Difference between revisions of "Penny a kWh"

from HTYP, the free directory anyone can edit if they can prove to me that they're not a spambot
Jump to navigation Jump to search
 
(19 intermediate revisions by 3 users not shown)
Line 1: Line 1:
−
==Penny a kWh electricity== (split to another article?)
+
==Penny a kWh electricity==
  
−
For 40 years, people have talked about solar power satellites. Based on an entire industry in space, it looks like they could deliver power for a penny a kWh or less. However, the complication of building up the industry and the long lead times are discouraging.
+
Why try to get [[space-based solar power]] (SBSP) down to a penny a [[kilowatt-hour|kWh]]?
  
−
Done via a huge fleet of rockets hauling nearly a million tons a year to geosynchronous orbit, recent studies make a case for 5 cent per kWh, highly dependent on the cost of transportation into space at around $500/kg. The energy in rocket fuel is paid back in 40 days. It's the high cost of aerospace hardware that makes it expensive, not the energy.
+
This has to do with [[wikipedia:price elasticity of demand|price elasticity of demand]].
  
−
A moving cable space elevator gets very close to the minimum energy to lift a kg, 14.6 kWh. The energy payback time is just over a day (30 hours). Alas, we don't have the nanotube cable yet, though progress is being made.
+
Some points on the curve:
 +
* At '''a dollar a kWh''', the demand is near zero, a small number of military camps that would draw a few [[megawatt|MW]].
 +
* At '''ten cents a kWh''', SBSP could pick up Hawaii's electrical demand of a [[gigawatt|GW]] or two except that SBSP power doesn't easily come in small blocks. 
 +
** For the rest of this analysis see http://www.coal2nuclear.com/energy_overviews.htm (and ignoring the small difference between a Quad and an EJ)
 +
* At '''four cents''', SBSP gets most of the electrical power of the US, about 400 GW plus the rest of the world for another 1600 GW.  (Half average price for power.  Distributing electricity costs too.)  Building 2000 GW (four hundred 5 GW power sats) takes 4-5 years. Gross income from power sales (the power satellites could also be sold) at 4 cents per kWh would be would be 2,000 GW x 4 x $80 M/yr/GW or $640 B a year, though some of this would have to be sold for less as off peak power.
  
−
Laser propulsion isn't as well developed as rockets, and does not develop a lot of thrust but it is very efficient on propulsion mass while using a lot of laser power.
+
In the early years, extra [[rectenna]]s will allow a premium for switching blocks of peaking power around.  Later, North American off-peak power in excess of base load can be switched to Canada to make hydrogen to upgrade tar sands oil.  This would raise the effective amount of oil since current extraction/upgrades burn one barrel for each 3 produced.  
  
−
If we were willing to build an 8 GW laser (cost estimate $80 billion) we could use it and a low performance "pop up" rocket to 250 miles to push close to a million tons of construction materials out to GEO at a cost well under $100/kg. (The laser takes 15 minutes to accelerate the payload into orbit from a sub orbital trajectory.) The energy pay back is about two weeks.
+
As the cost of power declines to a '''penny a kWh''', space-based power picks up the entire oil and gas markets. The only source that competes is installed [[hydroelectric power|hydro]].
  
−
This allows building power satellites for under $800/kW. The electricity could be sold for a penny a kWh or less and synthetic gasoline from the electricity sold for about a dollar a gallon.
+
==Cost of Space Based Solar Power==
  
−
On a crash program, we could be adding upwards of 500 GW of new power per year, in as little as seven years. That would be more than enough to compensate for the fall off in oil. Or as the military study concluded last year:
+
The rough cost of energy from pace is based on:
 +
 
 +
1.  8000 hr/year (90% on time)
 +
 
 +
2.  payback of capital in ten years (ten percent of cost per year in revenue).
 +
 
 +
So the entire production cost is to be paid by 80,000 hours of revenue.
 +
 
 +
A power sat production cost per kW is $/kW for the rectenna plus $/kW cost for the power sat parts, plus kg/kW x $/kg to GEO
 +
 
 +
This ignores production labor even at GEO.  If the construction rate is a million kW/day, a labor cost of a million dollars a day is only a dollar/kW.  At around $1000 per installed kW, that's about 0.1%
 +
 
 +
The key here is that kg/kW is just as important as lift cost.
 +
 
 +
Energy cost (in cents/kWh) would be production cost in $/kW divided by 800.
 +
 
 +
To get penny a kWh, the production cost of the power satellite cannot be more then $800 per installed kW.  (Compare with current estimates for nuclear power of about $8000 per kW.)  The target here will be 2 cents per kWh with off peak power eventually being sold for 1 cent per kWh to make hydrogen. 
 +
 
