Hacker News Reader

Comment by malchow | Hacker News Reader

High-level grid interconnection isn't so useful if one (or more) of the lower-level grids being interconnected is so badly maintained that it frequently has to be deenergized.

The future probably looks like microgrids, with MID/neutral-forming transformers [1] which generate their own 60 Hz pilot signal and allow multiple producers, batteries, and consumers to coexist on a common protocol even in the absence of the utility grid.

[1] https://enphase.com/sites/default/files/2021-06/Enpower-R1-Q...

View on HN

It's not an either or thing.

The nice thing of interconnected grids is that you can route around the bad bits. There's no such thing as a global shortage of power generation. Blackouts happen when there are local shortages. Which in turn usually means problematic local suppliers and a lack of connectivity to external suppliers. The key challenge on e.g. the European grid is moving renewable over production to where the demand is. E.g. Southern Germany firing up more coal/gas when the north has ample wind production is because they lack the transport capacity (i.e. cables). Grid interconnectivity increases the profitability of renewable.

Microgrids and batteries are indeed popular in much of the developing world where grids are very unreliable and power generation lags way behind demand. India, the middle east, much of Africa, and probably South America, etc. Grids are much less reliable there and investing in private capacity is essential and something that people do as much as they can so they can keep the fridge on, their phones charged, the AC on, etc.

In developed markets, people do the same but more for cost than resilience reasons. Though I can imagine Texans might be considering both after this year.

> The nice thing of interconnected grids is that you can route around the bad bits

But in the meantime you can get massive blackouts if the problem propagates, like in 2003. Has this been improved on since then?

I'm not sure what massive blackouts you are referring to. There were none where I live that I can remember; or in the years since; or before. Just not a thing in Northern Europe.

https://en.m.wikipedia.org/wiki/Northeast_blackout_of_2003

The electrical engineer in me wants to see bigger and better grids that allow us to better utilize more renewable (and intermittent) sources. Show me that transcontinental or intercontinental or transoceanic HVDC backbone that lets electricity slosh around all over the planet. Just in the US I fantasize about an HVDC line that runs the length of I-40 alongside an aqueduct covered in solar panels to knit together the eastern, western, and Texan grids while also bringing the southeast's excess fresh water to the southwest.

The pragmatist in me (and the witness to the difficulty of getting anything built, and the greater difficulty of getting anything maintained) thinks grid investment is both unlikely to happen and even more unlikely to work well. In particular, transmission is low-value, high-risk, and expensive. It's low-value because distributed generation and storage are getting cheap. It's high-risk because high-power-density things are dangerous (check out all the Western fires started by electric utilities' transmission lines and switchcraft). It's expensive partially for good material and access reasons, but also for bad political reasons (NIMBYism and the fragmentation of responsibility for large land areas) and simply real estate rights cost. It's like trying to build California's high-speed rail but with less value-add, so it's going to be a horrible uphill slog of questionable merit.

So yes, I agree with you. More microgrids with more distributed generation and storage are inevitable. And I think that they're probably going to destabilize and likely kill the large-scale electric utility as we know it in ~50 years. I often wonder why more power companies haven't already become telcos to utilize their poles to distribute internet access.

In Europe it is distinctly less risky as you get arbitrage between different parts of the grid. You make money as long as their is a price difference. The technology is well understood. California is hardly a good case study.

I think the way grids develop does still depend on local factors. In the UK rural substations have quickly become constrained and have limited export capacity. Urban areas have more capacity and are seeing peaker and battery installation. But large arrays of solar and wind just need a big connection to get power from the middle of nowhere into big cities. Places where land for batteries or peakers will be super expensive and where solar and wind are impossible.

Also, if you are going to setup this kind of generation why bother selling to the public anyway. Find a ceramics factory or an steel works and run a private wire. You get a guaranteed customer who will agree prices years in advance.

> In Europe it is distinctly less risky as you get arbitrage between different parts of the grid. You make money as long as their is a price difference.

Yes, and in theory, the market can decide whether spatial arbitrage via long interconnections or temporal arbitrage (via batteries, shutting off industrial consumers, pre-running air-cons, etc) is better.

Perhaps a combination of approaches will prevail:

The different arbitrage opportunities mostly make money off the price spikes they can smooth.

Simplified: the first long range cable you install earns the most money, because it can pick off the highest spikes. The second cable will cost just as much as the first one to install, but will have to find its profits in a world with already slightly blunted price spikes.

Similarly for batteries. But eg batteries and long cables can pick off slightly different spikes, and the back-and-forth flow in cables doesn't have to average out to zero (like batteries do).

> I often wonder why more power companies haven't already become telcos to utilize their poles to distribute internet access.

I think there is a business side to this wherein big cable co made some compelling economic argument and exclusivity arrangement with the power company.

Clearly, utilizing the pre-existing infrastructure and doing it all in-house would yield high-quality engineering outcomes.

The German railway company Deutsche Bahn used to have its own telephone network. It's since been spun out, I think.

I don't see how grid transit is a long term solution. Ultimately it doubles down on the problem of renewables. If Florida is consistently exporting solar power to the North East then it's two regions that go down if Florida is unusually cloudy.

You typically don't interconnect over long distances via alternating current anyway. You interconnect via direct current.

See https://en.wikipedia.org/wiki/High-voltage_direct_current

sounds like a dismal future. I want to live in a future where we invest in infrastructure, not some dystopian world where tech-bro fantasy provided the rational for systemic disinvestment

Economic progress means doing more with less.

That includes physical and human capital as well.

Economic progress means doing more with what you have, and the reason why economic growth is exponential is that you are creating more capital as part of the production process.

Capital accumulation is one big part of how we can produce more. But it's far from the only one.

See eg https://en.wikipedia.org/wiki/Total_factor_productivity

Doing more with any given amount of capital is very much part of economic progress. (Of course, if you can grow your capital, that's maybe even better. But capital ain't free: what goes into capital production doesn't go into consumption. And what goes into capital production for purpose A, like infrastructure, doesn't go into capital production for purpose B.)

> Doing more with any given amount of capital

yes, I'm glad you agree. It's not about doing more with less. It's about doing more with what you have. Capitalism does not promote reduction of either the capital stock or human capital -- in fact, just the opposite!

Yes. See eg https://en.wikipedia.org/wiki/Dematerialization_(economics) however: we are really using less stuff. Especially per capita.

If you arbitrarily limit yourself to developed countries, but the planet has no subdivided boundaries, it’s one system.

The developing countries are roughly going through the same stages, but faster.

Reply on news.ycombinator.com