Skip to content

Comment on 'Breakthrough' geothermal tech produces 3.5 megawatts of carbon-free power

Comments

For reference:

3.5MW is a single largish wind turbine.

PV solar is ~200W/m2, so 17,500 square meters of solar panels.

A single nuclear reactor is commonly >1GW (1,000MW).

I know you're trying to compare geothermal to these sources, but I'm now in shock from the comparison between solar and wind. 1 large wind turbine (3MW) is equivalent to 7,500 x 400W solar panels. That's mind blowing.

Well... A large wind turbine is... a very large thing. I was very surprised the first time I saw one up close - I thought they looked small from the distance, but they kept getting bigger and bigger while I was still quite far.

yes, that too. I think another mistake people make is assuming they're all sort of the same size. But the offshore ones are immense.

The W rating of wind turbines is misleading, because it's the max output, which is very rarely attained. The real average output is usually just 10 to 20%.

Nuclear can typically deliver 90% of its max output.

They can also use power in certain situations. If the wind is especially weak, they're usually kept spinning using electricity because it takes more wind to start one from a stop than to keep it going.

I've seen one that's sometimes a net-loser on days where the other turbines are standing still. (based on what a landowner said, from the stats the operator gives him)

Nuclear can typically deliver 90% of its max output.

Even more than 100% rated power, if you allow it to Chernobyl itself.

This would be a carbon free option for places where the wind doesn’t blow reliably, the sun doesn’t shine enough, and either doesn’t need the scale of a nuclear reactor or needs a solution quicker than the long lead times for nuclear plants.

Notice how they carefully omitted the word "fracking" and went with "injecting rocks with chemicals to create fractures".

The number of megawatts doesn't really matter - what matters is the cost per megawatt.

If it is cost competitive, then millions of these sites can be installed to get as much power as needed.

I would argue the lower amount from fossil fuel sources matters most. Cost per megawatt should be secondary since the costs of adapting to a warming planet in the future will swamp any marginal per unit costs today.

The are already developing much larger plant in Utah

(Hard to say how much larger as the tech journalists only use "households powered" units)

They specified a conversion rate of 1 MW : 750 "homes". So if the Utah project can power 750 "homes", that's 400 MW.

It's common to use a conversion rate of 1.4 kW for a home, so that's really close to the 1MW == 750 homes equation.

Cool. Wind turbines are unreliable though

Depends strongly on location as do the other sources. Some locations have both strong and consistent wind. There is no one size fits all.

In practice, the amount of power you can "rely" on a wind farm to produce is so small as to be almost useless from a grid planning POV. Yes, dispatchable generation can have unplanned outages too. But the differences in those probabilities and the amount of power you can plan on getting is so vastly different that they just aren't in the same class.

In practice, Altamont Pass is still one of the largest concentrations of wind turbines in the world, with a capacity of 576 megawatts, producing about 125MW on average and 1.1 terawatt-hours yearly.

As long as it reduces the need to ramp up fossil fuel plants, it's a win.

Ironic you put it that way, since it actually increases the need to ramp up and down dispatchable generation. Average is literally a smoothing function. If you look at an hourly output graph of wind farms it's often hopelessly spiky.

IMO VRE farms should be required to use energy storage to establish some semblance of control of their output. Instead of being allowed to simply dump whatever energy happens to show up and making it everyone else's problem. It would reduce emissions too since gas peakers emit less in steady state.

And they are usually connected to a larger grid. If you cover an area large enough, the numbers will tend to be more constant.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.