 +
(Because power satellites at $1600/kg and no fuel cost are the least expensive source of power, they will eventually be built out to peak demand.  The difference between peak demand and baseload will be fed to hydrogen electrolizers.)
 +
 
 +
The [[rectenna cost]] estimate is $200/kW based on [[power inverter|inverter]]s costing $60/kW (same as PC power supplies which have the same parts). We can spend $1400 per kW on parts, lifting them to [[geosynchronous orbit|GEO]] and assembly.
 +
 
 +
Allocating $200/kW to the rectenna, $900/kW ($450/kW before 50% transmission loss) to buying power sat parts and $500/kW for lift to GEO, then the lift cost cannot exceed $100/kg for 5kg/kW power sats.
 +
 
 +
==Thermal alternative to PV==
 +
 
 +
Most power satellite work has assumed PV cells because they are already well proved in space.  Having no moving parts helps for communication satellite that cannot be serviced at all.  For 10 GW scale power satellites thermal turbines may be attractive.  They would probably need to be installed in counter rotating pairs to minimize angular momentum problems. The Excel file here [radiator_temperature_extended.xls‎] shows minimum mass at ~130 deg C. 
 +
 
 +
'''Next''': [[Hundred dollars a kg]] (cost to [[geosynchronous orbit|GEO]])

Latest revision as of 22:42, 21 May 2012

Penny a kWh electricity

Why try to get space-based solar power (SBSP) down to a penny a kWh?

This has to do with price elasticity of demand.

Some points on the curve:

  • At a dollar a kWh, the demand is near zero, a small number of military camps that would draw a few MW.
  • At ten cents a kWh, SBSP could pick up Hawaii's electrical demand of a GW or two except that SBSP power doesn't easily come in small blocks.
  • At four cents, SBSP gets most of the electrical power of the US, about 400 GW plus the rest of the world for another 1600 GW. (Half average price for power. Distributing electricity costs too.) Building 2000 GW (four hundred 5 GW power sats) takes 4-5 years. Gross income from power sales (the power satellites could also be sold) at 4 cents per kWh would be would be 2,000 GW x 4 x $80 M/yr/GW or $640 B a year, though some of this would have to be sold for less as off peak power.

In the early years, extra rectennas will allow a premium for switching blocks of peaking power around. Later, North American off-peak power in excess of base load can be switched to Canada to make hydrogen to upgrade tar sands oil. This would raise the effective amount of oil since current extraction/upgrades burn one barrel for each 3 produced.

As the cost of power declines to a penny a kWh, space-based power picks up the entire oil and gas markets. The only source that competes is installed hydro.

Cost of Space Based Solar Power

The rough cost of energy from pace is based on:

1. 8000 hr/year (90% on time)

2. payback of capital in ten years (ten percent of cost per year in revenue).

So the entire production cost is to be paid by 80,000 hours of revenue.

A power sat production cost per kW is $/kW for the rectenna plus $/kW cost for the power sat parts, plus kg/kW x $/kg to GEO

This ignores production labor even at GEO. If the construction rate is a million kW/day, a labor cost of a million dollars a day is only a dollar/kW. At around $1000 per installed kW, that's about 0.1%

The key here is that kg/kW is just as important as lift cost.

Energy cost (in cents/kWh) would be production cost in $/kW divided by 800.

To get penny a kWh, the production cost of the power satellite cannot be more then $800 per installed kW. (Compare with current estimates for nuclear power of about $8000 per kW.) The target here will be 2 cents per kWh with off peak power eventually being sold for 1 cent per kWh to make hydrogen.

(Because power satellites at $1600/kg and no fuel cost are the least expensive source of power, they will eventually be built out to peak demand. The difference between peak demand and baseload will be fed to hydrogen electrolizers.)

The rectenna cost estimate is $200/kW based on inverters costing $60/kW (same as PC power supplies which have the same parts). We can spend $1400 per kW on parts, lifting them to GEO and assembly.

Allocating $200/kW to the rectenna, $900/kW ($450/kW before 50% transmission loss) to buying power sat parts and $500/kW for lift to GEO, then the lift cost cannot exceed $100/kg for 5kg/kW power sats.

Thermal alternative to PV

Most power satellite work has assumed PV cells because they are already well proved in space. Having no moving parts helps for communication satellite that cannot be serviced at all. For 10 GW scale power satellites thermal turbines may be attractive. They would probably need to be installed in counter rotating pairs to minimize angular momentum problems. The Excel file here [radiator_temperature_extended.xls‎] shows minimum mass at ~130 deg C.

Next: Hundred dollars a kg (cost to GEO)

Retrieved from "https://htyp.org/mw/index.php?title=Penny_a_kWh&oldid=18011"