Hacker News Reader

Comment by tremon | Hacker News Reader

Load shedding is the correct technical term for it. It's not just a South African thing either, industrial consumers in Europe have the same stipulation in their power contracts: in case of emergency, the power company (network operator) may choose to suspend power delivery to protect the grid.

However, in the EU this 1) isn't a regular occurrence, and 2) the grid has a tiered system for load shedding. I can't find a reference for the tier classifications right now, but if I remember correctly: tier 1 consumers (the ones who will be disconnected first) are heavy industries with huge power draw, tier 2 are other commercial uses, tier 3 is residential, and tier 4 is critical infrastructure (emergency services).

I don't think I've ever experienced a load shedding event that affected residential areas. There have been local blackouts but I don't recall any grid-wide emergency events.

View on HN

The better technical term for what South Africa has is rolling black-outs that increase in frequency (and ultimately duration) depending on the amount of load that needs to be shed.

South Africa also has actual load shedding, where industrial customers are asked then forced to reduce usage, but the black-outs are entirely different. They are region-by-region complete loss of power for all but the hospitals (but often take out hospitals and emergency services too).

Compounding this issue, the grid is not designed to be switched on and off frequently. While there is a set schedule for the rolling blackouts, transformers and other equipment will blow under the sudden loads, meaning your regional blackout may last a lot longer than planned. You may also get blackouts sooner or completely out of schedule for region, for which no explanation is given.

Texas had tier 3 level of load shedding recently.

Based on the incompetent decisions that ERCOT has made over the past few "severe weather events", I would say ERCOT and ESKOM are eerily similar.

All my reading said that the decisions ERCOT made during the weather event were all very competent. They kept the grid from collapsing, which was their job.

The fact that extensive load shedding and brownouts etc needed to happen was not a result of decisions made during the weather event, but lack of infrastructure investment over decades (that may or may not have been under the control of ERCOT, rather than the legislature and state government regulators, but ERCOT certainly was on board with/lobbying for the conditions of deregulation that led to them).

During the weather event... ERCOT, the grid operator, did their job very competently. Making more capacity come online or reducing consumer demand was not really something that could be done during the weather event. They kept the grid from collapsing -- and it nearly did collapse, it was a close call. That was their job during the event.

> All my reading said that the decisions ERCOT made during the weather event were all very competent. They kept the grid from collapsing, which was their job.

I respectfully disagree. Demand might have been twice what supply was, but the way the load shedding was done wasn't smart at all. The rolling blackouts were done on a roughly 18-48 hour basis, i.e. if you lost power you probably lost it for at least 18 hours and I know plenty of people who lost it for much longer, up to 48 hours.

If demand was double what the supply was then it should have been possible to give people power for say 2 hours every 4 hours. Or maybe 6 hours every 12 hours. If demand outstripped supply by a factor of 4 then 3 hours every 12 hours.

But that's not what happened. A lot of houses have gas furnaces but they need electricity to run the fans and the ignition system and such. People died in their houses either from the cold or from running their cars in their garages trying to stay warm. A lot of pipes were burst due to not enough heat in the buildings.

I can think of at least a dozen ways in which the load shedding could have been better managed. I wouldn't agree that it was very competent.

> If demand was double what the supply was then it should have been possible to give people power for say 2 hours every 4 hours

The problem is if that electricity is being used for heat, the draw of the 2 hour on periods will increase to make up for the time off. Perhaps even resulting in more total usage, because when power is on, people will set the thermostat to 80 instead of 65, anticipating the power going off and it getting colder. When really you need everyone's expectations of comfort to change to 40 degrees (half the delta T).

The only way to win and supply some power to all houses is to keep the duty cycle so low that the amount of resistive heating equipment becomes the limiting factor (1 hour every 6 hours?). I don't know if they could have managed this, but it's certainly more complicated to figure out than simply lowering the period of the rolling blackouts.

