Ridiculous 'blue sky' ideas

[DenzilPenberthy]

In the thread about fusion power a week or so ago I rather  (over?)confidently stated that powering most of the globe from renewable power was an easier problem to solve than developing commercial mass-scale fusion power.

It seems to me that this would be an achievable goal with more-or-less current technology but it would require a truly never-before-seen level of investment and international cooperation. I realise that this puts it in the realm of my imagination rather than reality But unlike fusion, I think it's only a pipe dream for financial and political reasons. These seem to me to be easier to overcome than fighting the laws of physics.

AAAnyway... I don't want to talk about the relative merits of windmills vs tritium. What are your most out there, bonkers, pie-in-the-sky solutions to distributing and storing energy on a massive international scale?

The thought that always springs to my mind is a 'mega grid' of ultra HVDC power lines spanning the whole of the globe (or at least from Portugal to Kamchatka and all the way to the end of SE Asia). These would smooth out a good deal of the 'what happens when it's not sunny or windy where I am' uncertainty. Plus on top of this some massive scale storage. e.g. build some *huge* channels and use the Dead Sea as pumped hydro storage.

It doesn't have to be practical, what are your crazy ideas?

[HAL-42b]

You want crazy? I give you crazy! Matter of fact I was looking into that a few days ago but didn't dare to post it here.

How about a superconductive power line encircling the whole planet? We don't have room temp superconductors yet but we have some which can be cooled by liquid Nitrogen.

So this grid can take energy from one part of the world and transport it to another part which is in peak demand. Since sun is always shining somewhere and there is always somewhere with peak demand. Thus the entire energy demand of the world can be solved with solar and without the need for storage.

Crazy enough yet?

Matter of fact people are already trying to make this happen.





Actually if we listen to these guys we might not need superconductors at all.

More on Elpipes and large scale HVDC grids:
http://large.stanford.edu/courses/2010/ph240/hamerly1/docs/hartley.pdf

http://www.geni.org/globalenergy/library/technical-articles/generation/elpipe/Chinese_Journal_HV_Engineering_004_November+footer.pdf

http://www.ciwg.net/files/70091162.pdf

https://www.google.com/patents/US8796552

[Kleinstein]

There is an old (e.g. started in the 19320s or 1930s) crazy idea of building a huge dam at Gibraltar from Africa to europe, thus cutting off the Mediterranean see. As rather little water flows to the Mediterranean see, the level can drop by maybe 1000 m allowing for massive hydro power and possibly storage. Tectonics is on our side, so in a far future this might get very practical as needed dam gets shorter.

There is also a crazy calculation, using tide energy in central america between the pacific and caribian see. The down side is, this would slow down the earths rotation over time, faster than normal. So expect a 25 h day in the future.  The channel planed in Nicaragua might be a fist step.

[Tomorokoshi]

So expect a 25 h day in the future.

I could sure use that now!

[DenzilPenberthy]

These are exactly the sort of things I had in mind  

Pacific-Atlantic tidal through central America is an interesting one I'd not thought about before.

I seem to have it in my mind that mixing the two via the Panama canal was not good ecologically?

[Kilrah]

The thought that always springs to my mind is a 'mega grid' of ultra HVDC power lines spanning the whole of the globe (or at least from Portugal to Kamchatka and all the way to the end of SE Asia).

Transmission is pretty much the largest limitation at this point.

Already 10 years ago there were numbers floating around that the entire Europe could be pretty easily powered by a covering a relatively speaking small patch of always sunny north Africa desert with solar installations, but... there is no way to carry that where it's needed.
If I remember well it quoted that the longest practical point-to-point link like that with acceptable losses/cost was only about 1000km.

Overcoming that somehow would indeed be worth researching quite extensively. Simple environmental factors make it that places where most energy is available are those it's not comfortable to live in, so there will always be massive geographical separation between the bulk of potential production and usage.

[Philfreeze]

There is an old (e.g. started in the 19320s or 1930s) crazy idea of building a huge dam at Gibraltar from Africa to europe, thus cutting off the Mediterranean see. As rather little water flows to the Mediterranean see, the level can drop by maybe 1000 m allowing for massive hydro power and possibly storage. Tectonics is on our side, so in a far future this might get very practical as needed dam gets shorter.

