historical reasons for the way they operate (ala NTSB and FAA).

It also allows the market to decide what risks justify the cost and which ones don't via the insurance market. These advisories are front and center in lawsuits when dealing with repeat occurrences so the pressure comes from economic forces. Thankfully, there are limitations to what the market is allowed to decide so this may eventually become a regulation through public pressure or agency review.

Oh man. My PhD work is focused on helping people communicate clearly about the measurements they are making and how to interpret them and where the pitfalls are in a research setting. Every time I read about something like this I wonder if I can use what I am working on to help people automatically detect when the assumptions of their measurement model could be violated. For example, if you can explicitly model the fact that the measurement is not the actual value and that they can become decoupled (and can thus brainstorm and model potential reasons), then you can start to help people realize that the map (measurements) are not the territory (reality), in ways that aren't just cute phrases without guidance for practical implementation.

I'm curious: within which field are you doing this PhD? I could see you as a computer scientist, systems designer, political theorist, perhaps an economist or philosopher.

Is any of your work available to read online?

I'm technically in neuroscience, neuroinformatics more specifically, but was raised by a political theorist. Unfortunately I have a bad habit of writing code rather than prose, and don't keep a blog or similar, so the first thing that I could point you to would be my thesis, which is still months away.

Well, hurry up then :)

Seriously, this sounds like a very interesting subject. Best of luck with your thesis writing and submission (mine's still a couple of years away).

Post the thesis to HN when you're done, or PM me (my Gmail username is the same as my HN username) and I'll host it and post it. Better communication tools for statistics and technical subjects is seriously something that the HN community could use and promote.

I see parent has responded, but it could be "human factors" as well.

If you havent yet you should check out Sidney Dekkers writing. He does a lot of work around human factors and perception in the area of safety & operations management.

While they were replacing an old system with a new system, they had inadvertently created an interim (hybrid?) system that wasn't as safe as either the old or new:

> The cause of the accident was not a leak or an equipment failure or a design flaw or a worker turning the wrong valve. The pressure didn’t just creep up beyond safe limits while no one was paying attention; the pressure was driven up by the automatic control system meant to keep it in bounds. The pressure regulators were "trying" to do the right thing. Sensor readings told them the pressure was falling, and so the controllers took corrective action to keep the gas flowing to customers. But the feedback loop the regulators relied on was not in fact a loop. They were measuring pressure in one pipe and pumping gas into another.

"I admit to a morbid fascination with stories of technological disaster. I read NTSB accident reports the way some people consume murder mysteries. The narratives belong to the genre of tragedy."

The author is not alone. Over a decade ago, there was a theater piece called "Charlie Victor Romeo" that consisted solely of actors re-enacting cockpit voice recordings of aircraft that crashed. http://charlievictorromeo.com/

What I still don't understand is why the pressure regulators and over pressure valves on the house gas meters didn't kick in? Were the homes involved so old that the meter protections failed or weren't there in the first place? It seems that ultimately you've got to have failsafes at the homes themselves. Without such failsafes, what would stop a bad actor from purposely over pressurizing a residential branch?

Those regulators work within a set range of input pressures, and need a certain delta to work properly. Over-pressure them severely and they will simply fail, and most such failures will lead to gas escaping from the regulator body. That is why you will usually find them outside of the premises they protect so they vent into the outdoors rather than into some enclosed space.

As for your bad actor: what would he pressurize your lines with? An air compressor? He'd have to dig up the lines or disconnect them first and gas from the line would likely escape in quantities large enough to discourage such tricks.

Just like in theory you could disconnect the mains from a house and then send a high voltage pulse down the feed lines, in practice pranksters and miscreants tend to avoid doing stuff that might get them killed instead.

>What would he pressurize your lines with? An air compressor? He'd have to dig up the lines or disconnect them first and gas from the line would likely escape in quantities large enough to discourage such tricks.

On the gas meter on my house anyway, the underground pipe mates at a valve. It doesn't seem hard or particularly dangerous to shut that valve off, disconnect the meter, connect whatever, then open the valve again.

If you leave the valve attached to the mains line then you can access the house lines, usually there is even a special port for this that you could use without disconnecting the mains line that is used for leak inspection (they evacuate the lines and measure the rate of seepage). Once you connect something in the line and you open the valve again all pilot lights will have been extinguished and won't re-light due to safeties.

Anyway, if you want to destroy someone's house there are much quicker, less obvious and easier ways to do so.

What would those devices do? A pressure regulator is designed to handle a range of input pressures. If the pressure really got 75x higher than the design pressure, the regulators probably just failed.

Are there really no pressure release valves that will blow out at a certain over pressure? Like a spring release that will give way and release all of the gas outside the house.

Supposedly, old lines like these are low pressure lines to the house. Thus nobody has a pressure regulator in their home. It seems crazy today but these lines are antique, from the days when electrical wire was run by knob-and-tube.

Residential gas piping hasn't changed much over the years. Threaded steel pipe from 1910 should be able to hold a couple orders of magnitude more pressure than it should ever see.

The real determining factor is not age but whether the installer was lazy and hand tightened it all (which you can get away with at 1/2psi, but not much more).

If there's no pressure regulator (due to systems being designed that way a hundred years ago), and the pressure spikes, it doesn't matter how the pipes were put together- your pilot lights become blowtorches.

It does seem odd, and I’m guessing that such protections simply do not exist or the article might have mentioned it. Kinda like running a network with no firewall if you ask me. “I’ll accept whatever comes down the pipe” is a bad idea whether it’s a literal pipe or Ethernet.

I had believed such a catastrophe was all but impossible. The natural gas industry has many troubles, including chronic leaks that release millions of tons of methane into the atmosphere, but I had thought that pressure regulation was a solved problem.

Yeah, natural gas tech is old and over time has been worked into a multi-layer system that is close to "inherently safe". I remember years ago trying to light an old heater the most stupid way possible and only singeing my eyebrows in the resulting explosion.

But these events just show that if a company skimps enough on needed maintenance, ignoring constant smaller leaks and the similar despicable maneuvers done by this gas company, then the overall potential of piping an explosive gas into a city can be realized and people die and are injured, building are destroyed and so-forth.

In this context, it seems pretty obvious this isn't a "technological tragedy" but a "people doing bad things tragedy" and I hope it's obvious people deserve to go to jail for this and if they don't then other problematic aspect of this society are then to blame.

The gas company leaks may be insignificant compared to the way natural gas is obtained. In particular, the way it isn't. Dakota fracking just burned off the natural gas, because in the USA its politically incorrect to build pipelines. Result: terawatts of natural gas burned off at the source or released into the atmosphere. More damage to the environment than a century of leaks. The burning flumes could be seen from space, appears as a megacity of light for years.

I think our political process has completely lost touch with science and pragmatism.

Your argument connecting blowing-off natural gas and the current antipathy to pipelines today is false and disingenuous.

Essentially, it follows the logic that the only way to keep corporations from devastating the environment is giving them some positive market incentive to not do so. But of course, given zero regulations or morality, there's always going to be an economic incentive to toss some poisons in lakes and some pollutants in the atmosphere - because some things just aren't useful and if you have zero-cost disposal, that the (amoral) path of least resistance.

I'd go the opposite rout. Polluting in whatever form should be illegal. Polluters should be fined or go to jail, whatever is appropriate. "I could make money with my stuff so I tossed it/burned it/whatever" should never an acceptable explanation or excuse.

Uh, its true and the direct cause.

Billions of cubic feet of gas is not a reasonable disposal problem - there's no place to put it, that isn't worse than burning it. Except a pipeline, where it would be ultimately burned anyway, but now yielding energy.

I wish we could just separate technology and human factors like this. In reality, each system is designed to prevent neither technological malfunction nor miscommunication or errors in process.

