Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Solar power and power loss over distance

  1. May 28, 2015 #1
    I was having a discussion with a friend, and my side of the argument was that for solar to work you also need a lot of fossil fuel running in the back ground just during the day, making it very expensive. Let's take New york as an example.

    Let's assume they have enough solar panels to produce 150% of the power needed on average during a sunny day. But when it gets cloudy , you could have a lot of power outages if the sun suddenly goes away, meaning you need fossil plants in the back ground, wasting a lot of energy. Because it takes time to fire up power plants.

    Now the question is, could you set up a network between all cities with a range of a 2-3k miles where if it becomes cloudy and rainy in one city (so solar can only sustain about 20-30% of power needs), other cities within a range of hundreds, or possibly a several thousand mile can jump in? Assuming ofcourse every city in this network has power left over during sunny days.

    How much would be lost per mile? Or would we really need batteries to make this sustainable.

    Could it be possible to set up massive solar farms in the desert in california? And power all of the US?

    Edit: my understanding is that A = V/ Ohms

    So if you generate a certain amount of Amps (that is electrons per second right?), and you want to push a lot of them over large distances, you need to send a lot of them at once? Meaning Voltage is high? So if differences would be small, and only small amounts of power would be sent at the time, more energy would be lost due to resistance? Or am i misreading that formula?
    Last edited: May 28, 2015
  2. jcsd
  3. May 28, 2015 #2
    You could use the same methods that power distribution companies use. To send a lot of power over long distances with low loss, the power company steps up the voltage to very high voltages using transformers and then steps it back down at the site where the power would be used.
    power lost = I^2 * R
    Because P and r are fixed values, less power will be lost if high voltages are used.
  4. May 28, 2015 #3

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    such a network exists, called "The Power Grid" .
    All generators connected to the grid are in synchronism,
    regions with surplus power pull ahead by a small fraction of a turn which pushes power into regions with a power shortfall.

    Ypu can watch that almost realtime here


    See the swath of yellow along Mississippi river? It's ahead of most of the rest of the country.
    I'd say at that moment Mississippi Valley was exporting power in all directions. Not a surprise, there's lots of coal plants there and in mild weather less efficient plants are shut off. Nukes are refueled in mild weather too, so as to be able to replace their generation with inexpensive fossil.

    old jim
  5. May 28, 2015 #4
    Way back in my power class they said as a rule of thumb 50% of power is lost every 500 miles. I don't know how true that is, but it's a number.

    Higher voltages do mean less current. But higher voltages don't mean more power flows. For technical reasons power flows by phase shift instead. And there's a limit to phase shifts and power flows.

    There's a type of power plant called a peaking plant. These can be turned on and off quickly. Mostly these are gas plants, but sometimes hydro plants are used (I think).

    Utilities seem to like a mix of plant types. It's not so much a question of whether a particular type will be used, but how much it will be used.
  6. May 29, 2015 #5


    Staff: Mentor

    That is a widely quoted number. I don't know under what circumstances it is true but it is wildly misleading.

    The world bank lists electric power losses for many countries here. http://data.worldbank.org/indicator/EG.ELC.LOSS.ZS In the USA it is 6%. Instructive are Sweden (7%) and Brazil (16%). In both of those countries their hydro power resources are located thousands of miles away from population centers where the power is consumed. During July, nearly all of Sweden's power generation is located 1500 miles from the loads. If what they said in your power class was true, their losses would be 87.5% instead of 7%.

    Back to the OP's question. California to NY no. But there are proposals to sell midwest wind power to NY and New England. But the extra power transmission capacity would cost billions. Wind generators in the midwest want to sell the energy "on our loading dock" and have the easterners to pay to transmit it. The easterners say that the westerners should pay transmission costs and to sell their energy "delivered to my door." The easterners also prefer the better deal they can get buying hydro power from James Bay in northwest Quebec, 2000 km away.

    Finance, not power losses are the important limitations.
  7. May 29, 2015 #6

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    either way, the owners of the transmission lines get their transport fee.
  8. May 29, 2015 #7


    Staff: Mentor

    That would be true, but in this case the lines never got built and midwest wind power is not sold in the northeast. It is very difficult to get investors to invest in power transmission nowadays instead of (for example) buying Apple stock.
  9. May 29, 2015 #8

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    It's a growing sector.


    one driving force is
    It's attracting European capital.
  10. May 29, 2015 #9


    Staff: Mentor

    Thanks for that Jim. Glad to hear that it's moving in the right direction.
  11. May 29, 2015 #10
    Do you know a number? I don't. The class I got the number in was a transmission line design class (undergraduate, balanced lines only). While 50%/500 miles seems excessive, 7% over 1500 miles seems way low.

    Finance isn't the big problem for new transmission lines. Regulation is. Few land owners want a new transmission line run anywhere near their property. That is the sort of thing that turns nice neighborhoods into slums. Since running such a line involves a serious decline in property values of thousands of registered voters, it is almost impossible to get right of ways approved.

    This in turn causes finance problems. Building a line involves buying up land. So a company starts buying land, then tries to get approval, then get bogged down in litigation for some number of decades. Finally, a corridor might get approved. But probably not, at least historically. So the finance company has spent perhaps a billion dollars, then carried that money for decades. In the end they have nothing to show for it.

    Yes, finance is a problem. But it's a problem because democracy is three wolves and a sheep deciding what's for dinner.
  12. May 29, 2015 #11
    So in theory you could put enourmous solar farms in the midwest then? And sell it to large surrounding cities within a 1-2k mile radius? If those panels become cheap enough.
  13. May 29, 2015 #12


    Staff: Mentor

    Yes in theory, but it makes no sense economically. The reason that we have the big grid in the first place is to support areas that are temporarily short of power by importing power from neighboring regions.
  14. May 29, 2015 #13
    Enormous ?? LINK -

    Each of the corner dots is an inverter - my estimate 1 to 1.5 MW each....

    As for long haul HV transmission -- wikipedia... LINK or basically at 765KV 0.5 to 1.1% per 100 Miles.... the 50% maybe should have been 5%?
  15. May 29, 2015 #14
    That sounds like a much better number. 17% at 1500 miles. That's not an unreasonable loss for shipping power. It could be lowered by about half if there were a dedicated (DC) line over long distances.

    So large non-photovoltaic collectors in the South West could power East Coast areas -- given the political will.

    BTW, 2013 overnight costs (pg. 10): Coal $2,934/kW; Clean Coal $4724/kW; Natural Gas $917/kW; Nuclear $5530/kW; Wind (on shore) $2,213; Photovoltaic $3,873.

    Gas costs 3.6¢/kWh in fuel, while the sun and wind are free (ish).

    Gas is the clear winner here. Those who wish to change our reliance on fossil fuels need to change those numbers through tax incentives, etc.

    There are unaccounted costs such as environmental damage. But which is worth more, the large footprint of a solar plant, dead birds, or climate change? These are questions we need to decide on a political level.
  16. May 29, 2015 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

  17. May 29, 2015 #16
    One disadvantage of a solar-wind-fossil fuel power grid is the increased startup cost of fossil fuel generators. Peak power demand occurs at about 5 pm when solar generators are under far reduced output. This prompts fossil fuel sourced power suppliers to ramp up standby generators at increased cost to consumers. It also means there must be more generators dedicated standby status at further increased cost.
  18. May 30, 2015 #17
    A very "hot" topic now is large scale energy storage for just this reason - to maximize energy harvest from these sources and to locally level the demand better. The industry is changing - and many consumers are willing to pay more for cleaner as well, so the cost needs to be competitive but does not need to beat.
  19. May 30, 2015 #18
    People are willing to pay more for clean until their electric bill goes up 400%. Then they make laws giving themselves subsidies. When these get large enough utilities start to lose money. Then they stop investing, rolling blackouts occur, and economic growth ceases for the region they serve.

    Perhaps that is the correct solution. That is a political question and this is not the forum for politics. Still, engineers need to be aware of the costs of their designs, and not just in dollars.
  20. May 30, 2015 #19


    User Avatar
    Science Advisor

    The 50% energy loss quoted is the probably the maximum loss over the whole system from generator to wall socket.

    Loss to 50% of the generated energy will be a drop to 70.71% of the voltage.
    The maximum acceptable voltage drop is usually considered to be 5% on any section of the system.
    For n sections we then have; 0.95n = 0.7071 ; then, n * Log(0.95) = Log(0.7071) ; so, n = 6.75 sections.

    An efficiency audit of those possible sections, each with a maximum of 5% voltage drop due to I2R could be:
    1. Generator, transformer and switch yard.
    2. First EHV Transmission line section.
    3. Middle section of EHV line.
    4. Last section of EHV line.
    5. Substation, local 6kV or 11kV distribution to final transformer.
    6. Final transformer and power line from street to structure.
    7. Meters, CB protection and within the structure.
  21. May 30, 2015 #20


    Staff: Mentor

    No, the losses I cited and linked in post #5 were end to end losses including pilferage. They are much less than 50%.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook