Would drilling holes deep enough theoretically allow us to harvest thermal energy on a meaningful scale?
Just need a way to efficiently extract the heat from whatever depth and convert it to electricity by usual means.
https://en.wikipedia.org/wiki/Beppu
It's really an almost surreal experience, with encrusted hot spring water pipes and steam everywhere in the district. Sometime you even need to be careful where you step to avoid to be scolded by steam coming up from the ground, likely due to overflow from the hot water boreholes flowing to the rainwater drainage system.
And it's not just the modern hot water boreholes they use to supply the hot spring baths & other users. One time we even saw a traffic code & couple sandbags placed on a random hillside next to the road, as there was steam escaping out of it. :D
This also discounts stuff like the need for maintenance, that is hard to achieve at the bottom of a hot & likely very humid if not water filled borehole.
For that reason, most geothermal systems pump water down & then back up again (possibly using multiple wells) & have the heat engines at the surface, where they can be easily serviced & a good heat differential can be achieved, via air or water cooling.
This in the end, is generally an industrial operation though, not really something suitable for every single house. Still helps with maintenance, as you can provide energy for many houses & don't hat to maintain the geothermal power production equipment for each house separately.
And yes, potentially much more heat than we'd know what to do with currently.
Interesting hard sci-fi (well, basic physics principles mostly) about it by Isaac Arthur: https://youtu.be/jZQP2oNDkAM
Disclaimer: absolutely not affiliated with the man but deeply hoping that such perspectives become maintream, normal expectations. Not holding my breath, but one hacker at a time, we'll get there!
An example: https://www.cnbc.com/2018/11/06/drilling-to-start-at-the-uks...
Not quite there yet: https://www.uniteddownsgeothermal.co.uk/future-programme
In both construction time and cost it's a bit meh compared to wind/solar, its only advantage is dispatchability.
Could they cover the hole, pour water into it and use air pressure changes in the hole to generate electricity? There must be a way to make the hole air tight. I suspect that the rock near the bottom of the hole would already be air tight.
Also I never understood why steam engines release all the hot stream into the air? Doesn't that waste energy to let the hot steam out? Isn't it better to keep the heat trapped inside the system and generate electricity from the pressure only?
They don't. All practical steam engines have condensers that recover most of the water and as much of the heat as current technology and the laws of thermodynamics allow.
For this reason most steam engines in ships and elsewhere generally did have condensers & reused the steam as feedwater, as you describe.
Steam engines are old technology. They've been replaced by modern steam turbines, in which the steam is either cooled and recirculated, or used for other processes, or both.
Geothermal steam turbines typically use a heat exchanger and release the original steam, as geothermal steam tends to be very corrosive.
If you want a single project at a huge scale, check out https://news.ycombinator.com/item?id=15596350
There's the added issue that rock's thermal conductivity is low, and any thermal borehole would have a limited effective lifespan as it reduced the temperature of adjacent material.
Geothermal energy is a viable and widely tapped energy resource, where it's available. In almost all such locations, it's been substantially exploited, with two notable exceptions: the African Rift Valley (mostly in Kenya), and the Yellowstone supervolcano, a national park in the US.
Substantial developments exist in California (The Geysers), Hawaii, Iceland, Japan, the Philippines, New Zeland, and quite probably elsewhere. 1GW+ plants are possible, comparable with the largest practical thermal and nuclear power plants (generally 1-4 GW, though multiple plants or reactors may be co-located). Worldwide capacity as of 2015 is about 12.5 GW.
The two principle variants are standard and enhanced geothermal. A standard plant utilises naturally-occurring steam, and is far less expensive to develop. "Enhanced geothermal" involves boreholes and often water injection to provide power generation.
I'd followed the case of one such project in Australia, the Geodynamics Habanero project. I'd first read of that in 2014 through a grossly misleading and fatuously optimistic report which struck me as both odd and curiously fact-free. Digging showed that in reality the project was running years late, at 1/50th originally-planned capacity, well over budget, and with significant technical challenges.
https://old.reddit.com/r/dredmorbius/comments/1wpa90/how_not...
Checking now, it appears the firm plugged the remaining wells in 2015 and cancelled the project.
http://www.thinkgeoenergy.com/geodynamics-plugging-wells-and...
Even had the project gone as initially scoped, the wells would have had a useful life of about 20-40 years, after which all available useful thermal energy would have been extracted, and would have to be replenished over ... long time, possibly centuries or more. There's a reason the Earth's interior remains molten -- rock is a very good insulator.
I'm not an opponent of geothermal power -- where appropriate it's highly useful, dependable, safe, and proven. In Africa it stands to make a tremendous difference, where even a small plant would make a tremendous increase in the availability (and probably reliability) of electricity. I'd encourage consideration of developing even such normally off-limits natural park resources such as Yellowstone (specifically excluded from a USGS geothermal resource survey I'd checked on some years back).
But enhanced development through borehole-based wells looks like a very long shot.
Wikipedia's treatment of geothermal is good: