Hacker News new | past | comments | ask | show | jobs | submit login
Desertec (wikipedia.org)
70 points by mooreds on March 23, 2015 | hide | past | favorite | 27 comments

What, no discussion? I frankly was more impressed by the description of the proposed Atlantropa project than I was of the Desertec (though both are ambitious!). That would be a dam stretching ~13 km and between 300 and 900 meters in height. That seems like an very impressive (and difficult) civil engineering project. The Three Gorges dam is 2.3 km long and about 200 meters in height, so Atlantropa would be significantly larger than the Three Gorges.

The failure scenario for that is pretty scary. At least I assume it is.

Well... probably not. The Med is large enough that you won't reduce its level a significant amount, so nobody will build houses on the new beachfronts, so a sudden dam collapse should send a wave a few centimeters high a few meters further than the tide already goes.

> The Med is large enough that you won't reduce its level a significant amount, so nobody will build houses on the new beachfronts

The description of Atlantropa in the linked article clearly notes besides hydroelectric power generation, the dam was meant to lower the sea level by more than 600 feet in order to create more land in the basin. A dam collapse would have been catastrophic.

Wow, when you look at this picture: http://en.wikipedia.org/wiki/File:Fullneed.jpg

You realize that our electrical problems could be solved in 5-10 hard working years.

This makes me think that maybe our current solar panels aren't as bad as they seem. Or we're really using them wrong.

It seems to good to be true. Also, if that project would succeed, I have a pretty strong feeling it will end up like us getting a larger hard drive. 1 year later people usage would raise above its capacity.

I'm dubious. First, it only works during daylight hours. Second, its ecologically disastrous (sterilizes the ecosystem under the facility due to lack of sunlight). Third, centralized generation adds transmission losses which are substantial.

I favor orbital generation, which mitigates all of these.

You got to be kidding. You can't seriously make the argument that we shouldn't do solar for all our power needs because it will rob sunlight from the ecosystem under the panels.

That's astounding. I mean, just a take look around you. Where's all the ecology gone?

Me and Greenpeace, yes. When we're talking square miles of ecology reduced to sterile dirt in the dark under panels, we're talking about disruption similar to plowing, or mining.

Ok, me and Greenpeace cofounder Patrick Moore

Continuing to burn large quantities of fossil fuels is also ecologically disastrous; given that the deserts are not processing large amounts of CO2 (compared to, say, forests, or phytoplankton), damaging them seems like the lesser of two evils. Orbital generation would be terrific, but seems more difficult to realize.

> ecologically disastrous

Assuming that it is - which has yet to be evaulated in each case - what would we lose if we plant these solar cells in a desert where noone really lives now?

What does it matter who lives there? That's absolutely irrelevant to the ecological impact of the project.

Desert doesn't mean the same as 'no ecology'. Its in fact more fragile that other ecologies, dependent upon scare resources and balanced with a very predictable annual cycle of sun and moisture.

Square kilometers of solar panels blotting out the light is a vast disruption of the desert ecosystem. There is no need to evaluate its impact on the ecosystem. Complete disruption is the result.

There are regions of deserts though where there is no ecology. For example, there are huge portions of the Atacama Desert that don't even have microbial life forms in the soil, so placing a solar facility there won't impact any known form of life. Additionally, there are solar panels that are transparent [1] which could also help minimize the ecological impact of solar generation. Furthermore, one can spread out the solar panels, to again minimize the impact in ecosystems. Yes, blindly using solar can be detrimental, but done right, you can mitigate the impact.

[1] http://www.bloomberg.com/news/videos/2015-03-23/invisible-so...

The interesting question, however, is not that. It’s whether that complete disruption is a worthwhile tradeoff. We make those tradeoffs all the time and just saying there is disruption is really not enough …

Solar has the disadvantage of disrupting ecosystems in geometrically huge areas, unexperienced in human history. Worse than strip-mining. Worse than city growth. On the order of the invention of agriculture.

Compare with Thorium reactors for instance, which could supply all our power needs for centuries, with negligible impact on the ecosystem. That's the tradeoff.

What would actually happen ? heat reduction under the panel will surely be a big change. Maybe it would allow an ecosystem to develop, being detrimental to the panel grid. Maybe it they will just share space nicely.

Compared to smaller solutions this project needs some hefty investment in order to get started. It might be useful to start building solar farms in the north african countries for their own needs and then they can start exporting excess electricity to neighboring countries in the south and - at one point when it's economically viable - via DC transmission lines to Europe.

Apparently a lot of the original shareholders pulled out, but it's still going forward[1]. I'm not sure how much the plan has changed.

1: http://www.reuters.com/article/2014/10/14/germany-desertec-i...

Is it not sufficiently obvious at this point that decentralization is better than centralization?

vegan, decentralised, and ecological energy ;) You know it is quite crazy calling electricty ecologically compatible, if produced by industrial waste and unethical factory farming (the power generated on giant solarplants at the roof of battery farms and the biogas produced by the waste of feedlots).

High voltage DC? What's that about?

It's more efficient for very long distance transmission.

With AC transmission at 60 Hz, the wavelength is 5000 km. Things get complicated when you approach that distance.

Also, with maximum peak voltage limited by arcing, DC uses the wires more efficiently because it's constantly at peak voltage, rather than oscillating up to it.

Also, AC current creates a varying magnetic field that pushes the current flow away from the center of the wire, so only the outer region gets used, increasing resistive losses. It's not a huge effect at 60 Hz, but it makes some difference.


But, the switching/control equipment is substantially more complicated, voltage conversion is hard due to transformers not working at DC, and you still need to suffer inverter inefficiencies to convert things to AC for domestic use.

Right, so it only makes sense for very long distances where the cost & losses in the inverters are less than the losses from AC transmission.

What happens when the transmission distance approaches the wavelength?

High voltage DC is actually cheaper and easier to do for long distances than AC. HVDC cables can be made cheaper because you don't have to worry about skin effect [1], and additionally, you don't have to worry about what phase each electrical system uses, nor the natural phase drift of electrical systems. It's why the west coast of the US uses HVDC for one of the paths of the Pacific Intertie.

[1] http://en.wikipedia.org/wiki/Skin_effect

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact