A Texas power official says they were "seconds and minutes" away from long-term catastrophe:
In an interview with the Texas Tribune, Bill Magness, the president of the Electric Reliability Council of Texas, said that if the utility did not cut power on Monday, the amount of energy going offline due to the storm, combined with a surge in demand amid the intense cold, could have caused widespread blackouts lasting for months, leaving the state in an “indeterminately long” crisis.
....In that disaster scenario, demand for power would have overwhelmed the supply of energy on the grid, which could potentially cause power stations to blow and equipment to catch fire. Once physical infrastructure takes such severe damage, it can take months to repair and would demand a slow process to return power sources back to the grid.
Wait. If demand for power is too high, power plants can blow up? Sort of like a bad Star Trek computer when you ask it a difficult question?
I know nothing about power generation, but seriously? You can make a power plant blow up simply by demanding too much electricity? I would sure like to hear an explanation for this. Is it common to all power plants and grids? Or just those in Texas?
ERCOT has a last resort option: ordering transmission companies to reduce demand on the system with rotating outages for customers....Usually, those outages are limited to less than 45 minutes. But this week, the outages lasted days. That’s likely because after ERCOT ordered companies to stop providing power to customers, even more power generation tripped offline, and it was not able to “roll” the outages effectively, Johnson explained.
So if the power deficit is too big, they can't even roll their rolling blackouts, as they usually do? They just have to choose some unlucky schmoes and cut their power for days?
It sure sounds like Texas power utilities are badly fucked up. They apparently have almost no ability to fail gracefully, something that every engineer in the world counts as a high priority. But I guess it would have been too expensive, and who expects bad weather in Texas, anyway?
It's almost as if regulations are promulgated to protect the common good when the need becomes apparent and, without those regulations, bad things can happen. Nah, that can't be it. It was the windmills.
Think of it as a short circuit. If there are too many consumers plugging in and too few generators on line the remaining generators will be be providing current beyond what they are designed to. The result will be overheated equipment and yes, oil filled transformers exploding. Normally there are circuit breakers to prevent this sort of thing, but this is Texas where people replace blown fuses in their pickups with live 22 rounds.
I was thinking about the power outage that darkened the Northeast a while back. A line went down in Ohio, I think, and that led to cascading failures. I don't recall power plants themselves blowing up, as Kevin suggested, but they did disconnected themselves from the grid and went into emergency shut down. The coal and nuclear plants take a while to restart.
here's a write up:
https://www.theatlantic.com/photo/2018/08/photos-15-years-since-the-2003-northeast-blackout/567410/
"but this is Texas where people replace blown fuses in their pickups with live 22 rounds."
Absolutely true. I saw it happen in a house in West Texas, not the town.
They got what they voted for, again and again.
What was that Mencken definition of democracy?
I don’t see what the big deal is. If the electricity goes out all you need to do is fly to Cancun and stay at the Ritz Carleton. Problem solved.
Will Cancun be Ted Cruz's Macaca Moment, or will the Storm rally to Sen. Qruz & his Qanqun Summit to plan March 4, 2021, when Donald Trump's 2nd Inaugural happens?
I'm not an expert on this, but yes, grid instability can cause damage to equipment on it. It's all interconnected. I think the linked story is probably equal parts sensationalism and reality. Sensationalism because this is probably true in any case of large scale grid instability, reality for the same reason. The problem is that power plants have to go offline when the grid goes unstable to protect equipment, which makes the demand/generation ratio worse, which makes the grid more unstable, etc.
Texas is probably only worse because plants aren't winterized in the right way (this is probably reasonable in many respects) and they are much more isolated then the rest of the domestic grid. Yes, that makes events potentially large in the rest of the country, but you can also get support from neighboring states. Texas can't do that.
Very good detailed report on the 2003 blackout. Ch. 6 is particularlyy good as it documents GW of power flowing in around the NE in seconds.
https://www.energy.gov/sites/prod/files/oeprod/DocumentsandMedia/BlackoutFinal-Web.pdf
When stations blow, it means a transformer, right? Let's see what happens when a station blows: https://bityl.co/5g5f
Now, what happens if one substation's blown transformers cascades into one substation after another blowing?
Yeah, power used has to be in balance with the power generated. Local power companies are tied into interconnections that handle the balancing task using generators and transmission lines among several companies, which helps the process. Usually they do it well, but extreme conditions can screw it all up. Like J. Frank Parnell says, the system's parts have breakers that will trip in order to shut off flow and protect the equipment, which can blow up, melt, etc, if it gets overloaded. The idea is that it's less destructive to shut sections down and restart later than it is to have to replace blown-out system parts and rebuild the whole thing.
Think what would happen in an old house if you replaced the fuses with pennies and had a bunch of octopus outlets hooked up and drawing too much current for the circuit. Not uncommon before breakers became almost universal, and it caused a lot of house fires.
The thing with TX is that most of the state isn't tied into larger regional interconnections-- the interconnection system is entirely in-state (ERCOT). Frigid weather brought a huge demand surge, but it also froze pipes and wellheads that take gas out of the ground to the generators, and also froze pipes and valves and controls in the generating stations (including at least one nuclear plant that's exposed to the elements, I've read). So they couldn't generate nearly enough power to match usage and couldn't get power from outside, and the systems shut down to protect the equipment.
Of course doing that saved electrical infrastructure at the expense of human life and health, and of destruction by broken plumbing and fires of an awful lot of homes and other structures. Either outcome would take months or years to recover from and now I wouldn't know which of the two is worse.
This is completely man-made, an avoidable catastrophe that responsible parties in TX had been warned against for many years. Deep freezes aren't that rare there.
A little nit-picking. Failure to spend the money to winterize more of the generation/grid was putting profit over people. Rolling blackouts were not putting infrastructure over people since the alternative would have been everyone being in the dark.
Basically I wouldn't disagree-- if the winterizing failures weren't so complete, and the freeze so widespread, that the system was too compromised to maintain rolling blackouts, then we wouldn't be seeing the burst pipes and flooding, potable water shutoffs, CO poisonings, house fires, etc, that we're seeing, or at least not to anywhere near this extent.
But that is where we are now. And I'd agree with you and with Kevin's next post that the individual operators put profits ahead of service, reliability, and people they serve, and regulators let them get away with it or encouraged it.
It was a gamble that whatever bad stuff happened would be containable. But this freeze has been too widespread and too long-lasting for that. Maybe no amount of winterization would have been enough for this kind of perfect storm-- which I doubt because plenty of systems work fine under worse conditions-- but we'll never know. What we do know is that when operators and regulators lost the gamble, so did the people, structures, and functions these operators and regulators are supposed to serve.
Haven't read the next post, but my understanding is that Texas doesn't want to be on a regional grid specifically because they don't want any damn regulators telling them how to run their state. They've also had some dangerous cold spells in recent years, investigated the failures, wrote reports, and didn't do any winterizing. So now they reap the whirlwind.
Several suppliers who aren’t part of ERCOT winterized their plants after a cold snap in 2011. So, second time in a decade. Probably more frequent going forward, thanks to climate-change effects on winds aloft.
It's very bizarre. How could they not segment their grid to be able to provide power when and where possible? How do they restart this?
They disconnected from the rest of the country to be free of regulations. Turns out regulations are mostly useful.
The rolling blackouts, which reduce total demand, are supposed to kick in before the individual power plants reach their limits. But however well a system is designed, it needs to be tested to see that it actually works. This is not something that can be done easily for power grids. Colder places experience stresses on the system more often and presumably have greater knowledge of how to do these things and are also more willing to spend money on the required system.
First, power plants shouldn't "blow up". They just automatically disconnect from the network, and that's been true for decades.
There are three networks in the lower 48 states, Eastern Interconnection, Western Interconnection, and Texas Interconnection. All the networks produce power at the same frequency, 60 Hz, and voltage, 240 VAC, 3-wire split phase, and within each network all the produced power is in-phase. The in-phase requirement is what makes it easy to "share" power when needed. It is also what makes it practically impossible for Texas to get power from other grids. The power from another grid would have to be converted to DC then back to 240 VAC and in the phase used by the Texas network. Alternately, variable frequency transformers could be used. There are two DC "ties" to the Eastern Interconnection, but they can't power the whole Texas network.
Note that El Paso, in West Texas, is not part of the Texas Network and had no problems during the recent storms.
Right, also, El Paso Electric winterized equipment after a cold snap in 2011, as did Entergy, which serves the Beaumont area in East Texas.
https://www.khou.com/article/news/investigations/texas-power-outage-counties-with-fewer-power-outages-are-not-part-of-state-grid/285-67277280-7fce-42ca-b55b-f8fc639c0889
No, you can’t blow up a power plant by overloading its output. You can blow up transformers, and replacing large ones can take a while. Which is why they should be protected.
It sounds like the Texas plants tripped either on frozen fuel or water lines, or tripped on overload. Trips are designed to avoid equipment damage, but then the plant needs power for its ‘parasitic’ loads to restart. Usually that is supplied by the grid. Oops.
AC power is delivered in sinewaves. The sinewaves have a certain voltage and frequency. All of the power plants on the grid phase-lock to the 60Hz sinewave. This means each plant dramatically varies its voltage 60 times per second. All the plants have to be in exact sync. If all of the sudden the frequency drops just a little due to an unstable grid, a plant will then try to drive the grid to a higher voltage when the rest of the grid wants a lower voltage. This plant will then self-protect and trip off. If it fails to trip off, thermal overloads or other failures can occur to various pieces of equipment.
But also, if the voltage drops overall due to a very high load on the grid, and a power plant is asked to deliver more power than it was designed to supply, it can trip off and self-protect too. If it fails to trip off, various equipment can start to overheat and be damaged.
If a grid is overloaded and could go unstable, grid operators start shutting down parts of the grid to protect equipment. Otherwise a cascading blackout can occur, and a much bigger area can be affected. In the 2003 NE black out, 265 power plants shut down during the outage.
These problems are universal to power generation and grids and not just a Texas problem. The Texas problem was due to inadequate winterization of equipment. After this fiasco, surely that will be rectified.
Well, that may have been the proximal cause, but the lack of an interconnect to either the east or west interconnections made things much worse than they otherwise would have been.
Live by the disconnected grid, die by the disconnected grid.
Then Texas falls under the federal regulatory regime. I really don't know the impact of that. But I really doubt the Republicans who run the state would agree to that.
Another factor is, even though Texas produces a lot of gas, it doesn’t have a lot of storage for the natural gas. Most of Texas is very dependent on real-time production of gas. And the gas production infrastructure, like the electric power infrastructure, has been hobbled by freezing.
BTW, below are retail electricity prices in the five largest states (cents/kWh). How much of this is really due to the unique policies of Texas, I can't say.
California 16.89
Texas 8.50
New York 14.34
Florida 10.44
Pennsylvania 9.81
Those must be average prices, or spot prices in normal conditions? From what I've seen, the low overall TX price is traceable to the mostly unregulated market there, which is a double-edged sword. They get cheap power in normal times, but it's from a system that doesn't invest in reserve capacity, in winterizing its equipment, or, as you say, in significant fuel storage.
To extrapolate, it's basically a just-in-time operation almost all the way from the ground to your home outlet, with real-time pricing, and with no provision for wide-scale contingencies. A test at real scale of the j-i-t manufacturing models, or an offering to the gods of the market (forget the prudential story about the seven lean years or any of that biblical stuff) or to the primacy of the profit motive.
And in order to preserve its freedom to operate that way, it's a system that's only minimally connected to the rest of the country. So with this particular wide-scale contingency, it looks like they've developed a system that made a difficult situation into their perfect storm.
Average retail price for the year.
https://www.eia.gov/electricity/state/
The average price is low in Texas. There are other states with lower average retail prices.
"All the plants have to be in exact sync. If all of the sudden the frequency drops just a little due to an unstable grid, a plant will then try to drive the grid to a higher voltage when the rest of the grid wants a lower voltage. "
I believe you meant to say 'phase'.
Why did Texas choose not to connect to the Eastern or Western interties?
We have a Governor (Abbott) that refused to follow the recommendations of a grid study after the last power disaster some 10 years ago, and we have a former governor that was head of the Federal Energy Regulatory Comm'n (Rick Perry) who, as one of his signature activities, killed the study and effort to build a "supergrid" that could deal with this.
See: Who Killed the Supergrid, The Atlantic, August 2020.
AOC has done more for Texas this week than any Republican ever has.
It all boils down to risk management. Nobody expects utilities to make their grids impervious to weather. To start with, it would mean putting all the transmission lines underground, which would be ruinously expensive (and bring its own sources of potential engineering failures). We accept that in a bad thunderstorm we'll probably lose power for a few hours. After a cyclone, it might be down for a few days.
The Texas situation was especially bad because of the deadly consequences of being without energy in a deep freeze. But the core issue remains the same: how to balance risks against costs to the consumer. I'm not sure in this instance it's helpful to blame evil corporate greed; publicly-owned utilities have to make the same calculations and have been known to get it even more horrendously wrong because politicians prioritise lower tariffs over security of supply.
But the outages in Texas were not due to downed transmission lines; they were due to frozen fuel and water lines at generating stations. Insulating and heating those is not ruinously expensive, it’s done all over the northern US. Wind turbines in northern states (and Canada!) are winterized and generate power year-round.
and who expects bad weather in Texas, anyway?
It's only by odd circumstance that it was a winter cold snap that caused this crisis. In summer, with demand for air conditioning, it seems like this could be an annual event.
In summer the system's generating capacity shouldn't be taking the enormous hit it took now because of widespread freezing. The basic problem this time has been power generation, badly compromised for all the reasons talked about upthread and all, or almost all, related to cold. Summer could well see brownouts, but the difference between generation and usage shouldn't be nearly as great and thus not nearly as likely to bring on this kind of catastrophe.
TX utilities were advised after a big freeze in 2003 that they needed to protect their equipment from cold weather. It seems that the ERCOT firms, or most of them, didn't do it. As mentioned upthread, El Paso's utility did; it's also tied into the western US interconnect. That part of the state did fine.
2011, not 2003-- got it mixed up with the Northeast blackout.
" Wait. If demand for power is too high, power plants can blow up? "
Kevin, remember it's Gell-Mann amnesia all the way down...
You've just pointed out the stupidity of journalist analysis as to "responsibility" in this particular disaster. But now you're going to turn around and accept that these same journalists are secretly EE's who are competent to make and judge claims about the TX power grid?
Pingback: Here’s Why Texas Power Stations Had to Be Shut Down – Kevin Drum