3 ms·

The $23BN Plan to Power Singapore from Australia

13 points by hindsightbias 5y ago

ggm 5y ago
The plan has obstacles. Finance is highly conditional on the sale of the power, but the power can't be sold until the project is complete for the long-haul subsea HVDC cable. So the capital works have to be done to completion before any income (not entirely unusual, but it stands out there is little phased income in this project as-is) It has to compete with other plans to use PV to run H2 production and sell the gas to other markets displacing gas and oil. Possibly this could be a revenue stream while the cable is being built, but it begs a question of "why build the cable" -And its other capital funding this alternate build. It might "cruel" the market for this project The cable has to go past several economies which would want a tap but who also extract ground-rent for subsea work. Indonesia is notoriously difficult about cable maintenance. So, there is potential for revenue to ID markets for electricity and at a significant discount of cable cost because it's so much closer: begs another question: why send the power to SG, when you can sell it to ID?
- Keppl8R 5y ago
  I wonder why they don't build and sell power to the local market in Australia and then use those funds to build the cable?
  - hazbot 5y ago
    A lot of the local markets in Australia are further away from the NW corner than Singapore!
    - ggm 5y ago
      There isn't currently a national network. There's the eastern seaboard with S.A. and there's W.A. we don't routinely send power these distances in Australia.
  - jlawer 5y ago
    The NEM (National Energy Market) is doing fairly well on rooftop and PV Solar. Current growth is actually causing problems as typical base load power (coal & combined cycle gas) are becoming uneconomical during solar hours. As the NEM covers the east coast and south Australia (at least major population centres), that doesn't leave much place to sell it. WEM (Western Australian Energy Market) is more regulated and have their own PV capacity. The only way this could be economical into Australia would be if it had an energy storage capacity. Someone might do the math and build a battery to be powered from this and inject into one of the grids, but longer term, if it is to supply singapore, it makes more sense to put the batteries at the singaporian end.
    - danfo 5y ago
      Here's a link to NEM mix this week in case anyone else is interested: https://opennem.org.au/energy/nem/?range=7d&interval=30m https://opennem.org.au/energy/nem/?range=7d&interval=30m Maybe power from further away could help toward replacing the base power. A little more sunlight for the east coast peak. Less chance of a cloud or a still day on the east coast causing an issue like Callide coal exploding the other day. Yeah it is a stretch (literally thousands of kilometres). If we're being ambitious, I'm excited about off-shore wind. If end game for NEM is 2/3 wind and 1/3 solar (with storage/hydro/gas in single-digit percentages filling the gaps) there are many inevitable projects and growing pains to come.
- toomuchtodo 5y ago
  > why send the power to SG, when you can sell it to ID? SG is wealthier and also burns quite a bit of natural gas (demand is roughly 6-8GW constantly [1]). [1] https://www.electricitymap.org/zone/SG?wind=false&solar=false https://www.electricitymap.org/zone/SG?wind=false&solar=fals...
dzhiurgis 5y ago
A HVDC cable from Sahara to NY would loose 20%, but solar is 33% more intense in a Sahara. Basically it saves on overnight batteries and is a hedge against weather and other factors. Also as solar install itself is around 50% of its cost, it's probably cheaper to do so in a dessert in a country with low labour cost than over your house. Dunno how to run the numbers tho. I'm sure it's relatively simple. Edit: Grid scale solar "levelized cost of energy" is 30-40 $/MWH, or 36-48 $/MWH if you account for losses across ocean (3.5% per 1000km). Cost of HVDC cable, based on Singapore link - 4,500km @ 3GW - $5.6b. Western Sahara to NY 5,600km = $7b. Wastern Sahara gives you 5.250 kWh/kWp per day. I now need to calculate how much power can you pipe thru the lifetime of the cable. Then compare that with grid scale batteries. 15.75 GWH per day or 172 TWH per 30 years. $7b/172twh = $0.04 / kwh for transmission, excluding maintenance. Battery storage is 115-150$/MWH or 0.115-0.15 $/KWH. So kinda looks it's cheaper to have HVDC.
- erentz 5y ago
  National security implications of relying on such a cable strung across an ocean to power your economy seem quite large.
  - dzhiurgis 5y ago
    This is de-risking. Also it’s a minuscule percentage right now.
  - chmod775 5y ago
    That ship has sailed long ago. I doubt there's a country on earth that could protect all of its critical infrastructure against another nation with modern weaponry and tactics. I wouldn't count on either country having much in the way of an energy grid left after a week or so. Additionally war will disrupt supply chains for a lot of industries, stuff you need for power grids included. This is a popular example to illustrate how little it would take to cripple a nation's power grid: https://en.wikipedia.org/wiki/Metcalf_sniper_attack https://en.wikipedia.org/wiki/Metcalf_sniper_attack Don't even need fancy long range missiles as long as you have clandestine operators in the target country. In a discussion of this (on HN I think) someone pointed out that replacements for these transformers probably has to be ordered from Europe, which can take as long as a year.

13 points by hindsightbias 5y ago

ggm 5y ago

The plan has obstacles. Finance is highly conditional on the sale of the power, but the power can't be sold until the project is complete for the long-haul subsea HVDC cable. So the capital works have to be done to completion before any income (not entirely unusual, but it stands out there is little phased income in this project as-is) It has to compete with other plans to use PV to run H2 production and sell the gas to other markets displacing gas and oil. Possibly this could be a revenue stream while the cable is being built, but it begs a question of "why build the cable" -And its other capital funding this alternate build. It might "cruel" the market for this project The cable has to go past several economies which would want a tap but who also extract ground-rent for subsea work. Indonesia is notoriously difficult about cable maintenance. So, there is potential for revenue to ID markets for electricity and at a significant discount of cable cost because it's so much closer: begs another question: why send the power to SG, when you can sell it to ID?

Keppl8R 5y ago
I wonder why they don't build and sell power to the local market in Australia and then use those funds to build the cable?
- hazbot 5y ago
  A lot of the local markets in Australia are further away from the NW corner than Singapore!
  - ggm 5y ago
    There isn't currently a national network. There's the eastern seaboard with S.A. and there's W.A. we don't routinely send power these distances in Australia.
- jlawer 5y ago
  The NEM (National Energy Market) is doing fairly well on rooftop and PV Solar. Current growth is actually causing problems as typical base load power (coal & combined cycle gas) are becoming uneconomical during solar hours. As the NEM covers the east coast and south Australia (at least major population centres), that doesn't leave much place to sell it. WEM (Western Australian Energy Market) is more regulated and have their own PV capacity. The only way this could be economical into Australia would be if it had an energy storage capacity. Someone might do the math and build a battery to be powered from this and inject into one of the grids, but longer term, if it is to supply singapore, it makes more sense to put the batteries at the singaporian end.
  - danfo 5y ago
    Here's a link to NEM mix this week in case anyone else is interested: https://opennem.org.au/energy/nem/?range=7d&interval=30m https://opennem.org.au/energy/nem/?range=7d&interval=30m Maybe power from further away could help toward replacing the base power. A little more sunlight for the east coast peak. Less chance of a cloud or a still day on the east coast causing an issue like Callide coal exploding the other day. Yeah it is a stretch (literally thousands of kilometres). If we're being ambitious, I'm excited about off-shore wind. If end game for NEM is 2/3 wind and 1/3 solar (with storage/hydro/gas in single-digit percentages filling the gaps) there are many inevitable projects and growing pains to come.
toomuchtodo 5y ago
> why send the power to SG, when you can sell it to ID? SG is wealthier and also burns quite a bit of natural gas (demand is roughly 6-8GW constantly [1]). [1] https://www.electricitymap.org/zone/SG?wind=false&solar=false https://www.electricitymap.org/zone/SG?wind=false&solar=fals...

dzhiurgis 5y ago

A HVDC cable from Sahara to NY would loose 20%, but solar is 33% more intense in a Sahara. Basically it saves on overnight batteries and is a hedge against weather and other factors. Also as solar install itself is around 50% of its cost, it's probably cheaper to do so in a dessert in a country with low labour cost than over your house. Dunno how to run the numbers tho. I'm sure it's relatively simple. Edit: Grid scale solar "levelized cost of energy" is 30-40 $/MWH, or 36-48 $/MWH if you account for losses across ocean (3.5% per 1000km). Cost of HVDC cable, based on Singapore link - 4,500km @ 3GW - $5.6b. Western Sahara to NY 5,600km = $7b. Wastern Sahara gives you 5.250 kWh/kWp per day. I now need to calculate how much power can you pipe thru the lifetime of the cable. Then compare that with grid scale batteries. 15.75 GWH per day or 172 TWH per 30 years. $7b/172twh = $0.04 / kwh for transmission, excluding maintenance. Battery storage is 115-150$/MWH or 0.115-0.15 $/KWH. So kinda looks it's cheaper to have HVDC.

erentz 5y ago
National security implications of relying on such a cable strung across an ocean to power your economy seem quite large.
- dzhiurgis 5y ago
  This is de-risking. Also it’s a minuscule percentage right now.
- chmod775 5y ago
  That ship has sailed long ago. I doubt there's a country on earth that could protect all of its critical infrastructure against another nation with modern weaponry and tactics. I wouldn't count on either country having much in the way of an energy grid left after a week or so. Additionally war will disrupt supply chains for a lot of industries, stuff you need for power grids included. This is a popular example to illustrate how little it would take to cripple a nation's power grid: https://en.wikipedia.org/wiki/Metcalf_sniper_attack https://en.wikipedia.org/wiki/Metcalf_sniper_attack Don't even need fancy long range missiles as long as you have clandestine operators in the target country. In a discussion of this (on HN I think) someone pointed out that replacements for these transformers probably has to be ordered from Europe, which can take as long as a year.