There is also a crazy calculation, using tide energy in central america between the pacific and caribian see. The down side is, this would slow down the earths rotation over time, faster than normal. So expect a 25 h day in the future.  The channel planed in Nicaragua might be a fist step.

It would take more than 1Billion (short system)/ 1Mrd. (long system) years to slow down the rotation of earth enough to get a 25h day.
I don't think we will be around that long

Note: this is just an estimate based on earths angular kinetic energy and the world electrical energy consumption per year

[dfmischler]

How big a water tower would every building need to be able to pump enough water while the sun is shining so that you could generate enough electricity at night (using the energy of the falling water) to power your home?  Forget about the problem of freezing temperatures...

[HAL-42b]

Power = Flow X Height X g
Energy = Power X Time

Also,

Typical office daily power consumption = 0.78kWh / m2
Typical office floor height = 3.5m

Therefore if we had 10 story office building the height would be 35m high.


So we need to generate  0.78kWh/m2 * 10 floors / 24 hours  = 325W / m2 roof area

So we have 0.325 = Flow * 35m * 9.8n/s2 
                   Flow = 0.94 liters/second
                           = 56.4 liters/hour
                           = 1353.6 liters/day

We only have 60% efficiency so better make this 2250 liters/day for every m2 of roof area.

Completely within the realm of possibility. Seems low though. Someone better check this with a soroban.

We usually consider 700kg/m2 live floor load though. Some adjustments will be necessary to the roof load bearing structure.

[DenzilPenberthy]

If I remember well it quoted that the longest practical point-to-point link like that with acceptable losses/cost was only about 1000km.

I wonder if this is based on traditional generation or 100% renewable because if the fuel is free then you can afford to have a much more lossy transmission link than if you're paying for fuel. (Obviously I understand that the power isn't free because you have to build and operate the means of collecting it)

[Tomorokoshi]

We usually consider 700kg/m2 live floor load though. Some adjustments will be necessary to the roof load bearing structure.

I think an excellent location for solar panel installation is on the top of municipal buildings like schools. Many are either 1 or 2 levels, which gives them a lot of surface area out in the open. The number of such buildings and their size scales well with the local population and hence the local power load. Power usage within the facility tracks well with solar availability. The payback goes into the local community.

[Kilrah]

I wonder if this is based on traditional generation or 100% renewable because if the fuel is free then you can afford to have a much more lossy transmission link than if you're paying for fuel. (Obviously I understand that the power isn't free because you have to build and operate the means of collecting it)

Your sentence in brackets is what matters... Considering only the fuel makes no sense, it's the total operational costs that matter. Which are currently still pretty much always higher with renewables than with traditional methods.

[Kleinstein]

Storing Energy or transporting it over long distances are somewhat competing techniques so solve the problem of uneven energy production and usage.  The price of the energy decides how important efficiency is - something like a 1000 km line can in this respect still compete with storage. So a low price might be more in favor of storage instead of long lines.

You will anyway need a mixture of both storage and large area grid.

[gnasirator]

My favorite: http://heindl-energy.com/

Our current problem is not power generation, it's storage! And this is just a beautiful solution because the storage capacity grows by radius^4 while build costs only grow ~to r^2 as only the surface of the rock needs working. More info on the homepage.

I happen to have studied at the inventor's university but I've never met him nor does he know I exist. But I love the idea for it is the only one I've seen so far that actually allows HUGE energy storage capacities as it would be needed for completely renewable world power supply.

[HAL-42b]

water density = 1

rock density = 2.7

This might actually work.

[Philfreeze]

Power = Flow X Height X g
Energy = Power X Time

Also,

Typical office daily power consumption = 0.78kWh / m2
Typical office floor height = 3.5m

Therefore if we had 10 story office building the height would be 35m high.


So we need to generate  0.78kWh/m2 * 10 floors / 24 hours  = 325W / m2 roof area

So we have 325W = Flow * 35m * 9.8m/s2 
                   Flow = 0.94 liters/second
                           = 56.4 liters/hour
                           = 1353.6 liters/day

We only have 60% efficiency so better make this 2250 liters/day for every m2 of roof area.

dfmischler only asked to cover the night. which means we only need about 1/3 of that or ~750 kg/night.
Or to put it differently: It should be realizable when we have only 5 floors (375 kg/night) but probably highly inefficient.

[Clear as mud]

Flywheel energy storage!  I know there have been researchers killed by disintegrating flywheels and this has put people off of the idea, but what about having a flywheel located in a hole in the ground, spinning on a vertical axis, so that if it flies apart, the energy goes into the surrounding ground and hopefully doesn't destroy anything expensive?

Rotational inertia I=mr^2, so to maximize this you put the heaviest part towards the outside, build some sort of a doughnut shape maybe.  To add a lot of mass cheaply, make the flywheel out of concrete.  It won't spin as fast as a flywheel made of some exotic material, but it's a lot cheaper.  Add a steel hoop around the outside for structural integrity.  Unfortunately, the actual amount of energy stored is proportional to the rotational inertia, but also the square of the rotation speed, so maximizing angular velocity allows you to store more energy, but as I already said, it would be a lot more expensive to build a flywheel strong enough to go horribly fast without tearing itself apart.  Maybe the cheaper option can still store enough energy to be useful?  I envision this thing being maybe 4 or 5 feet in diameter, and 3 feet tall, maybe supported by permanent magnets?  If I'm not mistaken, I think someone recently published a way to support things by permanent magnets in a stable configuration.  Nevertheless, if necessary a control system could keep it stable, because you would need some sort of motor/generator built in anyway to get the energy in and out.

So, what do you think of this crazy idea?  Enough energy storage for household use?  Probably not cheap enough to build.

[Marco]

My favorite: http://heindl-energy.com/

How do they plan to keep the piston centered? It doesn't show up on their plans.

[Marco]

My plan for Europe and Africa : Fill the Sahara with plain old unconcentrated PV (perfect optical clarity not required, robots can brush them clean without water use). HVDC distribution network. Use thermal storage in solid medium at >700 degrees to drive steam turbines, use the turbines with natural gas for backup.

Cost wise it will become possible soon enough, the geopolitical problems seem less tractable to me.

[Kleinstein]

Flywheel storage is somewhat practical for short times, e.g. a few hours or so to get the real peaks. As the energy is proportional to the speed squared, it is more useful to have a higher strength material and run it faster. Heavy an slow has quite some problems with the supporting bearings.  Permanent magnets don't get stable in all directions, though it can help, but it can still get expensive and has eddy current losses. There also is kind of a limited magnetic pressure, which make scaling to large size difficult. So it is likely not very practical.
 
For short time storage, e.g. over a day and shorter there are several solutions (batteries, hot salt, flywheel, superconducting magnets). However the more difficult part with renewable sources is storage over something like weeks, enough to compensate for wether patterns. That's exactly where pumped water storage would fit in - you just need to find suitable ponds.  As this storage would be used only a few times a year, efficiency is not that important, but the cost for the stored energy really need to be low. So I think even hydrogen might be an option - not as the only storage but for long term.

Anyway reliable energy supply will get more expensive so we need to get used to a price for electricity that depends on supply, so that some users can use energy when it is available, even if the means larger factories, that only run half the year.

[djacobow]

My plan for Europe and Africa : Fill the Sahara with plain old unconcentrated PV (perfect optical clarity not required, robots can brush them clean without water use). HVDC distribution network. Use thermal storage in solid medium at >700 degrees to drive steam turbines, use the turbines with natural gas for backup.

Cost wise it will become possible soon enough, the geopolitical problems seem less tractable to me.

Europe utterly dependent on a physical resource located in the middle east, with a few links (cut points) connecting them to that resource? What could possibly go wrong? :-)

I still find the idea of using 20% efficient PV to make heat pretty hard to swallow, especially, since it is possible to make sunlight into heat with near >90% efficiency. I personally believe that the economics of PV are not going to improve all that much more. The reasons is that the cost of the panels is now small compared to labor, land cost, site prep, mounting, cabling, inverters, and other balance-of-system stuff. Perhaps all those other things will get cheaper too, as have PV panels, but except for maybe inverters, they're all very mature technologies.

[Marco]

I still find the idea of using 20% efficient PV to make heat pretty hard to swallow, especially, since it is possible to make sunlight into heat with near >90% efficiency.

Concentrated sunlight produced through high optical quality tracking mirrors, keeping these arrays functional and clean in a desert seems folly to me. Static PV panels are much more robust.

I personally believe that the economics of PV are not going to improve all that much more. The reasons is that the cost of the panels is now small compared to labor, land cost, site prep, mounting, cabling, inverters, and other balance-of-system stuff.

All of those have the potential for massive cost reductions. For instance with a HVDC international distribution network there would be no inverter costs added by the PV itself ... DC all the way to the inverters coupling to the existing national distribution networks.

PS. Europe would still have the steam generators internally, which could run off fossil fuels. We already rely on fossil fuel supplies from rather iffy allies now, nuclear is the only way to significantly improve that. If autarky is the reason to go nuclear though, it means we have a lot more problems to deal with. For the purpose of this thread I was being a bit more hopeful.

[djacobow]

Concentrated sunlight produced through high optical quality tracking mirrors, keeping these arrays functional and clean in a desert seems folly to me. Static PV panels are much more robust.

You might be right, but you have to overcome a 4 or 5 to 1 ratio of reliability. The quality of optics does not have to be fantastic, depending on the temp desired. The old California SEGS plants mirrors look like crap -- cracked, missing chunks, etc, and the plants still run.

But I probably agree that PV is a better choice than CSP. What I object to is PV and storing its output thermally. A better plan is to try to use some of that HVDC link to ship its output to different time zones, and to use battery storage, of compressed air -- who knows -- for the rest.
 

All of those have the potential for massive cost reductions. For instance with a HVDC international distribution network there would be no inverter costs added by the PV itself ... DC all the way to the inverters coupling to the existing national distribution networks.

Please elaborate. Maybe (super cheap) robots can do the assembly? Maybe copper wiring in conduit that can withstand 30+ years in desert weather will get much cheaper.

In an HVDC, system, there will, of course, be no inverter, but there will still be power electronics to do MPPT and to step up to the HVDC voltage. Nobody is going to make a string of PV panels 500kV+ long.

PS. Europe would still have the steam generators internally, which could run off fossil fuels. We already rely on fossil fuel supplies from rather iffy allies now, nuclear is the only way to significantly improve that. If autarky is the reason to go nuclear though, it means we have a lot more problems to deal with. For the purpose of this thread I was being a bit more hopeful.

You alluded to a problem of security of supply. I was only suggesting that your plan would do nothing to enhance security of supply, and might make the situation even worse. I'm not talking about perfect self-supply, but about making yourself utterly and instantaneously vulnerable to decisions made in an unstable part of the workd. Most long-term plans for decarbonizaton also require electrification of many uses that are not currently electric. That means that the electric sector will grow and be even more important.

Europe has repeatedly hatched plans to create massive engineering projects to exploit supplies on other continents. DESERTEC is only the latest. They are maybe feasible from an engineering perspective, but still infeasible for many other reasons.

[daveatol]

Typical office daily power consumption = 0.78kWh / m2

So we need to generate  0.78kWh/m2 * 10 floors / 24 hours  = 325W / m2 roof area

I get the following:
E = 35 * 9.8 * m = 0.78kWhr/m^2 * 10 (stories) = 0.78*1000*3600 *10 J/m^2
m = 0.78*1000*3600 *10 / (35 * 9. = 81866 kg / m^2

So for a 10 x 10m building, you'd have 8187 tonnes of water above the 10th floor.

[hayatepilot]

daveatol's calculation is correct.

So we need to generate  0.78kWh/m2 * 10 floors / 24 hours  = 325W / m2 roof area

So we have 0.325 = Flow * 35m * 9.8n/s2 
                   Flow = 0.94 liters/second
                           = 56.4 liters/hour
                           = 1353.6 liters/day

We only have 60% efficiency so better make this 2250 liters/day for every m2 of roof area.

You forgot the minutes in your calculation: 
     Flow = 0.94 liters/second
             = 56.4 liter/minute
             = 3384 liters/hour
             = 81216 liters/day

This would equal a pillar of water 81m high above the building!
Absolutely unpractical.