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

The future of solar power

  1. May 21, 2017 #101
    So, the key to viable wind and solar is not in direct generation of electricity, but in using them to create a storable form of energy that can be used as needed by big power plants. Most are looking at hydrogen gas.

    According to this wiki article, storage of hydrogen gas for stationary power plants (as opposed to the mobile car engine) is not especially problematic:

    By this scheme, the grid and all its power plants remain intact, but they would burn hydrogen instead of natural gas and coal. Wind and solar farms would not be directly connected to the grid: all their output would be used to split water, and the resultant hydrogen conveyed to power plants much like natural gas is now.

  2. May 21, 2017 #102


    User Avatar
    2017 Award

    Staff: Mentor

    I don't know when you did your measurements and cents are a problematic unit for electricity production, but with $6000 minus installation costs in batteries and solar cells that just barely cover your average demand you'll have no power frequently.
  3. May 21, 2017 #103
    Not connecting PV farms to the grid is obviously wrong. That would waste energy on conversion even at times where it is not necessary.

    The right way is to connect all energy sources to the grid, and also connect all storage to the grid. When other demand is low (e.g. night), generation does not dial down, it continues to run and this energy goes to storage. When peak comes, storage puts additional power to the grid. This way, generation does not need to be built up to support peak load alone.

    Utilities have already figured out this elementary logic and are going with it.
  4. May 21, 2017 #104
    I'm not sure if I should start a new thread for this.

    I heard this from a questionable source, and trying to fact check it myself did not make clear results. Can anyone here help me? Someone said:

    This was in the context of the Iraq war costing $6 trillion, and "accomplishing nothing". That part probably doesn't belong on PF, but the quote above ought to be demonstrably true or false.
  5. May 21, 2017 #105
  6. May 21, 2017 #106
    I don't think that answers the question. Let's keep it simple:

    Assume you normally use a typical 1 MWH per month. Let's say you install solar panels that exactly off-set your usage, so your net consumption is zero. What would your utility bill be?

    It sounds like your bill would be near zero. It sure sounds to me like you are using the grid for (near) free. Why do you say otherwise?
  7. May 21, 2017 #107

    Below is a link to an cost estimate sheet from an Ohio Solar company for several roof top installation with battery backup. The estimated utility saving and income over 25 years however seems over optimistic

  8. May 21, 2017 #108
    It sounds non-nonsensical to me. Our electrical power is from coal, natural gas, hydro, wind and small amounts of solar, biomass. We don't use very much oil at all for electrical production (0.6%).


    Unless you want to convert our entire transportation fleet, and heating oil needs to electric power, we still need oil. And conversion of everything like long haul trucks, trains, etc just isn't feasible at this point. We just don't have practical storage systems - I won't do the math, because we don't even know what numbers we are talking about, but offhand, I don't think a half-trillion $ would go very far in terms of storage and transmission lines for the US. You could work backwards to see where you come up in terms of MWHrs, I'll pass.

    I don't want to get into politics, but I don't think that protecting our oil supply is the only reason we spend money on defense in the Middle East.
    Last edited: May 21, 2017
  9. May 21, 2017 #109
    1 MHhrs per month?

    First I would probably be generating more than I could possibly use during the time of generation.so I get paid for supplying the utility. When the sun gets low I need to buy power from the utility when I pay for the connection to the grid. If I have batteries to store the access power I go off the grid.
  10. May 21, 2017 #110
    It's a nice round number, and pretty typical depending on where you live. Some places the average is higher, some lower:


    Why don't you answer the question? It was a simple question - if you net to zero for the year, what would your bill be?

    I'll look at your link in more detail later, but $48,700 isn't appealing to me. Says install on a flat roof (so typically will cost more), and I doubt those batteries last 25 years, are replacements factored in? And is it really sized to store enough power to cut your grid connection? Hmmm, looks like about 1.6 days worth of battery power? Sounds like trouble to me, better add a bunch more batteries.

    And I don't count the "rebate" that doesn't change the cost, it only changes who pays for it. I can make the most hair-brained idea sound "cost effective" if I get somebody else to pay for it.
  11. May 21, 2017 #111
    Why is it "over optimistic"? According to the numbers there, ROI is ~10 years. That's about the same that I hear from many other places (even other countries and continents) - prices fell quite a bit, PV installations pay for themselves in 5-10 years depending on local conditions.
  12. May 21, 2017 #112
    I've never seen 5-10 year ROI numbers for a PV system that included a large battery back up. From what I've seen, 5-10 can be pretty optimistic even for grid tie, with no batteries (and then only in the highest $/kWh regions).
  13. May 21, 2017 #113
    Since solar installs rate doubles every 2.5 years and price falls by about 20% in the same time period, maybe you just did not see latest numbers and remember the "old" ones from 2-4 years ago?
  14. May 21, 2017 #114
    Looking closer at your source: https://www.renewableenergysolar.net/blog/can-i-sell-electricity-from-solar-energy-to-bge-and-pepco/

    that does not appear to be the case at all. You don't pay for everything you draw from the utility, you only pay for the net. And you are using the grid both ways, as a source and a sink. So if you net to zero, you pay close to zero (there may be a small monthly connection fee or something),

  15. May 21, 2017 #115
  16. May 21, 2017 #116


    User Avatar
    Gold Member

    Even assuming solar is perfectly predictable for a moment, it is not dispatchable. Demand is largely inflexible despite decades of attempts make it otherwise, e.g. the hospital must have power at 8pm during peak load.
  17. May 21, 2017 #117


    User Avatar
    Gold Member

    This effectively describes two grids (plus storage) running side by side at different times, in place of the single one in use now, an enormously expensive solution. I suppose some wealthy countries can attempt such a scheme for some minority fraction of the grid before cost puts a halt to the enterprise. More importantly, the developing countries (India, Indonesia, Turkey, Vietnam, etc) can't and won't attempt something so expensive and, without nuclear, will go on building hundreds of coal plants.
  18. May 21, 2017 #118
    I don't see where you get two grids. "The grid" is the electrical grid, which is already in place. The way I see it, the companies who are now mining fossil convert to hydrogen farming and sell their product to the utilities. How you get hydrogen from the farm to the power plant is the same way you now get natural gas from the fracking sites to the power plants. Every time they open a new fracking site, they have to repipe as it is, don't they? A stationary hydrogen farm, on the other hand, would only have to be connected once.
  19. May 21, 2017 #119
    While I think there may be merit to converting excess renewable energy to hydrogen to be burned for a turbine generator, this doesn't seem like an issue at all, until we actually have considerable excess renewable energy.

    Of the variable renewable sources, solar PV is still a small component, wind is a bit larger. But if we only have excesses on occasion, that won't justify the infrastructure costs to convert it. As much as most of us hate the idea of letting energy go to waste, I think the reality is that until there is considerable excess, and it occurs regularly (allow the imprecise wiggle words like 'considerable' and 'regularly' - I'm just talking very generally), no one is going to invest the resources to store it.

    Now maybe if we define 'storage' as just long enough for time-shifting, a few hours worth to handle peaks that occur in the late afternoon/evening, when solar has dropped. There have been plans to use pumped hydro storage for that, the round-trip losses of hydrogen will be far worse than pumped hydro, but hydrogen wouldn't be dependent on the geography (or caves) required for pumped hydro (or caves).
  20. May 21, 2017 #120
    Apart from subsidies (that are subject to change and unlikely to persist long term) I doubt any arrangements that allow solar customers to be "freeloaders" are widespread or will persist where they are in place. Charging methods appear to vary widely, but where I live the rate I pay for power purchased is about 3x higher than the rate I get paid for the excess PV sold back to the grid (and I'm on a better than average deal in that) - I have heard of early PV feed in metering methods that simply subtracted what gets fed to the grid from what is taken from it but suspect it's an urban myth that anyone is still operating under those arrangements (around here). Also there is a fixed Service Availability Charge that is in addition to the balance between what is bought and what is sold and incorporates the largest part of the costs of grid services. So I certainly pay for grid services - given my low usage and rate of feed to the grid compared to what I take the SAC is the greatest part of the bill and the effective rate per kWh (total bill divided by power imported) is much higher than the stated rate per kWh. I'm aiming to get that price up to infinity, but if it looks like I can succeed I'll be thinking about disconnecting.