Only so much of the load is heating, and heaters can only draw so much current.

Even 1 hour every few would stop food from rotting and let people do important tasks.

> Even 1 hour every few would stop food from rotting

I just put my food outside, in the car. I never lost power, but I didn't want to waste any electricity on cooling food when it was needed so badly.

ERCOT failed to adjust pricing appropriately after the initial emergency, which allowed energy producers to milk the situation for billions of dollars more than is fair or reasonable. It's not like their input costs went up by billions of dollars that required compensation.

> All my reading said

Comparing that to "All my expierincing" first hand. Not even close to the same thing. They would not have been in the stressed situtation in the first place had they enforced the recommended corrections for cold weather protection in the first place. The fact that they had to turn off power generation because pipes were freezing at the generating stations is an absolute joke, and is the prime cause of the stress.

The local weather forecasts remind people to move plants indoors, cover exposed plumbing, etc. Maybe they should also start including a friendly reminder to the local power plants to also cover exposed plumbing, but why would they listen to that when they've ignored government sanctioned reports from investigations into previous failures?

> ERCOT, the grid operator, did their job very competently "Turn off the power so the demand goes away" Check "Issue rolling black outs so that the demand is lowered and manageable, and give people a fighting chance" Nope. These were absolutely not rolling black outs, but static black outs. The method they chose to "save the grid" was just as incompetent as not maintaining their systems in the first place.

Thank you for more info!

Do you know of any good journalistic coverage of this, since apparently I've been reading bad stuff?

It seems important info for those who don't have first-hand professional experience to get.

I read articles from every where. Local stations had coverage of the damage and lack of "rolling" black outs, national outlets had coverage as well, and I'm pretty sure I remember seeing international coverage from BBC, Daily Mail, etc. I can't remember where I was seeing the scathing reports on ERCOT's ineptitude, but they were plenty.

https://www.texastribune.org/2021/02/17/texas-power-grid-fai... https://www.dailymail.co.uk/news/article-9270343/Texas-perfe...

just a snippet of a quick search from soooo many articles

> industrial consumers in Europe have the same stipulation in their power contracts: in case of emergency, the power company (network operator) may choose to suspend power delivery to protect the grid.

You can't really compare.

Industries can chose in their contract to be part of the load shedding system. It's done with their agreement and the system is purely based on the grid frequency. The disconnection is done on site in an orderly fashion. In a way, it's probably more correct to view it as industries electing to temporary disconnect rather than the provider stopping supplying them.

> I don't recall any grid-wide emergency events.

The European electrical grid is extremely resilient. It was accidently split in two a couple months ago and that was barely noticeable for the customers of either half.

American here--I've experienced load shedding once. The problem was a lineman started a wildfire that took out the primary source of power for the town. Until the line was repaired there was nowhere near enough power available, they directed what they had to things like the hospital etc and everyone else was in the dark.

There is also the mess we currently have in California. High winds blew stuff into high power lines and started a bad wildfire. In the lawsuits that followed the utility was told to shut down the risky lines under high wind conditions--which of course causes massive outages. Of course the utility is "at fault" because the ecology types blocked efforts to keep the vegetation away from the lines.

The utility is at fault because they didn't spend enough money on maintenance. As a for-profit company in an established industry, PG&E won't ever be able to "pop" like a SaaS company does, so profits and salaries for executives and shareholders have to come from somewhere. Spending less on maintenance won't bite you - until it does - but until then, thats money that can be put elsewhere like CEO pockets.

PG&E is at fault because they refused to properly maintain the power lines, and lines that had been hanging for years from the same supports where the supports failed are the cause of the fires.

See here: https://www.nbcbayarea.com/news/local/long-term-wear-found-o...

https://www.kqed.org/news/11792217/1987-report-suggested-pge...

I can't find it right now, but there was an absolutely fantastic RCA report that included all the information on the wearing down of the hooks and how this could have all been prevented.

Reply on news.ycombinator.com