The Texas electric grid failure was a warm-up
texasmonthly.com
texasmonthly.com
> In the hours after the blackouts, ERCOT tried to shore up electricity reserves to stabilize the grid. The computer system that runs the market, though, interpreted this as an oversupply (in the middle of blackouts!) and dropped prices. When ERCOT and the PUC realized what was happening, officials decided to bypass the market and, on Monday evening, manually set prices at the maximum of $9,000 per megawatt hour. (By comparison, the average hourly price in 2020 was $25.73.)
So, ERCOT officials decided to basically input a fake number to short-circuit the broken market, and now everybody is on the hook to pay that price for real? Am I understanding this correctly? It sounds completely insane.
It is the only market not to have one. So you get to set the price when you do run. The price limit is 9999.99 which is the DB column size for price AFAIK.
https://cpowerenergymanagement.com/why-doesnt-texas-have-a-c...
Capacity markets are "the open market". They're a market for a different product - the option to request power during a later period. And when we're discussing power shortages with catastrophic results, that lack of capacity is a lot more important than specific resulting spot price.
Environmentally it's better to keep an extra coal plant operational than to have everyone install their own backup generators (and ad-hoc fuel storage!) to mitigate the problems with "best effort" grid supply.
Are you saying with confidence that was the motivation? And that still isn't much of how they make the decision including how long to override it.
What sort of societal good is being offered by making it so that people are on the hook for a much higher price than what is needed for guaranteeing supply? What sort of value is added? The point of all this stuff is so that people produce electricity, and others can purchase it at a reasonable price. It's not to find some hypothetical equilibrium using a model that assumes atomicity of actors and some sort of instant feedback loop that didn't actually exist during the blackouts.
If the state's objective is to maximize corporate profits at the expense of ratepayers, they have succeeded.
But if the objective is to reduce the frequency of catastrophes, the state might consider reducing the incentive to produce catastrophes.
welcome to republican lead texas that deregulated the power grid and decided to isolate the grid instead of playing well with others.
There needs to be evidence that the incentives point to scrimping on maintenance. Usually the incentives for capitalists is to preserve their capital and keep it is great repair. The is a strong, rational incentive to solve problems like winterisation because the revenue is higher than normal if someone can (and potential profits are much higher). Capitalism will just try to do it cheaply and to a standard that customers are happy with rather than in a politically showy fashion with gold plated "see I've thrown the kitchen sink at this!" style. But it will nevertheless do it if there is any money at all to be had.
[0] https://en.wikipedia.org/wiki/Waffle_House#Disaster_recovery
That is the opposite of the collective action problem. There are enormous rewards for an actor the less other people act. Which is one of the reasons bigger corporations do disaster planning, it lets them scoop up profits when prices are higher.
Capitalists don't give up easily if there is a person willing to spend money on something. They have flaws, but motivation to provide goods and services isn't one of them.
And likewise, markets are amoral. If the "market" opts for a failure every decade over a higher-priced day-to-day run... this isnt "rational" or "correct" or "ethical" or whatever. The way a market happens to function is amoral.
You can make the argument that over fairly short time horizons in non-critical areas "meeting consumer demand" in most cases has a certain consequentialist justification. This doesnt carry over, a-priori to all possible cases.
Ie., it is quite reasonable for a society via a democratic expression of ethical judgement to say "the market is not here considering the long-term impact in the right ethical manner"... ie., that being without power when it's -20 outside & affecting the whole population in ways not priced in by the market is a *moral failure*.
It is the prerogative of the collective to say at once "here's how it should be", rather than via iterated interactions "here's how its happened to be".
We're not talking a percent or two here. We've got people claiming that a company whose sole tactical purpose is to get energy to consumers failed sorta-on-purpose. That claim is more wild than some people in this thread realise. It amounts to giving up free money.
If they'd charged outrageous prices that would be one thing. But failure to supply power, by a company that makes money supplying power? This isn't an incentives problem, they were genuinely caught by surprise.
Arguing incompetence is one thing, but people are calling out incentives here without having good evidence of an incentives problem.
Until a yet another black swan event happens that requires the government to step in and help them with the eventuality they didn't plan for. Like COVID. If there are two companies competing and only one of them pays to be resilient against disaster, they won't be able to compete on price and will lose market share.
We're not in a comments section about a theorycrafted threat of that happening, or a comments section about twiddling incentives to encourage WH-like behaviour, we're in a comments section discussing the consequences of capitalist abandonment: namely, they made billions doing it.
You're trying to argue that this quote + greedy heartless capitalism -> give up and let the energy grid go dark.
That needs strong evidence. Occam's razor suggests they were genuinely caught off guard and they'll be better prepared next time. $9k/MWh is a very tempting price for someone who can have a generator set up on standby. Some people like making insane revenue. Possibly even some Texans. The price of electricity was basically infinity here.
What actually happened is insufficient for you? Let me know when sweeping changes are made to prepare the gri--
>$9k/MWh is a very tempting price for someone who can have a generator set up on standby
Oh, I see -- the solution is to let the market "solve" it. The physical-world issues that led to that tempting price will wreak their havoc on the physical-world infrastructure the next time that tempting price is set, but the kind of physical-world prophylactics that could have prevented that tempting price are "too expensive" to implement. Rather than prevent the catastrophe, instead ensure that someone is poised to swoop in and profit from stabilizing some preventable chaos.
>Some people like making insane revenue
If the grid didn't go down, why, no one would be able to make $9k/MWh!
Yeah. Duh. That is why I'm arguing that they were genuinely caught off guard and they'll be better prepared next time - because I don't think what happened is reflective of the incentives here.
> Rather than prevent the catastrophe, instead ensure that someone is poised to swoop in and profit from stabilizing some preventable chaos.
That doesn't make a lost of sense. If someone swoops in at the last minute and prevents the chaos it will still be prevented. And I'm not saying that the Texans are going to do that, just that it is one option. The existing operators might learn from this and harden their infrastructure.
As far as I can read this section, you seem to be hypothesising that there are two options:
1. Overengineer the average case.
2. Have a contingency plan.
And then asserting that 1 is the solution that is acceptable to you and that 2 would somehow be a bad outcome. I don't really understand your position if so - if having a contingency plan for emergencies is the preferred method of the market then that is cool. Good enough, probably cheap. Overengineering solutions is exactly the sort of thing that the free market is set up to avoid - it is wasteful. If it doesn't make economic sense to overengineer then it is much better to have a contingency plan - the average case is cheaper and there are plans and supplies on hand to deal with a crisis when one occurs.
No, I said that someone would "swoop in and *profit from stabilizing some preventable chaos*". You even fucking quoted me saying it! The comment is right above us! You even encouraged this outcome ("Some people like making insane revenue. Possibly even some Texans.")! That incredible misapprehension certainly doesn't make any sense to me, unless--
>I don't really understand your position if so
It's pretty clear you're doing ideological battle here. You're unwilling to face the recorded facts of the situation, let alone what your conversational interlocutor is actually saying, so you replace that with your own imagination. Indeed, you make an argument that you imagine me making, and then admit that you don't understand it -- isn't this definitionally a strawman? Why bother with this?
No matter your position on my position, it's pretty clear we're only ever going to talk past one another: your axiomatic assumption that the free market is the correct way to run a natural monopoly like a utility is IMO as wrong as your position on how the free market should have responded. It's only "overengineering the average case" if your instrumental goal is to "make money by providing power" -- if your instrumental goal is to "provide power, in exchange for money", the calculation changes entirely. Suddenly, it's not about winnowing another few percentage points out of a maintenence budget so some MBA can get a new car -- it's about securing the provision of a necessity of life, for a dollar or two added to the end-user's bill.
If you're still interesting in prosecuting our "discussion", I suggest you go answer this guy's question instead [0] -- if you do genuinely think that they were caught off guard, and that they'll improve the (proven to be fabulously profitable!!) situation in future, go explain it to him. It's not about what should happen or what might happen now, it's about what's already happened and what's just transpired.
If someone has stabilised the crisis then it isn't really a problem for the consumers. In this case, stabilising the crisis involves keeping the grid up and running in an emergency situation. There isn't any point sitting around thumbtwiddling watching a brownout happen, electricity grids often operate at speeds measured in either Hertz or 15 minute windows.
> isn't this definitionally a strawman? Why bother with this?
It'll help you explain yourself if you have some hints about how I am reading your comment. If I just say "explain further" to someone then it is unclear what they need to expand upon. Plus it makes my position clearer and I enjoy typing.
And the dude's comment isn't something I feel a need to answer. I'm not arguing in favour of the Texas electricity providers, I'm arguing that "they've got an incentive to give up and let the grid go dark" is a pretty extreme guess, is probably wrong and at a minimum needs real evidence to support it. Maximum profit probably isn't achieved if the grid goes down. It is much more likely that the incentives align to keep the grid running 24-7-365 with a high tolerance for outlier events and this outage was just run of the mill incompetence in the face of change.
Then explain the recurring failure of the Texas energy sector to do so.
You can see as whatever way you want, but Price is signal and it is better overall for the system to let Price to spike to certain insane levels. Should there be some kind of common-sense regulation for curtailing bad behavior perhaps.
If you need people or companies to build in redundancy - you have to pay the price or just run the efficient but fragile system and enjoy the black outs.
For example, at $500/MWh, I'll shut down my glassworks and pottery factory. At $5,000/MWh, I'll shut down the datacenter. At $50,000 I'll hook up a diesel generator from my shed. At $500,000 I'll personally get on a treadmill and start running, because the 100 watts I can generate that way gives me a better hourly wage than I can find elsewhere.
In fact, the reason for grid failure when prices aren't capped is entirely people on 'fixed price' schemes who do not modify their behaviour when faced with extreme prices.
The future is electric meters which have a knob on the front for "max price I'm willing to pay" and shut off if prices exceed that. If someone is happy to pay more, they can set their max price higher.
I think the core issue here (which is honestly my problem with surge pricing as well) is that there are limits in these systems. People in the cold need some heating, so no matter how high the price is there's going to be a floor of how much energy will be consumed. Peaker plants might exist, but there are only so many of those, so no matter how much demand there is supply can only vary by so much.
Given those factors, it feels like the reasonable thing to do is to set price floors and ceilings that are "reasonable". If there needs to be exceptional pricing, honestly this is where the State can step in (and it can buy insurance or whatever for this) to help. But $9000/mwh isn't helping to relieve problems (especially when people don't even know!).
Even in your cutoff knob, it's like... OK you are going to set a max price, and cut off your thing. Are you setting a max price for your fridge or your heating in winter? Or rather, would most people? Hell, what commercial enterprise is "shutting down the datacenter"?
If we are going to say that pricing is a signal for controlling consumption, you need a feedback loop. That implies communicating price changes and giving people time to adjust for that. Generally this has been solved through the easy-to-understand "at these times it's cheaper" and "at these times it's more expensive". You can also have fun with seasonal variations as well. It's not 1-to-1 but hell, just charge more or less right and seasonally it should all average out nicely, and it doesn't require nearly as much futzing around.
Market rate feedback could be done purely at the grid level, without exposing that to customers in real time, and mostly work. It removes arbitrage opportunities for energy providers, but that's a feature in my book.
Not really... If the price were too high, people could decide to move in with a friend for a bit, or live from their car.
This is far better than rolling blackouts, because each individual person can decide if they want to spend $$$$$'s on heating, or to make big quality of life sacrifices. In the rolling blackout case, there is the same amount of electricity distributed, yet users don't get the choice anymore.
The immediate ramp to the $9000 cap was designed to alleviate this. Yes, it's boneheaded, but I imagine any other manual override of the algorithm would have taken longer to get people to agree to.
The conclusion I keep coming to is that so many free market maximalists are absolutely terrible at designing markets.
There's wisdom in this statement but a better phrasing would replace "free market maximalists" to "humans."
Anyone who thinks they can "design a market" assumes all rational, predictable (aka not-human) actors.
This isn't a greedy capitalism problem, it's a govt manifested price hike and long term reduction in capacity, as part of the attempt to kill fossil fuels with no mature, viable, reliable, and affordable alternatives.
The fact checkers clouded this issue by claiming it "false", not because the $1500 floor price isn't true, but based on a manufactured straw-man argument that Abbot requested the DOE waiver instead of ERCOT. And a few other equally ridiculous (but expected) "fact checks".
They aren't allowed to make a judgment call that $7k is affordable but $8k is too costly, and just shut off power to their customers.
There were still govt effects limiting supply and mandating purchases, even at those prices.
And that's not an accident of policy; that's the intent, in artificially manipulating markets in support of agendas they can't outright mandate.
It never ceases to amaze me just how absurd capitalism can be. It's the best system we know of so far but it falls massively short of good enough in its current form.
What you've said follows perfect capitalist logic. And yet, what you've said incentivises those plant operators to ensure there's more outages. Nothing they can do - whether that be to make their operation leaner or make their generation more efficient or whatever else, nothing can compare to a 350x increase in pay per megawatt hour produced for $0 investment.
> best system we know of so far
It's hard to look at the trajectory of society and the world right now and expect the prosperity boom of capitalism to continue. A lot of people just don't see that boom at all. They're living objectively miserable lives under capitalism, and there's no capitalist incentive to change that.
Market based capitalism can handle the negative effects of climate change if producers are forced to pay for the negative impacts they produce.
>it's built-in wealth-concentrating tendencies.
Which economic system doesn't have this?
>A lot of people just don't see that boom at all. They're living objectively miserable lives under capitalism, and there's no capitalist incentive to change that.
>A lot of people just don't see that boom at all.
Who do you have in mind here?
Market-based capitalism could have handled climate change if we had perfect advance information about the infinite horizon adverse effects before it started, and had priced that into actions by pigovian taxes from the beginning of the industrial revolution.
We don't have that information now, and the people against whom such costs would be assessed have incentive to (and actively expend resources) preventing consensus based on the information we do have to prevent the information we do have from being applied that way.
I mean, there are many branches of socialism devoted quite strongly to this. You could call them a core tenet.
> Who do you have in mind here?
You have to have a serious failure of imagination to believe that everyone's just having a party under capitalism. To pick just one example out of thousands, have you heard of the homeless?
Your incredulity is painfully inauthentic and detracts from whatever argument you think you're making.
Ah yes, the intelligence of automated systems with limited visibility
Looks like someone forgot to add grid frequency as an input to the price engines
(Emissions limits are especially difficult in cold weather due to a less complete burn off and less atmospheric mixing. But this isn't to spinup new dirty power, so much as to run existing plants at full capacity without having to reduce output to more ideal levels for emissions purposes.)
To dispel a few other points that get repeated without much thought:
Texas' grid isolation didn't contribute. Surrounding grids including in Louisiana and Oklahoma were also undersupplied and operating under emergency conditions with requests for voluntary reduced consumption, and were also dangerously close to blackouts. Connections to grids without surplus power doesn't solve anything.
Wind power was frozen in and decreased over 97% over the previous 2 weeks as blades iced over or wind speed exceeded allowable limits. Solar was also iced over. Which led to a premature demand for gas peaker plants to offset this loss via a 450%+ increase in generation. So peakers were needed to offset intermittent renewables instead of as a buffer to meeting peak demand.
"Winterizing" NG plants, wells and lines is complex. What's simple and mostly ignored, is an overloaded NG system causes pressure drops, and pressure drops contribute disproportionately to condensate (both water and methane hydrates) freezing and reducing flow, which is a runaway condition. Pressure drops affects this much greater than temperature (even propane tanks in summer ice up when used at a high rate for an extended time). An increase in NG supply (more wells and pipes) would have greatly improved things without any "winterizing".