Unfortunately, programmers seem often to be isolated from this really complex part of engineering, so this interplay might not be appreciated properly.

Donald Norman's Design of Everyday Things helped cure my technophilia, maybe even turning me into a humanist. http://a.co/d/dSsOfxJ

Aha, looks like there's a new version (dated 2013).

Ah, referal link. How sneaky.

Crap. Unintentional. I just cut & paste. I'll be more careful in the future. Thanks for the heads up.

I suspect the remote-sensing aspect of their regulation system was not well known, and thus the consequences of possibly opening the control loop weren't taken into consideration. It's a subtle but important point --- and I remember a friend telling me a story of destroying some very expensive electronics because the sense line of the power supply came loose.

What I see in those incidents is severe lack of backup sensors.

Why after switching to new pipe nobody checked/could check the pressure in it?

Why there were no other kind of speed sensors on a plane, simpler, less accurate, slower but not that vulnerable to icing?

Why in Three Miles Island incident why people couldn't double check if reactor has water in it or not?

Properly installed residential gas piping should be able to handle triple digit pressures. Appliances should be able to handle double digit pressures because almost all appliances that can run on NG also can be equipped to run on propane which outputs at up to 30psi depending on the regulator. Manufacturers slap a safety factor on top because it's a consumer product and you don't want someone to blow up their house and sue you because they used the wrong regulator.

There's a reason some houses went bang and some houses didn't.

I know it's fashionable to blame the megacorp (and it sure looks like they have plenty of blame in this case) but there's plenty of blame to go around here. There's a reason some houses went bang and others didn't. I've worked residential construction in MA and plumbers as a group have a pretty bad reputation (well earned, at least based on my experience, you should see some of the corners these guys will cut). I would wager that the houses that went bang weren't the ones that had their gas pipes done by the apprentice who still does everything by the book but the "seasoned professional" who knows exactly what you can get away with when your work only needs to hold 1/2psi.

Like any large accident there are many dominoes that need to be lined up before they can all be knocked down at once. I think it's foolish to act like some contractor dutifully carrying out Colombia Gas's faulty work order is the only cause of this.

Even my air compressors have blow off valves, did something get lost in the advancement of technology from the boiler days ???? it seems to be a pretty stupid design that doesn't have relief valves.

I don't know much about the specifics of gas distribution, but I came across a comment on an entirely different discussion recently which I think is apropos:

> you should be very suspicious of any conclusion that requires you to assume that all the world’s experts have missed something extremely basic.

It seems to me most likely that the answer is simply that releasing natural gas into the air is an even more dangerous failure mode than overpressure - natgas is not air in a compressor. But regardless I'd bet you dollars to doughnuts that, for one reason or another, blow-off valves are a bad idea in the context.

I’m not sure I agree with the statement—humans overlook basic details all the time. Most computer bugs are “extremely basic”. Multiply times possible failure points and it’s easy to see how basic flaws can easily cause systemic failure. Experts DO miss extremely basic things daily. You have to actively build proccesses to avoid this. Expertise is not enough!

> you should be very suspicious of any conclusion that requires you to assume that all the world’s experts have missed something extremely basic.

This is hugely mistaken. If you're an expert in compliance in one field, often your insights are valid in another field.

For example, anybody who has certified a data center knows that Fukushima was both in the wrong location and had wonky power distribution.

Anybody in aviation knows the Tyndall F-22s were in the wrong location after Kermit Weeks' collection was destroyed in Florida in 1990.

Expertise and common sense are always in short supply. It's 2018 - we've collected enough hindsight for the next 100 years.

You could flare it, but you would need some combination of space or a high enough tower.

Use a pilot light to ensure it can always function after the release valve/disc blows.

I see your logic, but the fact that overpressure demonstrably caused a catastrophic failure sets the bar pretty high for blow-off having worse consequences.

Compressed air can be released back into the atmosphere from which it came without much danger, but gas is not so harmless. Relief valves exist to prevent pipes from exploding due to the pressure, and during this incident they were probably wide open, but they just continued to release gas which eventually exploded anyway.

> it seems to be a pretty stupid design that doesn't have relief valves.

They have, the problem here was that the relief valves vented way too much gas - they're in the houses of people... and one spark with the right amount of oxygen and stuff goes kabooomm. Gas fires are no joke, gas explosions even less.

There are pressure relief valves, but they (a) can't handle that much overpressure and (b) many of them are located inside the houses (at the gas meter, which on many houses around here is still indoors).

This is why all the new gas meters in SF have their relief vents piped outdoors - much lower risk of explosion. If you see those little squareish-looking pipe ends with a fine screen mesh on the end that’s what they are.

Why not install the new gas meter outside? Or do you mean replacement gas meters?

The usualy idea of the valves is for them to open before the overpressure exceeds their design capacity.

A blow-out disk isn't going to fail to operate. My pressure-cooker has one. It's just a piece of material that will catastrophically open the vessel to the atmosphere before massive overpressurization.

Well...that's a very unfortunate example. Pressure cookers' blowout disks can get clogged by whatever you're cooking (beans specifically) and the whole lid blows out (embedding itself in the ceiling and painting the kitchen). Scary stuff.

I have a Presto pressure cooker. The blowout disk is a piece of rubber/plastic built into the cap. I have a hard time imagining it getting contaminated in a way that surpasses the strength of the stainless steel lips on the lid/pot.

Its design hasn't changed since 1977.

It's not a tube that a bean could get into. Maybe you're thinking of the vent pipe? When that gets clogged, the overpressure plug is what blows away to release all the pressure.

Well...I have seen the whole lid, firmly embedded in the ceiling. IDK what the brand was, but apparently the failsafe (a rubbery thing, off-center of the lid) failed to fail-safe in that case.

Looking into RAPEX, I see multiple recalls for this specific issue (pressure buildup leading to uncontrolled blowout); all for brand names that are unfamiliar to me.

Of the 12 recalls I found on that database, none specifically mentioned failure of a blowout disc/plug1.

1 depended on the rubber gasket itself as the blowout valve. Several could be opened while under pressure, and some were deemed structurally deficient, which I take to mean that the vessel wasn't (consistently) built to withstand the design pressure (and a margin of safety).

[1] 1 did reach 290kpa (42psi), nearly triple the typical pressure cooker. Not sure what happened on that one.

  Next year we may see the first models without a 
  steering wheel or a brake pedal—there goes the 
  option of asking the driver (passenger?) to take 
This is bad news for everyone. This is the network effect exploitatively writ large, such that, surely you can prevent yourself from being inside such a thing, but as a lone individual, without laws in place to ban such a thing, you, your children, your family, your friends are all endangered by a blameless force that everyone can simply shrug at, and point to statistics claiming that there is even just an incremental improvement over ordinary human performance.

You can stop yourself from posting selfies on social media, under your real name. You cannot stop other people from taking pictures, which your acquaintances discover and tag with your real name.

This is kind of horrific. At least as horrific, or moreso than, current traffic statistics, because the moral hazard in play is abysmally worse.

There was an effective arms race that took shape with SUVs and road rage in the late 1990's. I think something similar will take shape, as self driving cars ramp up. It may surprise some, to find that a self driving incident won't be accepted as blameless, glitchy software errors. Owners may see themselves villified directly, for things a car they chose to own, had subsequently carried out.

I think there are three turns of consequence to a botched self driving car deployment.

One: some will choose harm the legal owner as an individual, lawfully or lawlessly.

Two: others will harm dealerships, mostly through sabotage.

Three: overt action against the manufacturers. At all levels, and not limited to ordinary civil disobedience.

These consequences are nothing to be sniffed at. Aviation shows us that spotty disasters don't result in civil unrest, but with a human in the loop survival was incentivized. Automotive deployments like this will be a fire and forget scenario, and the corporations loosing the reigns, have a demostrable history of neglect. I wonder if they anticipate, in the rush to market, just how severely the general public might react to finding their roads on the receiving end of software glitches that kill their firends and relatives like deer?

