When Power Grids Secede: What Happened in Texas

I didn’t really want to write about the electric grid in Texas, but figure it is necessary. I spent two years working for a utilities consulting firm–specializing in…

I didn’t really want to write about the electric grid in Texas, but figure it is necessary. I spent two years working for a utilities consulting firm–specializing in electric utilities–so I certainly understand what has happened.

Second, I’ve been exasperated by the coverage. Actually more by the liberal-left voices than the wingnuts. The wingnuts don’t have any of the facts right, but they are correct in one sense: the explanation is very simple.

(The socialists are correct, for once. This is very simple and the explanation is the one they apply to everything. Well, even a stopped clock is right twice a day.)

It’s a case study in three very common human traits: the industry tendency towards greed and denial, enabled by idiots with political ambitions.

The only thing difficult in any way is that I need to explain some terms that you almost certainly don’t know. They aren’t hard to understand– you’ve just never heard them.

+++

Let’s start by peeling back some disinformation. Here’s your first takeaway: Almost all the country experienced freezing temperatures. Many areas were blasted much worse than Texas.

Arkansas and Louisiana have been hit badly by weather. Residents in those areas still have power. Their water still works. They’re not dying in droves. Neither are people in Mississippi–or the part of Alabama in the blizzard’s path.

No, the problem is confined to Texas–a very specific area.

So that’s takeaway #2: The problem area seems to be defined by the boundaries of one of the grids.

The US electrical power grid is divided into three parts. It was originally split in half (for logistical reasons) by the Rocky Mountains– then the area that doesn’t have power decided to secede.

That’s takeaway #3. The rest is just background info.

1 THE EASTERN INTERCONNECTION is by far the largest (in both area and in amount of power) region. The EIC doesn’t stop at America’s borders– it reaches into part of Canada (excluding, naturally, Quebec). It also gets the northernmost portion of Texas.

The EIC was divided into eight regional councils (six US and two Canadian)– basically carved into fiefdoms dominated by one (or more) major utilities. It is maintained by 36 “load balancing authorities”– basically the largest power companies.

“Balancing” means exactly what the word implies. Power is rarely consumed in the same place it is generated. Usually a utility located in an area with access to cheap fuel (such as coal) will generate far more than its customers can use– then sell the excess to a utility located somewhere else.

Takeaway #4: The grid is to electric power what the highway system is to cars.

There are ALWAYS multiple methods to move power from point A to point B. That’s by design– to make sure one event in a single location can’t disrupt power to everyone else. As is true with roads, some transmission lines are better than others: they move power more quickly, can handle higher capacity, deliver with higher efficiency (when you move power, some of it dissipates) or are more reliable. But they all work.

The job of the balancing authority is to choose the best route to move the power, based on origin (where it was produced), destination (where it needs to go), volume (how much has to be moved) and conditions along the way. These choices ensure that none of the cables (or the transfers stations) ever get overloaded or underserved if something bad happens.

In the EIC, “something bad” usually means blizzards, tornadoes, hurricanes or floods. It can also mean equipment failures or human incompetence. (In 2003, Cleveland’s Firstenergy took down much of the EIC due to a series of cascading screwups).

The EIC handles the vast majority of North American power, moving between 650 gigawatts (in winter) and 750 (summer; summer is always higher, because of air conditioning).

2. THE WESTEN INTERCONNECTION serves the area from the Rockies to the Pacific– including parts of both Canada and Mexico.

Because this area is more sparsely populated, demand is MUCH lower: roughly 150 gigawatts in summer and 125 in winter.

You would think the smaller volume of power being moved would mean the WIC needs fewer balancing authorities. You would be mistaken. Remember, “authorities” are simply major utilities. Because demand is lower (less money to be made)– and the terrain is infinitely more treacherous (more wires to run and higher cost to maintain)– the spate of mergers that happened in the east never occurred out west.

The WIC, as a result, has 37 balancing authorities (34 in the US, two in Canada and one in Mexico). They deal with different “bad things”– besides blizzards, they get earthquakes, avalanches and florist friars. (Their version of Firstenergy is known as “Pacific Gas And Electric” and sometimes “SoCal Edison”.)

+++

I hope all that was educational. It would be helpful if more people understood how power highways worked. But it is not really necessary for a discussion of what went wrong in Texas– except to give you some perspective,

Because, at the risk of repeating myself, nothing went wrong in either the EIC or WIC. Despite the horrible weather that occurred in their service areas, nobody lost power for days.

Arkansas and Louisiana have been hit badly by weather. They aren’t having trouble. Neither is Mississippi– or the part of Alabama in the blizzard path.

The reason? They’re all served by the Eastern Interconnection. As is Oklahoma.

New Mexico and Arizona get power from the WIC. Residents have all kinds of issues– traffic stalled, fatal accidents– but nobody is freezing to death, because they don’t have power.

3. THE TEXAS INTERCONNECTIONis nothing like the other two ICs. It is run by ONE balancing authority: The Electric Reliability Council of Texas.

(Yes, you’re allowed to giggle at the name’s similarity to Disney’s EPCOT and the use of the word “Reliability” in this context.)

ERCOT serves only 60% of Texas’s land area. But it has most of the state’s population in its service area and roughly 90% of its power consumption.

ERCOT supplies about 65 GW (winter) and 75 (summer). That’s half the load of the WIC; 10% of the EIC. It’s a lot of power for one state– but not a lot of decision-making, compared to what transpires there.

Why does Texas have its own grid. Literally: They seceded from the US grid. ERCOT was created ENTIRELY for political reasons.

The United States regulates electric power with a simple carrot and stick: If a utility wants access to interstate power transmission (meaning that it can get power from a utility located in another state)– it MUST agree to comply with federal regulations.

Texas utilities didn’t want to be subject to federal power regulations. So Texas refused to join the grid.

In case of emergencies, they created an intermediary (or, if you prefer, a “fence”), They let North Texas (only 10% of total power consumption) connect to the EIC and bend the knee.

ERCOT connects to that “demilitarized zone” with two DC lines. Because both the origin and destination point are within Texas, this connection does not subject ERCOT to federal law. But it also means that ERCOT can’t get power from any of the states in the EIC.

(This has only been waived once: after Hurricane Ike hit Texas in September, 2008.)

ERCOT has no connection (direct or indirect) to the WIC; it does have two links to Mexican utilities (neither of which are connected to the US grid).

Takeaway #5: Secession meant that if anything happened in the ERCOT service area, the rest of the country couldn’t bail Texas out.

+++

Yes, this was extremely stupid; some people did recognize it. In 2009 (in the wake of Hurricane Ike), New Mexico announced the Tres Amigas SuperStation– whole goal was to connect the EIC, WIC and ERCOT using three 5-GW superconductor cables

It went nowhere, because neither Texas politicians nor eastern and western utility companies wanted to work together.

+++

The problem is worse than I’ve made it seem. Texas isn’t merely on its own logistically. It is also working without any checks and balances on its decision-making.

The other grids have A LOT of cooks stirring the pot. To repeat myself: the WIC has 37 balancing authorities moving 126-150 GW of power. The EIC has 34 balancing authorities for 650-750 GW.

Having the grid run by committee– with three dozen members– is a really bad idea in many respects. That structure is the main reason our grid never gets modernized. That many entities can never agree on anything; when a task force generates a report that one utility doesn’t like (it would cost them money), they refuse to cooperate and stall things.

But that unwieldy structure has one benefit. It ensures one idiot can’t destroy the grid.

If a utility starts doing something unsafe (or just plain dumb), the other authorities can decide (and have) to shun it. They can decide not to run power through its territory because they don’t trust its lines. They can choose not to buy power from it (because they don’t think it can fulfill the contract), or not to sell to it.

There is safety in numbers.

The Texas interconnection has only one balancing authority: ERCOT. If they make a mistake, everyone suffers.

And because ERCOT was created by politicians– and is run by for-profit businesses– it did.

+++

Before I get to what it did wrong, I’d want to explain how ERCOT works. Unfortunately, I can’t tell you as much as I’d like to. Since the blizzard hit, and the situation blew up, ERCOT has been pulling information offline.

ERCOT is run by a 15-member board of directors, plus eight alternates, who step in if someone is flying to Cancun. These members must come (this is defined by the statute that created ERCOT) from three different categories:

1) Six representatives from each electric utility segment: (a) investor-owned utilities, (b) municipally-owned utilities, (c) power cooperatives (usually for rural areas), (d) independent generators, (e) power marketers and (f) retail electric providers.

+++

I don’t want to do too much digressing, but the last two categories are intermediaries. They buy energy generated by someone and sell it at a profit to someone else.

Independent generators don’t have customers– they build plants and sell 100% of what they produce to someone else.

+++

2) Six representatives from the three consumer segments: industrial (factories and manufacturing), commercial (businesses that don’t make goods) and residential (people) customers.

3) Three independent members, who are supposed to be unaffiliated with the electric power industry.

ERCOT has a chair, vice chair and 13 board members. They pick a CEO, who is supposed to run things… but the board (to quote from the site) “sets overall goals and policy direction, has approval powers over the organization’s budget and market rules and oversees ERCOT operations.”

In theory, ERCOT’s decisions can be overruled. The first step is an appeal to the Public Utility Commission of Texas (PUC)– which can be overruled by the Texas Legislature. You can guess how often either happens. To make matters worse (for customers), the three commissioners of the PUC are appointed by the governor.

+++

I’ve watched entities like this work for decades. Here’s how everything plays out. First, the directors from the industry propose doing things that make their companies the most money. Then the directors representing the customers oppose it.

After some dueling proposals, two of the three “independent” members support the utilities. Easy-peasy, Japa-neezi, as my uncle used to say. All the companies need to do is make sure the “independent” members can be bent and they get their way.

Texas is a worst-case scenario. Let me quote from a story published in the Austin American-Statesman on the 16th, after things blew up and the media began to look into who was running ERCOT:

“Five [ERCOT] members, including the chairwoman and vice chairman, do not appear to live in Texas.”

“Sally Talberg, chairwoman of ERCOT’s board of directors, is a former state utility regulator who lives in Michigan, according to her biography on ERCOT’s website.”

[ERCOT naturally pulled down the list of members, as well as their bios after the story broke. Gregory Lee-Taitz might call that behavior “cognizance of guilt.”]

“Vice Chairman Peter Cramton is a professor of economics at the University of Cologne in Germany and at the University of Maryland. His Linkedin profile lists his location as Del Mar, Calif…

“Vanessa Anesetti-Parra, a board member, serves as vice president of regulatory and compliance at Just Energy and lists her location on Linkedin as Toronto.

“Board member Terry Bulger spent his career as a banking professional in the United States, Canada and in Europe. His biography on the ERCOT website states that he is a resident of Wheaton, Ill, a Chicago suburb.

“Raymond Hepper, another board member, retired in 2018 from ISO New England, which operates the electric system and wholesale markets for the six states that make up the area. His Linkedin profile lists his residency as Auburn, Maine.”

Note that the members who live out of state include all three of the supposedly “unaffiliated” members. So they have zero incentive to do anything for the customers.

+++

So what happened in Texas? As was true in 44 of the 48 continental US states, the weather in Texas got very cold. But in Texas— unlike almost all of those states— none of the facilities required to make and deliver electric power were prepared to handle that.

Let’s begin by explaining WHAT facilities Texas uses to make electric power— which will also let us dispel the myths right-wingers have been creating.

+ 45.5%— nearly half—was created by extracting natural gas from the ground and transmitting it through pipes to generation plants. The plants burn the gas to produce steam— which spins turbines, which generate electricity.

+ 22.8% was created by using windmills to spin turbines to generate electricity.

+ 17.9% was created by burning coal to generate electricity.

+ 10.9% was created by nukeyouler reactors (I believe that is how they are spelled in Texas).

The remaining 2.9% of the power was created by every other source— solar, fuel oil, power dams. (The largest share— about 2%— is solar.)

Before you try to share those figures, STOP. Those percentages reflect the amount of power each source actually used last year. They are not— repeat NOT— the percentage of total generation capacity each source can produce.

ERCOT’ estimates that, if every generator connected to its grid were running at maximum safe level, they would produce 86,000 Megawatts of power (86 gigawatts) per hour.

If that were happening, generators burning natural gas would produce an estimated 53%. But since Texas never needs to use 100% of its capacity, utilities can choose which generators they run. (They use the ones that are cheapest for them to run.)

+++

Here is a scary thought. ERCOT estimates that it has 86,200 MW of maximum supply. Peak demand was 74,820 MW, on August 12, 2019.

86 MW capacity and 75 MW of peak demand is 87% capacity, which is far too close for comfort. Remember what happened to the boiler in the book version of THE SHINING when it ran at 90% in the last chapter?

+++

To reiterate my takeaway. Wind turbines DO NOT produce over one-fifth of Texas’s total capacity— they produced 22.8% of the electricity that utilities chose to buy in 2020.

That number has been rising because it’s been getting windier in Texas (due to global you-know-what) and because wind turbines have been getting less and less expensive to run. In 2009 (the first year for which I have data), it cost Texas wind farms between $101-169 to produce one Megawatt of electricity for one hour. By 2018, that had dropped to $29-56.

2020 was the first year that sales of wind power represented a higher percentage than coal. For 2019, it was 24.8% coal and 18.6% wind. I’m gonna guess that continues— more and more wind farms are being built— but I’m not gonna project based on one year.

(Texas is also seeing a boom in solar. Cost per MwH has dropped from $323-$394 in 2009 to $40-46 in 2018. Like every other technology in history, the more we work at green energy, the more we learn and the less it costs.)

I need a takeaway to close this out, so here are two:

+ Texas relies on Natural Gas for 40-50% of its power.

+ Another 40% has been about equally split between coal and wind.

+++

So what happened in February 2021? We know what ERCOT thought would happen, because they are required to produce energy forecasts— how much power they think they will need to deliver and how much they expect companies can produce— every quarter.

(This is the only nice thing about having only one balancing authority. If this were the eastern or western interconnection, I’d have to look at the power forecasts for three dozen balancing authorities.)

ERCOT created its projections for the first quarter of 2021 by looking at the data from the same months between 2004 and 2018. After adjusting the earliest years to reflect economic growth (fewer people and businesses in 2004), it predicted it would need 57,699 MW at peak demand.

Because ERCOT expected it would have 82,513 MW of generation capacity available, it assumed they’d be running at 69.93% capacity this winter.

Unlike what Tucker Carlson and the rest of the right-wing idiots have been telling you, ERCOT wasn’t relying on wind and solar. They reduced estimates for wind and solar to an extreme degree— because they aren’t as stupid as the people on Faux News. They realized the sun wouldn’t shine as much and wind turbines don’t turn as quickly when it is cold.

(They also assumed other generators would be offline, because Texas utilities use winter— when customers aren’t using AC— to perform system maintenance.)

According to ERCOT’s final production estimates (dated November 5) for the period— which are (last I checked) still posted on the site, they assumed:

A. Only 43% of capacity from offshore or coastal wind farms.

B. Only 32% capacity inland.

C. Only 7% production from solar.

(You split estimates by location because the wind blows more on or near the ocean.)

All in all, ERCOT only expected they’d need 6 GW of wind, and less than one gigawatt of solar. They assumed natural gas, coal and nuclear would give them at least 67 GW.

As you might guess, this estimate was a smidge mistaken. Here’s what went wrong:

1) ERCOT WILDLY UNDERESTIMATED THE SEVERITY OF THE WINTER. Even their “worst possible outcome” scenario was way too low.

As I said above, ERCOT assumed peak demand would reach only 57 GW— that was based on what winter had been like over the last 15 years. If it were an unusually bad winter, ERCOT thought demand would be 62 GW. Based on the last three years, they thought 61 was more likely.

Their “perfect storm” projection—the absolute worst-case, where the most extreme situation occurred— was 67.2 GW.

By noon on February 14, demand was 70.3 GW. Twelve hours later— as temperatures continued to drop— ERCOT and the utilities realized they couldn’t meet demand. They simply shut off power to every zone that didn’t have a hospital located within it.

So we don’t actually know how high demand might have gotten— how badly they actually missed on their demand estimate. But that was nothing compared to theit supply miscalculation.

2) ERCOT WILDLY OVERESTIMATED HOW MUCH GENERATION CAPACITY IT ACTUALLY HAD. They assumed they could count on— no questions asked, fuhgeddabout it— 67 GW of generation available. If they broke out the gerbil treadmills due to emergency, ERCOT believed it could summon up 82 GW if they absolutely had to.

They were able to produce about 40 GW last week.

Wind and solar were off 16 GW from their summer peak— about 15% lower than ERCOT had expected winter output to be. Natural gas, coal and nukeyouler were down 30 GB— that’s 50% LESS than they imagined.

When actual results are off by 50%, you have lost the right to call your estimate a “projection”— or even “hope”. We’re in what the military calls “Scientific Wild-Assed Guess” or “Nate Silver Predicts 95% chance of Hillary” territory.

+++

Why did they fall short? Because none of the Texas utilities is run like a business— not one in a competitive market, at least.

1. Texas has the most wind capacity (did you expect it to be Rhode Island?), at 33,133 MW. None of the next nine states are having this degree of shortfall. (Those states are, in order: Iowa, 11,660; Oklahoma, 9.048; Kansas, 7,016; Illinois, 6,409; California, 5.922; Colorado, 4.692; Minnesota, 4,299; North Dakota, 3,989; Oregon 3,737.)

Those states aren’t (pardon the pun) sucking wind as badly for the same reason their airports don’t shut down as often in winter: they recognize the chance they might have bad weather, and install the myriad devices designed to keep turbines running at or near full capacity.

Canadian wind turbines— which EXPECT to encounter icing up to 20% of the time between October and March— rely on water-resistant coating (the new thing is carbon fiber), heating elements in the blades and insulation for the components (battery, motors, gearbox). (Some companies even offer drones designed to de-ice blades, though that sounds like a shuck.)

Turbines can operate with 50% efficiency or better in temperatures down to -22F. But you have to buy the equipment.

Texas utilities don’t. (No, I doubt that will change now.)

2. To oversimplify a lot natural gas can’t freeze— but the pipes that carry it can. If pipes freeze, water crystals can form inside them; the crystals (aka “hydrates”) can cling to the gas molecules and plug the pipes. Depending on the amount of pressure on the gas in the pipes, gas hydrates can form above water’s freezing point of water.

It’s not difficult to prevent this from happening. But you have to either bury the pipes or buy shielding— and that costs money.

Guess what percentage of natural gas lines in Texas are shielded. If your answer was in single digits, you’re close enough.

But wait— there’s more. The motors in the pumps that move the gas out of the wells and through the pipelines can freeze. Winterizing ALSO seemed like an unnecessary expense.

Best of all: Take a wild guess at the TYPE of energy these pumps and transmission facilities use. You guessed it— electric power from your friendly neighborhood utility.

Do most gas wells and pipelines have BACKUP generators to provide electric power so they can continue to pump— in case something happens to the power company? They do in SOME states

+++

Fun fact: 15 states produce 94% of the total natural gas America uses. Based on what just happened in Texas, I would say that 6.8 trillion cubic feet— 23.7% of our total production— is at risk. Also, natural gas wells in Texas have to drill unusually far down, meaning that they require stronger-than-normal pumps.

(The #2 state is Pennsylvania— 5.5 trillion cu.ft; 19% of total— and is, by and large, properly prepared. The #5 state— Ohio; 1.9 trillion; 6.2%— also knows it gets cold in winter.)

+++

Since natural gas is necessary for production of 50+% of total generation, electric generation facilities that run on natural gas COULD decide to have a huge reservoir of gas onsite— just in case the flow from the wells is disrupted (it isn’t gonna spoil).

But building a tank would cost money (and, to be fair, would also be dangerous— the last reel of WHITE HEAT comes to mind). So all too many utilities rely on “just-in-time” production— AKA just assuming there will always be a continuous flow of gas.

3. Both coal and nukeyouler power plants also require electricity to operate. (Coal-fired plants, for example, require electricity to run the belts that feed the fuel into the furnaces.) You might think they would divert some of the energy they produce to run the machines they need.

You would be wrong. If they do divert energy, they don’t divert enough of it.

4. Last but not least, the Texas generators mostly seem to have skimped on insulation, and not put up their storm windows. When the weather got cold, their turbines didn’t turn much more effectively than the windmills people mocked.

+++

So all of this happened. As I say, current reports say 40 GW of generation was online. That’s probably wrong, but accurate reports are probably a month away.

Then we get to the last element— which I mentioned earlier in passing. American power grids are not “smart” to any degree. Texas grids are especially stupid.

When the utilities couldn’t supply enough power to meet demand— when they realized they had to shut down— they couldn’t decide to deliver power only to “essential services” and “residential” areas. Their switches read “On” and “Off”— with a single flag for “Emergency Power” (the last areas that get power before a total system shutdown).

Basically (and I’m sure much of this will get corrected or supplemented by better reporting in June or July) — if you didn’t live near a hospital or a police station, you weren’t going to have power. There were lots of photos— you probably saw them, if you have dumb hippie friends on Fakebook— showing office buildings or luxury condos in Houston or Dallas with lights on and nothing else, with captions like “These are the customers that matter to the 1%.”

No, that’s just how the system works. Nobody allocated money to build a grid that let utilities say “these grids serve the most people and need power most.” The utilities had no intention of spending the money— it would come right out of profits— and the regulators certainly weren’t gonna make their BFFs pay for it.

Yes, that is a reflection of society’s priorities. No, I’m not excusing anyone. But… put it this way. My brother-in-law, who is the CEO of Firstenergy, has parents living in Suffield Township (population 6,383) just outside Mogadore (in Portage County, about 15 minutes from Akron).

Both Richard and Lori Jones are in their 80s; both have cancer. If a major event happened here, Mark COULDN’T issue an order to keep their lights on. He’d have to send one of his brothers or sisters to transport them to some location that had power (probably a hotel serving the families of patients at Summa Hospital or Akron General).

They’re in a low-density grid with no essential services. Plus, to keep power running to them (out there in BFE), he’d have to keep power running to not just their segment, but (I think) 3-4 segments connecting them. And (it’s been 20 years since I did this), to keep those lit, he’d have to enable god knows how many other ones— because you can’t keep some quads lit without also having the adjoining ones up.

Our electric grids aren’t wired the way homes that billionaires live in are— use your phone to order the the wine cellar, the artisanal cheese refrigerator and the AV booth for the multimedia theater powered, but cut off the rest.

Our country is maybe a step up from the Chance homestead in Natesville (RAISING HOPE) or the Douglas farm in Hooterville (GREEN ACRES).

Texas just showed what can happen anywhere, given a few shocks to the system.


Print Share Comment Cite Upload Translate
APA
Geoff Beckman | Just another WordPress site (2021-10-17T13:02:21+00:00) » When Power Grids Secede: What Happened in Texas. Retrieved from https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/.
MLA
" » When Power Grids Secede: What Happened in Texas." Geoff Beckman | Just another WordPress site - Friday February 26, 2021, https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/
HARVARD
Geoff Beckman | Just another WordPress site Friday February 26, 2021 » When Power Grids Secede: What Happened in Texas., viewed 2021-10-17T13:02:21+00:00,<https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/>
VANCOUVER
Geoff Beckman | Just another WordPress site - » When Power Grids Secede: What Happened in Texas. [Internet]. [Accessed 2021-10-17T13:02:21+00:00]. Available from: https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/
CHICAGO
" » When Power Grids Secede: What Happened in Texas." Geoff Beckman | Just another WordPress site - Accessed 2021-10-17T13:02:21+00:00. https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/
IEEE
" » When Power Grids Secede: What Happened in Texas." Geoff Beckman | Just another WordPress site [Online]. Available: https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/. [Accessed: 2021-10-17T13:02:21+00:00]
rf:citation
» When Power Grids Secede: What Happened in Texas | Geoff Beckman | Just another WordPress site | https://www.museumpirates.com/2021/02/26/when-power-grids-secede-what-happened-in-texas/ | 2021-10-17T13:02:21+00:00
https://github.com/addpipe/simple-recorderjs-demo