Coal and nuclear plants don't suffer from this fuel supply limitation since they have on-site fuel supplies, and the recent closure of 20%+ of the states coal capacity surely contributed, as do the remaining closures going back a decade+ of increased EPA regulation.
[0]https://www.energy.gov/sites/default/files/2021/02/f82/DOE%2...
It's sort of like asking why Northern or Western coastal cities don't have better hurricane protection or levees. It's not needed or worth the investment, even if New England does get an occasional hurricane. Despite federal flood protection mandates.
Asking what Southern and Gulf Coast cities do to prepare is a valid question, but doesn't justify taking the same precautions.
A better comparison is perhaps a windless summertime high pressure system causing rolling blackouts in California or Chicago. When was the last time you heard the spot price publicized during a summertime blackout?
Was it blamed on the equipment not being sufficiently designed for operation during an infrequent heat wave? Or on lack of planning? Or just blamed on the "freak" heat and possibly global warming?
The Texas event is a bit more complex due to underlying and transitioning infrastructure buildout where NG plays a much larger role, but its roughly the same as far as expectations go.
And the bigger takeaway should be that overdemand/undersupply of NG in cold weather is a runaway condition, but mainly due to pressure drops at the wells, not lack of insulation. Expected spikes don't cause pressure drops.
The alternative, non-market way would be to guarantee revenue even in times when the excess capacity isn't needed (most of the time), by paying more than the spot market price would normally provide. Personally, I prefer that way. But, if you're using a market system for electricity, you need an incentive to build excess capacity. Then, when a non-market mechanism (the order to impose blackouts) interferes with that, you need to make up for that somehow.
Again, I'm not saying I totally buy this, but there is the rationale.
The alternative market based way is to a have an additional "capacity market" for which providers offer bids for committing to provide future capacity. Thus providers who can guarantee they will provide a certain capacity are compensated for it, without needing to have a shortage of electricity to make the dynamic pay out. This is how it is done for the rest of the US.
Texas uniquely forewent the capacity market as a subsidy for wind, since wind providers are unable to supply guaranteed capacity and thus unable to bid in the capacity market. This creates a much more raw dynamic where grid capacity is fine right up until it is not.
If I'm not mistaken, customers have little idea that a price surge is in effect and what the rate is until they get billed. It's not like going to a gas station, seeing a price, and being able to opt in to the transaction, seek alternatives, or plan to self ration your consumption, as you're prepaying for the energy. The way the Texas model works it seems to assume that customers have a financial buffer that they're able to tap into to pay the higher rates. Everyone should plan for such a buffer but for those that don't have one it places hardship on them. It's also a risk for the provider that they won't get paid their windfall. You can't send huge surprise bills and expect people to pay them.
A sensible approach is to charge a bit extra under normal conditions to build a communal buffer that is used to pay for emergency power production. As long as there's good governance, this derisks the system. Another alternative might be energy bill insurance, if such a thing exists. It's easier to plan for a fixed monthly premium than a surprise payment in an emergency.
The consumers who had been hit by this were the ones who opted in to receive power at the market rate from a handful of shady businesses. They explicitly had to change their service to this model because the TikTok ads told them that the current market rate is $0.02/KWh so they are "overpaying" if they are buying at the utility's price.
The utility rate was not always fixed/negotiated: there's at least one utility company (Griddy) had an arrangement to pass-through market rate + small premium. Before the storm, this mostly meant getting rates modestly lower than the competition. The same customers were exposed to the power market spikes and received bills over $10,000. IIRC, Griddy has since filed for bankruptcy as the Texas legislature was generally supportive of individual customers not paying their bills. Paradoxically, Texas's political leadership allowed the utility companies to recover their losses by billing customers over years, which seems mostly performative/hiding expense from customers, but still decoupled from market forces.
I agree with your "sensible approach". On a side note, this issue is much like the reason conventionally taxis normally charged more than Uber did, but Uber had surge pricing. If you want enough capacity for the peak times, you need to overcharge at all other times.
$9,000 was a cap. If it wasn't for the cap, I'm sure prices would have gone up much higher under market rules. (ie: It's not like the market operators pushed prices up to that point, the operators capped it at that point.)
> (By comparison, the average hourly price in 2020 was $25.73
Commodities markets are notoriously volatile when there is mismatch between demand and supply. (There are also cases where power prices go negative, when there is more supply than demand. What's happening there is that generators are paying for the right to say online, on the premise that a shutdown is more costly. For large baseload units, this can absolutely be the case.)
That's exactly what the article says happened. They brought on extra power generation which dropped the market price, which they then overrode manually.
> there is compelling evidence [1] that such extreme weather phenomena are becoming more common [0].
The links looked interesting, so I followed them. The second one was an abstract [0] saying, basically: "the last 50 years have seen more of these types of events", fair enough. The first one [1], though, had a paragraph that stuck out to me (apologies for the long quote):
> “It’s tough, though,” Butler continues, “because we don’t have a very long record of observations of the stratosphere. We’ve only been observing it directly since the 1950s. That’s not very long to understand what kind of natural variability the polar vortex might be capable of. One researcher did a historical reconstruction by correlating the overlapping portions of the North Atlantic Oscillation index—which goes back much farther—and the polar vortex record, and then extrapolating the polar vortex record farther back in time using the NAO index. It showed no long-term trend, and no big differences in recent decades compared to previous decades.”
and then near the end, as the conclusion:
> Still, by most of the metrics experts use to describe winter climate, Overland agrees the big picture is clear: on average, winters are warmer and cold extremes are less likely than they were a century ago.
How is this article cited as evidence that extreme weather phenomena are becoming more common? How is the science.org article able to claim that there have been more in the last 50 years when we've only been observing them for ~70 years?
Anyone able to shed some light here?
[0] https://www.science.org/doi/10.1126/science.abi9167
[1] https://www.climate.gov/news-features/understanding-climate/...
In the central US, winter temperature is highly dependent on wind direction and strength. We've been measuring that for a very long time. The polar vortex is very distinct. Strong winds from the north cause a temperature drop of 30+ degrees basically overnight. It stays like that for a few days to a week, and then goes away. The day/night temperature differential is much smaller.
Anecdotally, people in the upper midwest will tell you this happens more frequently than it used to. Perhaps 3 out of every 5 years, maybe? Honestly, do we need to get more scientific than that? No, seriously - do we?
This is why it matters: here in the northern parts of the US, we can handle it pretty well. It's within the operating range of our infrastructure. We have modern clothing. We have modern forecasting. When it happens, it doesn't last that long, and we move on.
What doesn't seem to get talked about is this: if you consider distance from the arctic circle, the polar vortex that froze Texas only went a few percent further south than the ones now happen almost every year. A few percent. To my dear friends in Texas: just spend the money to winterize and then get on with your lives. It's not that much money - you can afford it.
Honestly, this just reminds me of the mindset differences between the chemical companies and gas companies in Texas. All those chemical plants that surround Houston are giant rats nests of pipes because the chemical companies won't waste a drop of anything they can sell. They pull in chemical engineers from all over the country to come find ways to turn a waste product into a revenue stream. The gas companies, however, have their heads so far up their rear ends that they'd rather flare off their waste from gas pumps than bring an outsider in to help make money off it. You practically need to have been born into a gas family to work in that industry. Whatever, what I really mean is that winterization has a positive ROI, but good luck getting a gas person to see that.
Yes. Anecdata isn't science. People don't remember even the recent past accurately. We need hard data. I can't believe I have to say this on HN.
It also discusses how the models break both ways: half the models predict stronger polar vortices, half less.
I'm firmly of the opinion that if the measurements and the models disagree, then the models are wrong. But that's probably me just being old-fashioned ;)
So if the anecdata and (half) the models say that these events are more common, but the actual measured data says there's no change, then surely we can agree that The Science says there's no change?
And ultimately, it's the surface-level data that impacts us day-to-day - which indicates that yes, the surface becomes cold AF more frequently than before.
Oh, and as an aside, recorded anecdotes are perfectly valid for use in science. Hence why we can even have science for psychology.
But what exactly would this entail? My home was fine in sub 30F temps. It is fine in sub 20F temps as well. It's coming up on 40 years old and has never had any significant improvements since it was construction. It was a cheaply built home then and still is today.
The issue is, the entirety of the energy system here failed in Austin. City of Austin deliberately turned off power when commanded to by ERCOT. Natural gas supplies were not always sufficient. When those are out, my house is going to freeze. Again, this doesn't really do anything but make me uncomfortable. I can just run a bucket of water then shut off the main to make sure the pipes don't burst.
So what would I do to "winterize"?
After the first day, there was no infrastructure to clear the roads of snow and ice. So you couldn't even go buy food, water, or even a blanket from the store in many cases. It didn't really matter if your vehicle was capable of driving in the snow (mine is fine) because there were so many accidents it was basically roadblocks everywhere. Also, when some of the stores lost power they told everyone they could not make any purchases.
So what would we as citizens do to prepare? I can keep food, water, and other supplies but that is basically it.
Vote in politicians that will force the winterization of the energy grid. That's the only way.
On a side note if I lived in Texas with all of that sun, I would be installing a solar roof and tesla power wall. I don't think this issue is going to be resolved for a long time.
They were not speaking to home owners as much as utility and energy infrastructure owners. There is a great deal that could be done to make the system overall more able to withstand winter.
Here are a couple concrete suggestions:
* Wind turbines are available with an optional 'cold weather package' that lets them run more reliably in the sort of weather that Texas experienced last February. Texas wind generators often don't pay for this package on the basis of perceived risk, but maybe that decision needs to be revisited.
* The gas pipeline companies that own electrical pumping stations could file the paperwork that notes these stations as critical infrastructure that should not be cut off from power. (A lot of this paperwork was missing in February.)
* Nat gas well heads could be retrofitted with equipment that keeps them freezing due to water production from the well.
* Long term planning could be put in place to connect ERCOT into neighboring power grids to allow for better risk sharing. (Southwest Power Pool in Arkansas is part of the Eastern Interconnect that covers the entire eastern half of the country. Even though SPP had had the same weather as TX, they were able to import power generation from as far away as the PA area to help support their load.)
* The South Texas Nuclear Project lost a generator due to a frozen pressure sensor line. Large coal plants went offline due to frozen coal piles. I'm sure there are remediations that could be put in place for that sort of thing too.
* The state could make an ernest effort to bring in political leadership that does not try to boil down all of the above to "Green New Deal Bad".
There really is no reason for the energy industry to bother winterizing. They don't really loose much by just idling capacity for a few days, especially if they don't have to worry about the cold weather preparations.
The government is not going to require any industry to do something it does not want to unless it is filled up with their buddies that can get rich in the process through government subsidized programs.
I'm sure this is going to be followed up with "but you could vote in politicians that cared!". There have been some definite attempts at voter disenfrachisement in recent history, although the most prominent one has failed. Texas is a representative democracy. It's representative of the districts, not the populace. The districts do not want politicians who are going to prioritize such things. The general population may in fact want it, but that isn't how our government is structured. So when you say "vote in other politicans", what you really mean is "replace the state government entirely". I doubt that is going to happen.
Only a problem because EPA no longer permits self sufficient gas turbine compressor stations.
>Nat gas well heads could be retrofitted with equipment that keeps them freezing due to water production from the well.
Pressure drop is a bigger factor than ambient temp and insulation. Increasing capacity by increasing # of wells and pipe size does more than defrosting equipment. Being overly reliant on NG to pick up the slack when renewables go offline causes avoidable pressure drops.
>Long term planning could be put in place to connect ERCOT into neighboring power grids.
The neighboring grids were also under emergency reductions and had no surplus power. OK and LA were extremely close to also implementing blackouts. LA was able to use its peaker plants and reduce industry demand to avoid this, in part due to a lower reliance on renewables (and not needing to waste peaker plant or transmission capacity to make up for wind turbines shut-in and solar panels covered in snow). A connected grid just would have further spread a cascading grid failure.
>Wind turbines are available with an optional 'cold weather package' that lets them run more reliably in the sort of weather that Texas experienced last February.
Wind turbines shutting off caused Texas to use peaker power prematurely to replace renewable power instead of in its proper role as a buffer for surges in demand. Once the front passed through, there was little to no wind for the remaining ~week of cold weather (high pressure system).
>Large coal plants went offline due to frozen coal piles.
The state just lost over 20% of its coal capacity due to increased EPA regulations. But I have not heard of any plant in Texas shutting down due to "frozen coal piles".
>The state could make an ernest effort to bring in political leadership that does not try to boil down all of the above to "Green New Deal Bad".
$1500/MWh DOE floor "market" prices to generate under conditions permissible 10 years ago is bad policy. Over reliance on intermittent renewables without sufficient backup is bad policy. Regulations reducing the number of NG wells is bad policy. Regulations reducing coal capacity by 20% without on-demand replacement capacity is bad policy.
None of those are state leadership. They're federal "green" friendly regulations and mandates.
Why didn't they have this excess capacity before the 20% reduction rule went into effect? When will they have it now that is has gone in to effect? I get the gist, although if the goal is to encourage green alternatives and planning, a policy that requires a reduction is simpler than one that requires reduction and a plan. I would rather have the latter, but if the alternative to the former is no green policy, then that's not a real option. So what could that policy have been to improve the situation?
Investors don't want to risk entering the "new market" in case the "old market" isn't actually shut down or run out of town and remains competitive. Uncertain indirect subsidies via govt driving up your competitors costs is less enticing than guaranteed subsidies for building more wind and solar.
The best solution is not allowing arbitraty administrative agencies to take the place of legislative certainty. But that's been deemed unacceptable by those demanding change at all costs despite lacking popular or legislative support.
Blaming wind for the problems seems a bit disingenuous, when wind outages were 3,000MW and thermal outages were over 30,000MW. I don't deny that wind failed too, but this was a holistic failure.
> The neighboring grids were also under emergency reductions and had no surplus power. OK and LA were extremely close to also implementing blackouts.
You missed a chance here to point out that SPP actually did have blackouts, although only for a few hours. They did better outside of Texas because they had more options and better preparedness.
> But I have not heard of any plant in Texas shutting down due to "frozen coal piles".
I heard about it at the time through connections in the industry, but it's gotten some press too.
https://acee.princeton.edu/acee-news/andlinger-center-speaks...
> Over reliance on intermittent renewables without sufficient backup i
In 1989, I sat in a cold dark house in Houston, due to power shortages and forced blackouts much like those in 2021. You're concerned about an overeliance on wind power in 2021? There was 'no' wind power in 1989. You're concerned about shut down coal plants in 2021? They were running in 1989. You're concerned about electrical NG pipeline pumps in 2021? They were self-powered in 1989.
Most of my suggestions are really about ensuring sufficient backup and addressing problems that range far beyond just the number of wind turbines and solar panels in the state. What are your suggestions?
Don't do this... please don't do this. Just set all your taps (internal and external) to drip slowly. That's all you need to do. Shutting off the main could actually make it worse because it creates an area without expansion.
> So what would I do to "winterize"?
Consider putting in more insulation in your attic, R30 should do it. It will help with energy bills in the summer too. Fixing walls is nice but not worth it unless you're doing a whole house reno.
Also next time you get your AC replaced, consider swapping it for a unit that also functions as a heat pump. Get one rated to 5-10F, that means it will be within rated efficiency range until that point. It will still function and operate below that but will be more like the electric heat you have now below that. But still generally more efficient overall. That reduces load, less load overall means less chance of things going out. It may seem small but those two things can have an effect.
The only thing that could freeze when I shut off my water main is the water main in the street. If that freezes, that isn't my problem. Nor is it anything I can do something about.
Unfortunately, problems like that are very hard (expensive & time consuming) to fix.
But my house was warm due to excellent insulation and running the fireplace for about 8 days straight.
Can you please elaborate on this? Why wouldn't shutting off the water supply to my house and then completely draining the pipes by running the lowest faucet in the house be the safest way to protect against burst pipes? Or am I misunderstanding your recommendation?
FYI for warm-weather dwellers: that is not adequate preparation to prevent pipes from freezing and bursting!
Required prep depends on insulation and interior temperature, but in the extreme you want to drain the supply pipes and put antifreeze in the drain traps (which can also burst, though less catastrophically than supply lines!).
An alternative that works in medium-cold temperatures is to leave all terminal taps dripping slowly.
A burst pipe will leak water when the pressure is restored. This can leak inside a wall cavity causing a great deal of damage in a short amount of time, including follow-on problems like mold and rot if not fully dried.
A frozen/burst pipe (even with pressure) will rarely leak while still frozen. The leak begins when it thaws, and yes that can be catastrophic if everyone is asleep (rare, due to nighttime temperature) or if no one is home to notice.
If the pressure is OFF when the burst pipe thaws, unless the system is drained in advance, you'll still have many gallons of water in the system that could leak, depending on location of burst. Again if you are not aware when the leak begins, you can drain a large quantity/most of the system contents.
Even a few minutes of leaking can require replacement of a lot of drywall, wallpaper, etc.
Why on earth would I turn my water back on after a freeze without also checking for leaks?
This may be non-intuitive if you haven't lived through it.
Why do you think a burst pipe will flow at or above the rate of a single toilet fill? Sometimes it might, but some frozen pipe bursts create very small breaches in the pipe wall. A tiny, steady stream inside a wall will barely if at all register on the spinner, and might not be audible at the valve. But it could cause a lot of damage.
Supply line in the basement, a run to the water heater(s), then forking into two parallel lines (hot and cold) to supply kitchen(s) and bathroom(s), verticals to get to higher floors, sometimes horizontal runs if the bathrooms are not stacked, etc. Cold supply to the outdoor spigots.
Even small single-story houses with perfectly located utilities will have several dozen feet of potable water lines. Add stories or less-convenient utility locations and you can be well into the hundreds.
But also, there are often low points in the piping that are not drained by the lowest faucet. And people frequently forget traps, tanks, and toilets.
If you know the freeze is coming and have bathtub(s). Fill them up ahead of time, you may need them for water for flushing.
Spend money to properly insulate the windows/doors. This helps for the awful TX summers too. Get an IR therm gun and find the leaky spots in the house and get them fixed. Insulate the garage and garage door. If the water heater is in a closet in the garage, insulate the closet.
Spare propane tank(s) for the grill if you have electric cooktops in the house.
Get a powerwall/generac like unit for the house that'll give you a longer runway if power is lost.
Have an electrician wire in a tap on your main to allow cutover from city to a plugin generator.
Keep sleeping bags on hand.
For clothing; mitts, wool socks/hats, down jacket, over jacket/pants or shell.
Keep a set of chains for your car, no one in TX has winter tires. (chains are illegal in TX, but I'd still use them in an emergency)
A couple water filtration units like a Brita. If you lose water and run out of bottled water you may need to drink the bathtub water.
Phase 1 at least includes such amazing winterization requirements as, "Attest that you repaired everything that broke last February" and "Inspect your insulation". There are more useful requirements like adding temperature monitoring, wind breaks around components, etc, but there is literally no enforcement mechanism yet so it's all basically on the honor-system for now.
Yes, the fix for cold temperatures are insulation requirements and wind breaks, etc. Not exactly exciting stuff but still correct.
Perhaps a tiny stick is more motivation than a much larger carrot for operators that were so unsophisticated as to miss that initial carrot, however.
If you notice the ERCOT press release, "321/324 facilities" passed inspection. Last February over 1,000 generating units failed, and over 600 of these were gas generators. So 320ish generators passed inspection -- it's unclear how many of those were in the 600 that failed last year or how many total there are in the state (e.g. the 320 that passed could have already been winterized?).
And the daily fines are for "repeat offenders" who don't "remedy within a reasonable period of time" deficiencies that are reported to the commission. It's basically toothless because "reasonable" isn't defined anywhere and it's all at the discretion of the PUC.
Obviously things will be directionally better now, and likely even more so in the future, I just wouldn't want to be in Texas during a cold streak in the next couple years.
Any fine system has time limits and reasonableness clauses.
I will continue living in Texas, and continue not being very concerned during the winter other than our lack of salt trucks.
His counterpart said nope—the Texas utility companies are very much treating it as a one-off event, and the winterization company is not seeing an increase in business there.
That is very slippery language and the rest of the Yahoo/Reuters article reads like a puff piece.
According to Texas Monthly's investigation, sounds like there's fuckery afoot:
> When Schwertner sent his bill to the House, the legislation also created a committee to map the gas-electric supply chain and determine which gas facilities were critical to the operation of power plants. It authorized the Railroad Commission to use its hundred new employees to inspect and, if necessary, fine gas companies. When the bill returned from the House, though, the language had been revised: only companies “prepared to operate during a weather emergency” were considered critical. This created a troubling loophole. Once the bill had passed, the Railroad Commission was responsible for implementing it, and the agency proposed a rule allowing gas companies to exempt themselves from winterizing simply by paying a $150 filing fee and claiming that a facility wasn’t prepared to stay operational—a dizzying bit of circular reasoning.
The “cold snap” terminology is also confusing to me. Is Texas expected to have one this year, and if so, at what point does a “cold snap” become part of normal climate?
(I am a Texas native and never experienced anything like last year’s “cold snap”. I was living in California last year, and now I’m worried about another “cold snap” this year)
I expect hard freezes and occasional ice during winters in Texas but consecutive days of single digit highs I do not.
And it wasn't just the cold that was the problem -- it was the cold over 5 days and unprecedented amount of snow. It is rare for Austin to ever see snow.
Here in Maine we experience the weather Texas had basically bi-weekly in normal winters (it's been very mild the past few years) and IF your power goes out, it's back on in 4 hours, 20 if it's a particularly bad storm and you are a low priority area. Meanwhile the wind turbines up north will happily keep spinning, and at no point has generation failed.
88% of gas producers by volume are winterized.
Why would they:
> During the February freeze, the gas industry failed to deliver critically needed fuel, and while Texans of all stripes suffered, the gas industry scored windfall profits of about $11 billion—creating debts that residents and businesses will pay for at least the next decade.
I understand that it's a challenge to maintain 60-hertz frequency throughout the grid-restart process. But I'd like to know why exactly it'd take weeks instead of hours to restart the grid.
So you need to do this slow 'titration' of turning on power assets, connecting demand to them, balancing that load and then moving on to the next thing. Texas has something like 11 designated 'black start' generators that are meant to be bulletproof but more than half of them failed in some way last February.
I'm not sure how you would test your automation even if you tried.
"It’s a large stepwise process to build up load, build up generation, build up more load, build up more generation until they’ve got enough reliability to go to the next element of the system..."
There are added challenges in a distributed system. Black-starting a hydroelectric plant is easy but I doubt if Texas has any.
[1] https://www.wired.com/story/venezuela-power-outage-black-sta...
Black starting a plant is one thing which can basically be solved by having a working and large enough diesel generator on site to power the plant auxiliary systems such as oil and water pumps required for start up.
Reenergizing all of the loads in a large system from zero is another thing. Energizing transmission lines with no load are essentially giant capacitors that initially appear as a dead short, energizing transformers that are essentially giant inductors that initially appear as a dead short, energizing resistive loads requires all of the sudden opening whatever the throttle on the power plant is - water flow, steam flow which itself requires gas/coal/feed water/etc in to a boiler, or natural gas, that is as rough on equipment as flooring the engine in your car except you have to make sure the power plants connected are as large as the loads being connected. And the inertia in the rotating masses in the system has to be enough to supply the energy to the load until those throttles open and generate the additional energy - time delays here are measured in seconds. How do the fish in the river or the people camping on the river bank like it when the river goes from 5% to 100% all of the sudden? More restrictions and considerations there.
The whole process is a bunch of step responses to systems at unusual operating points. Operators of many facilities that normally just run at constant load points won’t be familiar with the unusual demands placed on the equipment. Equipment will be operating both at minimum and maximum limits in very short time spans. It’s a recipe for equipment failures unless it is practiced regularly.
Solar and wind unless supplanted by batteries are no help here either.
Solar systems are on inverters, so capable of very fast response - unless the controller decides it doesn't like what it's seeing and trips out again. A bad inverter trip was the cause of this UK outage: https://www.greentechmedia.com/articles/read/orsted-and-rwe-... (missing a software update from Siemens!)
Grid batteries would definitely help a lot here as they're similarly capable of fast response, and providing artificial "inertia" to the system.
Inverters don’t like to energize lines though since they normally follow the system waveform. The inverters would have to be connected with a low load set point and frequency regulation eg droop enabled and then they certainly could help with the system response.
When a single generator synchronizes to a normally operating wide area grid, the relative power balance is so huge that if the generator is not close enough to matching the phasors it will get pulled into sync as long as it's close enough to not trip offline again to avoid damage. With only a few generators on line, and very narrow extra capacity and loads, the margin of error is much smaller. This must be done methodically and carefully for every generator in the black start plan, and it must be done with the majority of the region not having power so transportation, communication, possibly fuel services for the generators, and basic services like food and water and toilets are also impacted. Everything is going to be slow.
Some plants also have ramp up and startup cycles that can take most of a day to produce close to their nameplate capacity, and most of them require a significant external power source to run. Coal plants require lots of moving machines that load and process and pulverize the coal and complex steam pressure control systems that don't go from 0 to 100 at a whim; Nuclear plant cooldowns after a trip can be days, and startup again may require another day or two as well. Modern combined cycle gas plants are relatively easy by comparison but still require several hundred KW of available station servicing external power to start the unit and usually take a few hours of careful checks to go from cold to ready to sync.
It would probably only take a few days to get most of the ERCOT region back on line, but a full restoration would definitely take weeks as much of the work can't even begin until the big generators in key locations are all up and stable, and most of the efforts can't realistically proceed until the cause of the blackout has been resolved, e.g. no pumps for the gas lines, those generators can't start anyway.
Imagine a Y junction switch that can take switch between the power from 2 plants into a local city grid. The switch's control circuits might only having a power supply on the side of one plant. (Dumb cost saving measure, which I heard is falling out of practice.)
I don't understand why the max price is so exorbitantly high -- if they capped it at, say, 10X the 30 day rolling average, then they'd still incentivize everyone with energy to sell to bring it online, but without bankrupting utilities and customers who have to pay over 300 times the normal cost of energy.
A family that normally pays $20 for a day of electricity in cold weather can probably manage to pay $200/day for a couple days, but probably not $7000/day.
Then there is the atypical amount of capacity you need for when something goes sideways. There is no wind for a week or it's cloudy and hot so there is less solar even though everybody wants to run their air conditioner.
The capacity to handle that rarely gets used, but you still want to have it. So the price you have to pay to the person who can provide it is really, really high, but (and because) it's only paid for a short period of time. Or nobody will bother to build that capacity and when that situation happens there won't be enough power at any price.
There's a lot of economic theory pointing to the economic efficiency of marginal cost pricing. But the wholesale electricity market is a pretty "synthetic" design, and combined with the technical limitation demand must match supply exactly every second, marginal cost pricing leads to huge volatility.
I'm not really sure what the solution would be, though. I'm not aware of any workable alternative (competitive) market model. There's of course ye olde regulated utility model, but, yeah, maybe we don't want to go back to that.
The cheapest insurance provider should then be a company that can provide emergency capacity to the grid, because the time they have to pay out claims is the same time they're making bank by selling power for >$100/kWh. The money from the premiums goes to providing the emergency capacity and then when the event happens the claims and the revenues cancel out.
Another alternative is to give customers a price threshold at which they don't want electricity. If the price exceeds $2/kWh, just immediately shut off my service. For someone with their own generator this could be a good deal (they get to pay a lower rate by not buying the price insurance), and then it drops demand out of the grid when there isn't enough supply.
Sure, that's one way, ultimately all the way to the end user paying a flat fee, which I guess is what a majority of residential customers are in fact doing.
But I wonder, if all the producers and consumers are hedged in one way or other against too low/too high prices, why have that wholesale price in the first place if it's just a fantasy number that affects nobody?
> Another alternative is to give customers a price threshold at which they don't want electricity. If the price exceeds $2/kWh, just immediately shut off my service.
This is in essence what RTO's are doing today, just for the whole region they are serving and not each customer separately. That is, there is a roof price, and instead of the price going over that limit in order to incentivize more generation to come online or customers to reduce demand, there will be rolling blackouts.
So in general wrt marginal cost pricing, I was thinking of something along the lines of https://www.researchgate.net/publication/323378820_The_Use_o...
This is only going to get worse as more and more generation will be near-zero marginal cost like renewables and nuclear.
It wouldn't create the hypothetical incentive to invest in capacity only used in black swan disasters, but I bet there's ~nobody doing that anyway.
As horrible as $7,000/day electricity is, it would be worse if the whole grid was down for a month to recover from catastrophic failure.
ERCOT has authority to load-shed, regardless of price. From the article:
> For the entirety of 2020, Texans paid $9.8 billion to keep the juice flowing. On February 16 alone, they spent roughly $10.3 billion. Costs for the month of February totaled more than $50 billion.
The price reaction time is generally too slow, and judging by the bills incurred, ineffective as a control mechanism - hence the massive outrage on incurred costs. Do the costs seem appropriate?
While I haven't read the whole article, blackouts mentioned were the effect of load shedding and the principal action that prevented grid failure. The price was fixed only after the critical period, by an unknown criteria and questionable need. Yes, you don't want price to drop too low after a load-shed, but do you need it maxed and do you need it for days?
Look, the simple answer to your question is no, of course. But, given that we understand the reason why controllers acted the way they did, these criticisms are taking advantage of hindsight, and hindsight is 20/20. The controllers needed to not just shed load, but to keep demand down until the storm passed. They weren't thinking about price/demand elasticity and how much Grandma's electric bill was likely to be. And furthermore, it's not correct to expect every public operator to second-guess every move they make, because it slows down the public sector, increases public sector bloat, increases cover-your-ass behavior, etc.
If you're going to apply hindsight to arrive at a moral judgement, then it's better to apply that hindsight not at the people who were time-stressed to prevent complete grid collapse, but to the gas industry that refused cold-weatherization for short-term profit.
I'm not unsympathetic to the people who got screwed with five figure electric bills. If a state employee screwed up on the pricing, the state should pay the bills. But I wouldn't throw the controller under the bus for it.
> the simple answer to your question is no
The answer is no and no. IMO, if they did only one of those things, it wouldn't be a big deal. When you set a fixed price at 100-200x normal price and hold it like that for days, you'd better have a very, very good reason and think through the consequences.
What you're saying does make sense, but I wouldn't give them a free pass, unless they can explain their reasoning. I don't know - maybe they did. If you're doing things with such implications, you'd better be able to explain your choices and reasoning. Not just to cover you ass, but to really consider the implications and possible alternatives. And we're not talking about a few critical minutes or hours, they kept the price for days.
I think it is an odd choice to blame the consumer in this scenario since customers just one state over did just fine throughout this ordeal.
Edit: Perhaps in the end consumers didn't have to pay. I haven't found anything saying they didn't.
There are different systems in different municipalities. Houston and Dallas both have systems where multiple providers can offer service which are distinct from the providers responsible for the actual utility infrastructure (power lines to your house). Austin and San Antonio both have municipally owned utilities (CPS Energy in San Antonio, Austin Energy in Austin). For the majority of households they either have a contracted rate based on their agreement when they signed up for service (just like a cell phone contract), or they have a legally set rate which requires a ballot issue.
This is WELL established. There's no dispute here, and if you can't "tell who's telling the truth", it just means you haven't bothered doing any research before deciding. The national media used the snowstorm in February as a political tool to kick Texas while it was down because they see it as a Republican bastion (even though the majority of the population lives in reliably blue cities in national elections and Texas is purple now), and used something truthful for less than 1/10th of 1% of the population to act like it was that way for everyone to make the issue seem larger than it was.
Also from your article "Despite the astronomical prices, Scott-Amos is considered one of the lucky ones, as millions of people across Texas are still without power while temperatures remain perilously low." Griddy customers pay the live wholesale price, they were also immune from forced load-shedding meaning they didn't go without power during freezing weather like most other Texans did, so true to their contract they paid the live wholesale price and in return received guaranteed service. Sounds like they got exactly what they signed up for at the risk they decided to take which 99.999% of Texans opted not to take for the safer (but generally more expensive) financial bet.
In regards to "The national media used the snowstorm in February as a political tool to kick Texas while it was down because they see it as a Republican bastion" I agree. However, it didn't help that this problem was spun around by the opposite side and used to blame solar and wind energy sources as the reason for the problem.
From what I gather(ed) the root problem was the lack of winterization to [all] energy infrastructure. But perhaps I'm misinformed on this too. Do you have additional insight?
Thanks.
This is mostly the case. The primary issue wasn't that generation infrastructure wasn't weatherized, it's that wind turbines were not properly weatherized and failed, and the NG peaker infrastructure /was/ weatherized and took over, but was having issues meeting demands because the NG well infrastructure was not weatherized, meaning the peaker plants couldn't draw enough gas to meet demands.
Another major confounding factor during the snowstorm was that grid didn't have enough high voltage pathways to support the amount of power that had to be shifted between generation locations and the generation locations were too spread out regionally (Texas is a large and mostly empty state) to maintain grid stability, so there was forced load shedding to prevent phase shifting out of spec or grid islanding.
Also, while we're at it, it's bullshit when people say that "Republicans deregulated power in Texas, which is why it has an independent grid." Texas has ALWAYS had an independent grid. When the national grid started forming in response to New Deal legislation in 1935 states could avoid being affected by federal regulations by not transmitting power across state lines, so Texas chose to not connect its existing grid to the national grid as it was forming. Originally Texas had two independent grids, but they joined during that time, choosing to remain disconnected from the national grid but sharing with each other. ERCOT was actually formed to begin internal regulations of power in Texas in response to a huge national power disaster in 1965 that didn't affect Texas, but made the state realize it needed /some/ regulation.
I don't make any judgements as to whether this situation is better or worse, and it was certainly politically motivated, but it's not new and it's not recent, so most of the things written about it are simply untrue.
Sure, this is true. That was never in dispute though, so I didn't address it. I don't think I have a "narrative". I'm just tired of every single thread about what happened last February in Texas turning into an opportunity for everybody to get their political kicks in on the basis of lies and twists. That behavior does exactly nothing to solve any of the problems identified, and it stokes feelings of undeserved self-righteousness.
The power shortages could have been avoided had wind generation been properly weatherized and had we had properly weatherized gas wells to supply the peaker plants. Neither of those things were true, and therefore this situation occurred. So far, we seem to have improved generation weatherization since, but the wells are still not weatherized, so it's possible this situation (or something similar) could occur again. It remains to be seen.
I do think the market failure here is a bit political, in that it's market fundamentalism to rely solely on spot prices, falling back to wishful thinking that short periods of extremely high prices will somehow incentivize new investment into the market. That decision to not have a capacity market is definitely a political one.
Agreed on the physical things that need to happen to prevent this from reoccurring, but the question is how to get those things implemented. A strong regulated market would just declare that they need to be done, and ignore the protests of how burdensome they are. Whereas a deregulated market dynamic really needs a forward capacity market or some other incentive for plants to do the preparation for staying online. And this applies to both the electricity and natural gas markets.
The bigger issue is that ERCOT /has/ made a regulation saying generators need to weatherize, which they have done so and been inspected and verified. But ERCOT can't control the natural gas producers, and they have thus far mostly refused to weatherize their wellheads, and therefore we may once again find that we can't drive peaker generation high enough to meet demand due to inadequate natural gas supply pressure. The oil & gas industry and electrical utilities are regulated independently in Texas, although they are often considered to be the same as "the energy sector", and so there is definitely more that can/should be done here.
I guess we'll see how it pans out. With better weatherized renewables, it's likely we won't see this situation re-occur. There was plenty of wind blowing during the crisis, so had the wind turbines been properly weatherized we wouldn't have needed as much peaker capacity anyway.
My utility bill didn't go up at all during last February, nor did the bills of the vast vast majority of Texans. The only individuals who had an increase to pay were customers of a company called Griddy that charges wholesale live pricing to customers, which is generally a discount, but it's customers taking a risk that a price spike may occur. Those customers entered into that agreement with a full understanding they were paying the live wholesale rate at all times for power and consequently Griddy had a very small number of customers compared to other utility offerings that use contractual set rates.
It's a tiny fraction of truth blown out of proportion in the national media that Texan individuals had to pay $9k/kWh. That simply didn't happen for nearly anyone.
[0]: Lest you think I'm being sensational, here's the formula: ([1]/[2] = ~0.001, or .1%).
[1]: https://www.houstonchronicle.com/business/energy/article/Gri...
Taken at face value: I work in the tech industry, have never had any association to the energy sector. I believe in basic personal accountability and the bounds of reality, which unfortunately makes me a rare individual in the current era.
From the article:
> The Legislature approved the issuance of what will likely end up being about $5 to $6 billion in bonds to pay back some of these costs. That form of borrowing creates an obligation of about $200 for every adult and child in Texas.
No, most families didn't pay $9000/MWh during the event. But they will pay for it in the form of future taxes or fees charged by the utility.
What's your alternative proposal? A hindsight 20/20 review would have obviously said to not have such a crisis happen in the first place through proper preparation. That didn't happen, and here we are. How do you pay for it?
A key piece is that due to load-shedding, the actual cost to most Texans is significantly lower than had they "kept the lights on" at $9k. Instead, we all lost power and the price was paid only to maintain the necessary generation to recover the grid without a black start that would have been a much larger disaster. Electrical power is common infrastructure, and the cost to sustain it is a common cost, raising that in such a way seems the most reasonable path forward. I'm not happy about my rates going up or my taxes increasing, but I also understanding reality. The net effect on me as a rate payer is still far far far less than had I agreed to pay live wholesale like the tiny fraction of people on Griddy.
This is going easy on the people directly responsible.
> “We’re going to get to the bottom of this and find out what the hell happened, and we’re going to fix it.”
There's no "going to the bottom of this". It was known what would happen a decade before it did, and most of the people responsible were in office for nearly that time. They were warned. They chose to do nothing.
A failure cannot become more complete, yet people still place trust in these individuals?
Between 2015 and 2019, Texas didn't even make it into the top 10 list of power outages; and they're in #10 spot for outage hours per outage. In a state with extreme temperatures and large, unpopulated areas, this strikes me as extremely good.
I have no doubt that their power grid isn't perfect. No power grid is perfect. I just like basing my opinions on the facts and, currently, Texas isn't doing badly.
Also, realize that all outages aren't the same. Thousands of homes scattered throughout Maine losing their power because an ice storm brings down power lines isn't that big a deal. Houses there have oil or gas heating, and town centers have more robust electric lines so people can wait it out there and still get supplies.
In Texas, the issue was the the whole grid went down and for two weeks. Hospitals, fire stations, grocery stores, ... were affected. There food and water shortages, and 250-700 people died because of it.
Most likely your backup heat is wood and/or a petrol generator. But you definitely have a backup plan to stay warm, as it being cold for months at a time means the question of outages isn't "if" but rather "when".
It turns out that power grids don't really care much about your political philosophy. They still require competent engineering and maintenance and planning and accountability. And that requires smart people doing their jobs with adequate resources, which isn't guaranteed whether you've voted for the deepest blue or the brightest red government you can get.
1. That's not "data"
2. What you need to make a judgment on is whether Texas is positioned well to handle a tail event, not how it handles the odds and ends of unexceptional power outages that dominate graphs like that. The families of the hundreds who died last winter probably have some opinions about how well Texas handled the last event, handling which was extremely bad and closer than anyone would have liked to power grid "Mad Max." It's a question of how much they've improved.
Why weeks? I “survived” the “great” northeast blackout and most areas were back up in 16-48 hours. And that involved a bigger network across multiple jurisdictions.
In a real "black start" none of that external capability and coordination exists. Imagine the entire US and Canada east of roughly Kansas being dark.
For weeks? Without ATMs, gas pumps, grocery stores, etc? It all runs on electricity. It would get pretty grim. According to risk assessments of EMP risk, 90% of Americans of die after 1 year of not having electricity.
So far nothing quite as bad as a real black start has yet happened, and here's hoping it never will.
solar flares are predictable with enough notice. If it’s an h bomb, we will have enough problems with or without the empl. So, please stop writing old fashioned fairy tales here.
> But the situation was only growing more dire. At the precise time of the third call, the frequency reached a critical threshold: 59.4 hertz.
> Automated turbines across the state began spinning even faster to produce more electricity, but when the frequency dips below 59.4 hertz, the turbines reach speeds and pressures that can cause catastrophic damage to them, requiring that they be repaired or replaced.
Is this actually a way to bump up power generation capacity - by getting the turbines to "spin faster" rather than spinning up more turbines? Also, if the turbines are spinning faster, why does the frequency of the generated AC drop?
I think frequency dropped because there was too much load on the grid. More load produces more impedance in the grid, this means that, without additional power on the turbines, they respond by turning slower. Like having to bike up hill.
To give the benefit of the doubt to the author, it is also true that a turbine shaft can be connected to a generator via a gearbox, in which configuration a faster turbine would correspond to more power delivered to the generator at the same generator RPM.
They were "dialing up" the turbines to compensate for overload but it was insufficient and causing the turbines to redline/exceed operating allowances.
The article might be explaining it wrong.
They want the turbines to spin faster because they're generating at 59.4 Hz when they should be at 60. Keeping at 60 is a balancing act. When there is more load, the turbines want to slow down and to maintain speed you need more pressure.
You can have a 50 MW turbine generate 25 MW of electricity. It still spins at the same speed, but because there is only 25 MW of load, it needs less steam pressure to maintain speed. When the load increases, they increase the pressure and it generates more power while spinning at the same speed.
But if you take a bunch of generation capacity out of the grid, you now have a 50 MW turbine with 65 MW of load. Putting enough steam pressure to generate 65 MW of power into a 50 MW turbine is out of spec and could damage it, but put less than that in and the frequency falls.
In general it's like this:
* Generators in the whole grid need to be synced up (frequency identical and phase difference small).
* If the generators in one area cannot keep up with the load in that area the frequency drops a bit (phase is shifted back).
* Powerlines transmit power from areas with more generation capacity (where phase is ahead) to areas lacking capacity (phase in the lacking area is pushed ahead and frequencies remain identical).
* Powerlines have limited capacity.
* If the transmitted power is not enough to satisfy demand in that area the generators slow down more (phase gets further behind).
* If one area gets out of phase too much it would lose frequency completely and there would be catastrophic failure in powerlines, transformers, generators, switches etc..
* To prevent catastrophic failure all areas agree on a small band of frequency variation inside which the regulation mechanisms work.
* If the frequency falls below that the grid disconnects into islands.
* The islands will either go dark completely (and power plants in those areas shut off) or they will have an oversupply of electricity (with these power plants reducing production).
* Regions with an oversupply go back to nominal frequency, sync up and reconnect. Then bit by bit the other areas are connected.
The AC power grid is mostly a historical accident; it's only been recently that we gained the ability to fully control DC power at grid scale.
In my view, electricity shouldn't count as a commodity. Virtually any other commodity in the world has some amount of buffer in the supply chain. Oil can be stored in tankers, grain in silos, etc. Electricity doesn't work like that. Production and consumption must always be in perfect balance. Sure, it's possible to address that by implementing grid scale batteries, but very few electricity grids in the world have done so. Australia is an exception, and the Telsa batteries are able to blunt a lot of the price gouging caused during periods of increased demand.
I can't see Texas investing in batteries to create a diversity of supply given how hydrocarbon-centric the state is, especially if they can't even mandate the easy/cheap fixes to their existing infrastructure.
Markets are great when they work, but governance matters when they fail.
What does this mean? I thought all electric heaters were essentially 100% efficient, due to conservation of energy.
Well, gas heaters are a lot more than 100% efficient in terms of heating per unit of electricity consumed, so better for the grid.
(Before anyone else points it out, yes, electric heat pumps are usually more than 100% efficient, too, but that's not relevant here because in severe cold, they are less efficient than direct heating.)
If this was about northern Canada or other such climates you would have a good point.
Mitsubishi's residential air-source ductless mini-split heat pumps that have their "Hyper-heating inverter" technology are greater than 100% down to at least -13℉ (-25℃), although their capacity starts dropping fairly linearly from 100% at 23℉ (-5℃) to 76% at -13℉ (-25℃), so unless you oversized your system a bit you might have to put up with your house being a few degrees cooler when it gets very cold outside for a long time.
Their efficiency remains above 180% down to around -5℉ (-20℃).
In most of the major cities of Texas these would have worked fine and remained around 200% or higher efficiency and around 90% capacity. Only a small number of people would have had them get down to 100% efficiency.
Fujitsu and Daikon have similarly performing heat pumps.
They look like large baseboard heaters, and inside have a bunch of ceramic bricks which can store a lot energy. During off peak periods the heaters can 'charge', i.e. heat up the bricks, then after that the heat is slowly released over time. Modern versions have a small fan to assist releasing heat, controlled by a thermostat or other smarts.
The thing is though, these can store a lot of energy. My parents had a new one installed last year which stores 24 kWh of heat:
https://www.elnur.co.uk/product/ecombi-hhr-high-heat-retenti...
Potentially these (the charging can be controlled by a smart meter) could have reduced the electricity demand during peak times.
What is striking to me, is that roughly 50% of Texas have gas furnaces and most of the rest have electric resistance heating. Nearly everyone has an AC. There was more demand for electricity from those using electric heaters during the cold snap than in the summer from AC's when the state has an average temperature near a hundred. It would probably be worth it to subsidize a heat pump for anyone stuck with resistance heating.
For example, geothermal electric heat pumps typically achieve greater than 300% efficiency in terms of thermal energy delivered per electric energy consumed.
https://blog.networkprofile.org
which details steps taken by the author to mitigate any future outages like what they experienced in the past. really thorough and interesting to think through what grid independence takes, plus any related costs/maintenance.