This is fascinating. For anyone interested in a slightly odd but unique and in-depth view of Systems design and failure, would like to recommend "The Systems Bible: The Beginner's Guide to Systems Large and Small" by John Gall.


I realise this is beside the point of the article, but gas distribution seems like an obsolete and dangerous technology to me.

For cooking, induction, for heating, heat pumps.

I built my house with solar, heat pumps, an EV charger... And natural gas service just for my stove and grill. It costs $500 for gas service, or $500 to put in the high amperage outlet for the stove, so it's a wash, cost-wise.

You can take my gas stove over my dead body. I'll happily pay whatever it takes for carbon neutral methane to come through my pipes. (Natural gas is mostly methane.)

The flame just makes everything taste better, and methane is pretty easy to produce without fossil fuels.

> so it's a wash, cost-wise

For installation. For ongoing costs, gas wins by a mile.

Nope. When you own your solar panels, they produce electricity at about the same cost per therm as gas. (After Federal tax incentives)

Meaning, running an electric oven / stove (edit: with appropriately-matched grid-tied solar panels) costs about the same as running a natural gas oven / stove.

The difference is that my loan never increases in cost, but the cost of natural gas is probably going to go up.

Anyway, I pay about $9-11 a month for natural gas to run my stove and grill. That's much less than the monthly premium to eat organic.

And in most areas, way more than a mile, especially with 97+% efficient appliances.

In an average house, there are no opportunities for an obviously winning 97% efficient gas burning appliance. A gas stove is nowhere near that efficient. A gas heater may be, but the competition is a heat pump that is far above 100% efficient (in the sense of heat delivered / energy consumed). Same goes for a water heater.

(Water heaters are odd. Where I live, there are no highly efficient gas-fired tank water heaters because no one makes one that meets the NOx rules. You can get highly efficient tankless heaters as well as indirect fired heaters. And you can get heat pump water heaters.)

You also need to take into account the efficiencies of delivery.

I know grid electricity is usually about 1/3rd efficient. (Rough numbers.)

Measured how? If you mean delivered energy / higher heating value of fuel, I’d believe it. But a good heat pump may have a COP greater than 3. And an increasing fraction of grid power comes from renewables.

(I’ve seen swimming pool heaters with heat pumps that quote a COP of 6 in mild weather. In appropriate climates, you can get a heat pump that pumps heat from your house to your pool, which can be extremely efficient.)

Induction stoves are unpleasant and ineffective to cook on, not to mention inefficient. And if you want to use a wok, well, you can't.

They are definitely better in appealing to first time buyers who may not know or care about their shortcomings. They look great and are easy to clean. But if you are into cooking, gas wins.

> Induction stoves are unpleasant and ineffective to cook on, not to mention inefficient. And if you want to use a wok, well, you can't.

With induction, ~90% of the energy from the electricity is used for cooking while only 40% of energy is used using a gas cooktop. Induction is more efficient, and you can use a wok if one is purchased that is designed for use with induction. Induction units can also regulate themselves based on feedback from the cooking device on the receiving end.

Induction is arguably superior to cooking with gas.



> On almost all counts, induction is faster, safer, cleaner, and more efficient than either gas or electric. And yes, we've done exhaustive oven testing in our labs to support that claim.

What about the 45% efficiency of a natural gas power plant?

Yes induction is efficient if you only look at the part inside your house. But most of the losses are outside your house.

If you live in a place with advanced wind or solar power like South Korea then electric probably is more efficient.

The electric grid tilts cleaner every year. Your stove will burn natural gas forever. A combined cycle natural gas generator is upwards of 50% efficient, already more efficient than your gas stove.

It's wasteful to spend money on gas infrastructure when it's clear electrical distribution is the future of home energy use. Infrastructure dollars are already in short supply. Just my two cents.

EDIT: You can stockpile energy with batteries, which is the likely outcome based on how much battery manufacturing capacity is coming online to build hundreds of thousands of EVs a year (which are also a great buffer for renewables and electricity in general).

It's wasteful to spend money on gas infrastructure when it's clear electrical distribution is the future of home energy use.

Maybe for cooking with a wok. But around here most homes are heated with natural gas. And that takes a lot more energy then heating chicken and vegetables.

New natural gas furnaces are up to 97% efficient. They're no longer allowed to sell furnaces that are less than 78% efficient.


You're only going to need a gas furnace (vs an efficient electric heat pump) in climates where it dips below 10F and an air or ground source heat pump would need to call on auxiliary heat to keep a dwelling warm (unless you live in a newer home with a very tight envelope and little unassisted air exchange occurs).

To your point, you can purchase very efficient gas furnaces, but they're more expensive and require a retrofit of the flue pipe to PVC due to corrosive properties of high efficiency furnace exhaust (not a concern in new construction).


Air source only heat with a heat pump is a slam dunk anywhere south of the 36°30′ parallel, and all of California.


We can't stockpile electricity for the winter. When days are short, demand for energy is much higher than what solar can generate.

We can stockpile natural gas, which is mostly methane. And we can figure out how to make methane without fossil fuels.

They are more difficult to cook with than gas - I find it hard to get a sense of how much heat is going into the pan. But they are certainly not ineffective. If anything, the problem is that they are too effective!

We love our induction range. It gets both way higher and way lower power than any gas range I've used. It doesn't dump a crapload of heat into the kitchen, which if you live in Hawaii is a big feature. And we power it with our PV...

> Induction stoves are unpleasant and ineffective to cook on, not to mention inefficient.

They're none of that, although they are less flexible than gas.

> They are definitely better in appealing to first time buyers who may not know or care about their shortcomings. They look great and are easy to clean.

They're safe, efficient and convenient for people who want to make food.

> And if you want to use a wok, well, you can't. […] But if you are into cooking, gas wins.

If you're one of the people who believes they need gas to survive, you can buy bottled gas for your range.

Compared to heating oil, which is the other viable option in the Northeast US, natural gas is cheaper, generally more efficient, and largely idiot-proof. With heating oil, a big, very common problem is people not scheduling deliveries appropriately, so the tank runs dry, and sucks up sludge that has settled to the bottom, and then you have to do expensive maintenance to clean the system, bleed the lines, etc. Worse, if you lose heat in a cold snap in February or March, you have to worry about water pipes freezing up and bursting, which cause even more damage. Piped in natural gas is always on.

I'd love to use electric heat. Too bad electricity costs in my state are stupidly high. It's just beyond insane to heat with electricity here. If my state could do away with the ineptitude, graft and taste for purchasing energy based on political optics that cause those high prices I'd consider electric heat but there's still the always-on issue that would give me pause. Gas infrastructure is much more "always" than electrical infrastructure around here which usually goes out when you need heat the most.

In my state electricity is relatively cheap (~$0.10 per kWh) but gas is more like $0.01 per kWh equivalent energy yield. Given that the US (with fracking) is basically the Saudi Arabia of gas, it makes sense to bet on gas for the long run. Even with very efficient air source heat pumps (~300% efficiency) it still makes sense to go with gas heat, especially in a place that has many heating degree days.

As to the "always on" issue, you can purchase a fairly cheap generator that runs on natural gas or propane for backup - assuming that the gas is running, you can run your furnace fan to heat the house. And yes my experience is that gas never goes out but electricity sometimes does.

Dangerous? How many people electrocute themselves each year?

At the moment this is also 3x as expensive to run, and the only way that gap will close is curtailment of gas exploration and fracking.

I find it bizarre that it’s not required by code for a pressure regulation station to have a sensor in that station.

I also wonder if the utiliy’s connection between the sensor and controller is on the internet.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact