The Texas Power Grid Is Hanging on by Its Fingernails | Hacker News Reader

The Texas Power Grid Is Hanging on by Its Fingernails

gizmodo.com

48 pointsrbanffy52 comments

If any other state’s power grid got even remotely the level of scrutiny TX did, it would feel like it’s broken also.

As a former Washingtonian who moved to TX let me tell you all the ways WA was bad:

- most power lines and grid infrastructure in major cities like Seattle is 50+ years old and fails - constant power outages in storms - we had a massive dam on the Columbia develop a crack, and most of the other dams could do with reinvestment - Hanford and everything with it is a disaster - our “green grid” of dams is likely responsible for collapsing the native salmon population and with it, the resident orca populations and a bunch of other critically endangered species - power generation in WA is a crazy mishmash of public utility agencies all of which have more in common with the Adeptus Mechanicum from 40k than a normal government agency

I think if you went into most any states bureaucracies with a fine toothed comb it could feel apocalyptic. Truth is, WA’s system is pretty fine (dam’s aside) and the same seems to hold up for TX. It’s very rare for any system to survive contact with a generational event unscathed.

I can go on, but I mention this because the article is short on anything concrete, confuses (as another commenter pointed out) how power generation and regulation works, and it feels particularly bitter when TX is pretty ahead on renewables compared to say, NYC or Chicago, where most of these writers hail from.

disclosure: i worked in mechanical maintenance for a large midwestern power distribution company.

power isnt a 'good' or 'bad.' co-ops and collectives exist to keep generators and distributors and maintenance crews honest and on the level but at the end of the day cities say "we need power" and power companies are responsible for it. theyre also responsible for communicating risks like salmon collapse and, if thats acceptable to the city, then proceed.

texas has a see-no-evil, hear-no-evil problem where plants do not perform risk assessment and do not communicate it, instead pocketing the cash and giving a big beaming thumbs-up. Texas has decided that instead of a transparent and working regulatory agency that helps navigate things like storm coverage, MTTR, and low income assistance, a rubber stamp will do.

no grids perfect. cities often put off maintenance for decades to save traffic headaches, prevent water and other utility outages, and to plan around other construction work they might be doing. Texas is a different and very cavalier beast entirely though.

Excellent summary of the Texas situation.

The reason people are shitting on on Texas is that Texas shit talks other states power grids.

I've lost power here in Maryland several times over the course of just a few years (usually during thunderstorms). And I happen to know that Pepco (our power company) basically has a "let hardware fail before we fix it" policy. A couple years ago we had a problem where plugging in a vacuum or turning on the disposal dimmed the lights in the whole house. We hired a master electrician to take a look and he concluded it was pepco's box outside. We reported it to pepco and they did nothing. 1 year later the box failed and we completely lost power until pepco came out and replaced something in their box.

So yeah, I seriously doubt many states could hold up to this level of scrutiny.

Michigan's Consumers Energy has had 14% of it's customers blacked out since Tuesday morning. 500,000 people across the state and it looks like another two days to fix it.

California’s grid management definitely has issues. I’m not very familiar with this grid management stuff, but I’ve read all CAISO’s reports on what happened in august 2020 that led to the rolling blackouts, and there’s some stuff in there that seemed pretty crazy.

The highlight for me is that CAISO’s day-ahead energy market had some features and “enhancements” such that it 1) ignored the constraint to make physically feasible schedules with enough supply to meet demand, and 2) developed schedules based on “virtual supply” rather than actual physical supply, and 3) develops the schedule based a bid-in demand figure that is consistently well below the actual demand. The result of this is that CAISO’s system generated mostly-binding schedules in which CA exported sort of a lot of power to low priority areas at the same time as CA was having rolling blackouts. I guess financial markets are generally pretty good for solving this sort of problem, but it feels like some optimization based on the real constraints would be useful.

http://www.caiso.com/Documents/Final-Root-Cause-Analysis-Mid..., the whole thing is interesting, the issues I mentioned are mostly described in sections 4.3 and B.3.

My understanding of this may not be correct, so refer to that report as the source of truth, and please let me know if my description is not accurate.

I was in Seattle for about 7 years, and then about 15 miles west of Seattle for 20+ years. In Seattle I only had maybe two or three outages, only one of which laster more than an hour.

15 miles west of Seattle, there are maybe 2 or 3 a year almost always during high winds or heavy (for this area) snow. Those are usually fixed within an hour or two, but every few years there is a storm big enough to cause so many outages that it might take several hours for mine to get fixed.

I did have a 24 hour outage once when a tree took down a line on my street.

The important thing though is that none of these outages covered the whole region. They were all collections of local outages spread around the region, with plenty of places spread around the region where power stayed on.

In every case I could get in my car and drive a few minutes and find someplace where power is fine. The only time where it looked like that might be possible was the one with the tree down on my street, because the tree blocked the street. But there was a neighbor with a half loop driveway with the ends of the loop on different sides of the blockage letting people further down the street bypass the tree that way.

Most importantly, I've always been able to easily find a gas station somewhere that still has power. If I fill my tank and go back home I can stay in the car until power comes back on. Idling in the driveway with the heater running sufficiently to handle the coldest day, listing to the radio or using my phone charged via the car I've got over 24 hours of run time before I need to go get more gas.

I don't know how well that would work in summer since I've never checked the fuel consumption rate idling with the air condition on. Most multiple outages here are in Winter (storms) or Spring (wind) when heating, not cooling, is the concern. Summer outages are usually of the "transformer blew up" or "someone hit a pole and knocked it over" type which can get a repair crew assigned and onsite quickly.

So, I've lived in Texas since the late 90's. There are certainly things that need doing. But this is not really a great analysis. A better analysis of the February grid failure is here: https://www.youtube.com/watch?v=08mwXICY4JM

Long story short:

1) the population has gone up a lot, so all infrastructure (traffic, electricity, water, housing, etc.) has had to increase rapidly

2) the mix 10 years ago was a lot of coal and oil, now it is much more natgas, wind, and solar. These can be reliable, but they are very different than coal and oil, and therefore a lot of new lessons have to be learned.

3) a big reason for the surge in renewables is the way the market is designed, which lets them pocket big bonuses at peak demand times. Since this _normally_ is in the summer during the day, this makes solar a lot more competitive financially. Natgas certainly failed in February, but all three of wind, solar, and natgas had big problems, it was just that the natgas and wind were unexpected while the solar was not (in the winter at night).

Thus far, the power grid is holding this summer, even though there is all-time record demand. We certainly do need to increase capacity, but that is mostly due to the surge in population, not any huge mismanagement. I'm sure it could be done better, but it's not unusual for record demand and a new mix of sources to put the system through a learning curve.

I’m in central Texas and my power has gone out twice this summer.

Not for a week like in February but I prefer the cold, you can put on a jacket. My house got into the 90’s and I could do nothing.

Could you install solar on your roof and a battery?

(Sorry if that's a hard thing financially, just a suggestion?)

What is this, a third world country?

If you live somewhere that encourages you to "pull yourself up by your bootstraps," grab hard and pull.

What type of actions on behalf of a John Q. Public would qualify as this 'pulling' that you're referring to? Just go solar, since wind really isn't feasible for a single family?

That 'bootstraps' is an old term that doesn't really apply to today's highly-siloed civilization. Everyone has their specialty and making anything out of bailing wire & bubble-gum just isn't as likely as it was in 1920.

Deregulation has been a major shift for most regions. Back in the day, the utilities would always plan the system for load growth and build new capacity and transmission when needed. This led to overbuilding as the utilities had a greater incentive to be reliable than to reduce costs. Later there was a big push to deregulate to let the free market make those decisions, but capacity is extremely expensive to build and prices have been very low in the past decade. Why that is the case is being hotly debated. The production tax credits for things like wind were very successful in getting many GW of wind capacity built, but they've distorted the market by letting those entities offer into the market much lower than anything else. From a green energy perspective, that is great, but there isn't enough storage yet (not anywhere even close), so conventional generators are still vital.

As far as the February event in ERCOT goes, this would've happened regardless I think if the utilities have zero desire to winter storm proof their equipment. All of that is incredibly expensive, so there would not have been a rush to spend money on something viewed as a hundred year event. With that being said. Hundred year events seem to happen every month now, so I'm fully supportive of new legislation and reform to the resource adequacy in the united states. I think it's hard for most people to understand that a huge amount of the natural gas fleet just couldn't get gas and that gas prices were in the thousands of dollars.

Yes, the ERCOT energy market doesn't pay enough for grid stability, i.e. idle dispatchable capacity.

Also the Texas regulator did not require enough weatherization. This is changing somewhat [0]. Also to be fair the Texas weatherization requirements currently exceed the NERC ones, though those are due to be revised this year. In any case they were obviously not good enough.

[0] https://www.texastribune.org/2021/05/26/texas-power-grid-ref...

> We certainly do need to increase capacity, but that is mostly due to the surge in population, not any huge mismanagement.

I’ve lived in Texas all my 40+ years and I fully believe it to be mismanagement for a utility to not account for growth. This state is always growing. Growth is neither unexpected or out of the ordinary. They should have a plan for an assumed level of growth with a contingency plan for how to handle their assumption being wrong. They have nothing except outdated and neglected maintenance because of short term thinking, which is only a side effect of mismanagement.

Well, I agree with the problem, but not the cause. Infrastructure maintenance and failure to plan for it are endemic in America. The only time it seems that ongoing maintenance is planned and funded from the beginning of a project is toll roads, where private-public entities can extract profit from the road.

My stronger point though, is funding. It isn't the fault of state utility operators that we don't have the capacity. It is the fault of local and state governments who have not reacted quickly enough to growth requirements.

Extremely valid point. I fear toll roads will be out of favor by the masses before their profit horizon kicks in. In a good portion of DFW, tolls are typically in the $300-$400 per month range out of a household budget. And, they keep adding more and more tolls. Some portions of town has an average household income that simply can not afford to get around town. Oh, and public transportation is a joke because... it's Texas.

ERCOT should be more vocal about their budgetary needs and the consequences of delaying planned expenditures. The growth numbers support it. Legalize weed to pay for it.

Houston area resident here. Largely agree with your analysis. Whether or not the great exodus from California is true or not, Im seeing lots of CA license plates around lately(along with more out of state plates in general). Maybe they drove here to visit friends and family, maybe they moved here.

In 2018/2019 there were rolling blackouts during the peak of summer, which to my knowledge we have not had on my section of ERCOT this year.

My beef with the grid is with the aging infrastructure. I live 40 miles from downtown Houston in what used to be a rural area, and now is becoming a suburban area. The local grid has not had any upgrades and has a lot more stress on it with the increase in demand. We lose power to a blown transformer or fuse on average of 4 times per month.

> the population has gone up a lot

Texas is growing ~1.5% annually over the last 10 years (15.6% for 10 years). This is all expected, it's not a surprise. And Texas is expected to grow to 40-50m by 2050. The power companies have access to the same info. It's ridiculous to blame the low capacity on population growth.

The real issue is that it is "unprofitable" to build spare capacity that will only be used 4-5 days a year. "population growth" is just an excuse.

goes to look at Texas' energy mix

https://www.eia.gov/state/?sid=TX#tabs-4

They've got nearly 30% non-hydroelectric renewables. That is a pretty remarkable achievement; it is quite hard to keep a grid stable with such high penetration of renewables. It doesn't say exactly what, so maybe more of it is dispatchable than I expect.

>it is quite hard to keep a grid stable with such high penetration of renewables

Perhaps worth pointing out the recent Texan outages were caused by gas freezing in the lines. An extreme weather event derived from climate change. So it's literally the fossil fuels causing the instability we've come to associate with the Texas power grid.

Then the governor came out blaming renewables.

I remain patient in waiting for the actual engineering report on the subject. I heard a rumour that that the reason wind was being given a free pass was because it was assumed that all wind generation would fail in an emergency - as it did - and therefore emergency planning was on the basis of no wind.

https://en.wikipedia.org/wiki/2021_Texas_power_crisis shows suspiciously low output by renewable sources. It isn't showing coal and nuclear being effortlessly surpassed by renewables (which is what I'd expect form the aggregate EIA stats)

The UK is running at 40% non-hydro renewables right now and we have a very stable grid https://www.gridwatch.templar.co.uk/

The current retail price is about £0.19/kWh or $0.26/kWH, Texas looks more like $0.12/kWh.

Yes and one of those is a stable supply.

True, I can only remember two power outages in my life time in London.

And people in the US commonly shell out hundreds of dollars per month for electricity. I've never even approached anything close to a hundred.

Is this due to fluctuating pricing or high usage? It seems that Texas bills vary between a monthly average of $100-200. That seems a lot to me, but the average unit price is pretty good.

https://www.eia.gov/todayinenergy/detail.php?id=47876

I had a quick search and immediately hit the headline “ Texas wholesale electric prices spike more than 10,000% amid outages”. That’s concerning.

https://mobile.reuters.com/article/amp/idUSKBN2AF19A

Consumers were insulated from this because they only pay the price in their contract, not the wholesale price. There was one provider that offered lower rates in exchange for not giving that protection, which did not work out very well.

In this case (power usage surge caused by air-conditioning use during a heatwave) I would have thought that solar was ideal, generating most power exactly when it's needed.

Texas is the biggest wind power generator in the country. North central Texas up near the panhandle has basically a constant 15-20 mph wind year-round. The entire landscape north of Lubbock is dotted with wind turbines.

> They've got nearly 30% non-hydroelectric renewables. That is a pretty remarkable achievement; it is quite hard to keep a grid stable with such high penetration of renewables.

I assume you mean it’s hard to maintain stability with a high percentage of non hydroelectric renewables?

I’m in New Zealand and we have a large grid for a small population. 84% of supply is renewable and we have very good stability (though we had some outages this week).

https://www.nzte.govt.nz/page/renewable-energy

> it is quite hard to keep a grid stable with such high penetration of renewables.

This isn't true (counterexamples: New Zealand, UK, Germany), and even if it were, saying it in this context is disingenuous because it implies that the problems with Texas's grid are due to renewables. They aren't: https://www.washingtonpost.com/politics/2021/02/17/energy-20....

It's nearly all wind, whose output fell all the way to 12.5% of installed capacity over a three-day period when homes were freezing (February 15, 16, and 17). [0] Output consistently stayed in the 15-30% range while I was watching it in ERCOT's live stats. Solar plants are nearly a rounding error. Every other source maintained 60-75% of capacity even with insufficient winterization.

[0] http://www.ercot.com/content/wcm/lists/181766/IntGenbyFuel20...

the comparison to installed capacity is interesting, but what really matters is a comparison to the projected supply from each of the sources, right? Was wind expected to be generating at 100% of installed capacity on those days, or was 15-30% in line with expectations?

The ERCOT projections I've seen are short-term [0], which means that after the most dismal days, the slightly less dismal 15-30% days "exceeded expectations". The fact-checkers reported this supposedly great performance without mentioning a single one of these details. I don't think these short-term projections are very useful for our analysis. It's as if your team finished the season 6-6, and all anyone mentions is how they won six games in a row so they're actually the best team in the league.

If you want to know "normal" output, you'll find that wind is very inconsistent. (I probably did find the mean at some point, but I don't have it at hand, and honestly don't feel that it matters). Wind is not "supposed" to carry the winter load, but ERCOT's record day for wind generation was set in February. Investors don't want to build power plants that could serve the expected winter load but will bleed money when wind has a good winter. Texas has built nothing but wind and solar for years because that's where all the profit is. ("Nothing but wind and solar for years" is according to D'Andrea's leaked phone call that got him fired, but I don't doubt his account of his own crony capitalism).

Wind is not "supposed" to be at fault for outages because it's expected to drop to as little as 1% output for short periods, but the reality is that we have built a power grid that does rely on wind. Our population is increasing dramatically, but we are not building any more reliable capacity because it's unprofitable compared to its subsidized competitors.

I'm happy that Texas is a world leader in wind generation, but we need a reality check. We have to build reliable capacity too. Winterization will help, but it's not going to do a thing to fix our problems during the spring and summer. We actually need more than that, because we occasionally have winter demand that matches peak summer demand. If we keep building so much wind, that will never happen unless wind happens to have a record winter in just the right year.

[0] http://www.ercot.com/gridinfo/generation/windandsolar

I don't understand how a hyper developped rich country like the USA tolerates so many grid failures over times. I do not mean that as a critic or boasting about ours or whatever, and I do understand (a bit of) the specifics that gives this end result, I just don't understand why there hasn't been a massive push by the population to make that issue one politicians had to fix.

Short of their pylons being taken down from storm/flood/... causing localized failures, the french grid virtually never goes down like this, and yet even then people complain if it takes too long to recover after a small localized blackout.

Or maybe this is a biased view I have from not living there and only getting fragmented news about it, and grid failures are actually very rare in the USA too ?

Could this be due to France's high use of nuclear energy (~70%)? I wish we had that, but it's nearly impossible to build here.

It probably does plays a part in why France's grid holds so well, but it doesn't explain why the US tolerate theirs.

How would we fix it without building nuclear plants? I don't think building a bunch of fossil fuel plants is going to go over well. Even Texas hasn't built anything other than wind and solar in years.

Maybe we tolerate it because fixing it is politically impossible.

The main problem is, because of various political idiocies, the US has put effectively no money in basic infrastructure in 50 years - so the lines are old and a lot of the more local base load generation has aged or is aging out of functionality. In addition to this, the power industry was one of the ones that got privatized in the 80's and regulation of it has been remarkably milquetoast outside of a few pushes like after the 2003 blackout.

The Texas problem has to do with ERCOT dodging oversight by keeping its grid separate from the other interconnects and within state lines, so they don't need to actually bother with much national regulation, like investing in heaters on their gas lines to keep the valves from freezing, for instance.

This is substantially misleading. Due to the 2011 outage, the Texan regulations are stronger and more strictly enforced than the federal regulations. Additionally, the primary issue with natural gas was due to suppliers not being able to get gas to the generators. The natural gas suppliers are federally regulated.

> I don't understand how a hyper developped rich country like the USA tolerates so many grid failures over times.

Because there is not such thing as the "USA grid". The US, unlike many other countries (FR, ES, UK, IT), is not a monolithic legal entity. It is made up of small legal units (states) that have sovereignty with-in their borders over certain matters:

* https://en.wikipedia.org/wiki/Federalism

There is no "US electrical grid" in the legal ownership sense. Rather there are a bunch of smaller grids tied together, each somewhat free to do what they wish with-in guidelines. Different areas may choose to tolerate grid failures more than others depending on how they want to allocate resources: some may want 0 issues per year and pay more for that level of reliability, while others may be okay with 1-2 so that prices are lower.

Apart from anything else, Gizmodo repeatedly confuses the PUC (public utilities commission) and ERCOT, both in its criticism, and factually when it refers to a PUC board member as an ERCOT board member.

Its just a psyops "punch Texas" article anyway, so getting the facts straight wasn't an issue.

Getting the facts right matters. At the same time, isn’t the Texas leadership that lead to the power grid fiasco deserving of intense criticism?

Here's hoping that D'Andrea was the biggest part of the leadership problem. He certainly represented himself as having too much power.

Title: "The Texas Power Grid Is Hanging on by Its Fingernails"

Second paragraph: "Even the high temperatures forecast for this week are within the normal bounds for the state in the summer: the average August high in Dallas is 96 degrees Fahrenheit (35.6 degrees Celsius), right around the temperature that’s expected to hit this week. In other words, ERCOT is prepared for conditions like this."

Clickbait much?

Some utility operator is going to have nightmares for the rest of his life over trolly probleming hundreds of people when he pulled the trigger to shed residential load... because ERCOT has been avoiding anti-frost equipment on its gas lines since the mid-90s.

Texas power is a mess.

Reply on news.ycombinator.com