    Now, I am treated as just another customer and expect supply when I need it, to be charged no differently than any other customer, but there is a case for PV/battery fitted homes to be special cases - where the service is only occasional backup power provision but with them most likely to occur during periods of high demand (from large numbers of PV users switched from self use to import). Given how low my usage is and that I still get a significant amount of solar power even during overcast conditions, I suspect my contributions still exceed the costs of the service provided - but inherent in that is that over time greater proportions of that service will be provided by "peaking" plants rather than "baseload". That "peaking" supply - at higher than average prices - is what storage will compete with.

    Some other points - if you have more PV than you use then it can be cost effective to use the excess for hot water heating, especially if the prices you get for exporting the excess is lower than the price you pay for electricity for water heating. If the hot water system is a high efficiency heat pump a little electricity goes a long way. Solar thermal systems may have better conversion efficiencies and simplicity, yet electrical systems are very reliable and can have low costs.

    It's a transition and most homes and businesses in developed countries are serviced by existing, mostly reliable, grids; there is no current need for those that install PV and batteries to be 100% self sufficient - and good reasons to remain connected apart from the convenience of making use of that backup, including some that can advantage grid operations. That they cannot currently achieve 100% at reasonable cost doesn't mean they should not aim for a lower proportion as an interim step, even where it's not yet known how the rest can be achieved.

    Customers can potentially become active as participants in an energy market, with smart systems able to manage buying, storing and selling power on our behalf and in a market like Australia's National Electricity Market they could bypass the retailers (and a large part of the costs that extra level imposes) and deal directly with the wholesale market. Going away for a few weeks? The batteries can be run lower, selling when prices hit the predetermined threshold to maximise returns. A linked weather and electricity demand prediction service could allow the system to pre-charge our storage during "off peak" periods ahead of prolonged overcast conditions or expected high usage. Equitable arrangements may be contracted that allows the grid operators access to a proportion of stored power, as virtual peaking power plants or for emergency services. The old way of dealing with electricity supply is no longer applicable or even tenable.

    I think it would take determined efforts to put the solar genie back into the bottle. It will disrupt a sector that needs serious disrupting in order to undergo fundamental change; guiding that change towards desired outcomes - not preventing them - is what it's about. Emergent market forces from growing proportions of intermittent renewables will tend to create the incentives that policy makers have resisted, that will make investment in storage - and demand management and efficiency - increasingly attractive. Fixed fossil fuel plant - and nuclear (unless it gets special treatment for it's low emissions) - will be forced into greater intermittency during renewables rich periods and have to charge higher prices outside those times to sustain their economic viability; it will (or can be treated as) a de facto carbon price.
  21. May 21, 2017 #121


    User Avatar
    2017 Award

    Staff: Mentor

    Is it?
    I could criticize that they only have an "advanced nuclear" entry, no rooftop solar entry, assume solar panel usage for more than 100 years (seriously?), and check in more detail how they arrived at their numbers, but that is not even necessary. Nuclear power is available 80-90% of the time, solar power 10-20% depending on where you are (average power divided by peak power), wind onshore is a bit better but not that much. That alone makes nuclear similar to cheaper than solar and wind (apart from a few very windy places). Add the storage issue and nuclear power wins by a huge margin over wind and solar. It loses against coal and some types of gas, but only because these don't have to pay for the massive environmental damage they cause.

    Conversion: For 80% availability, $10/(kW*year) = $1/MWh
    Germany is getting there, and there are pilot MW-scale projects to produce hydrogen from electricity. It is fed into the existing natural gas system, combined with gas power plants it acts as storage system.
    Solar panels plus installation and infrastructure are ~$2/Wpeak if we take the values of nikkkom's source,, one trillion would give you 500 GW peak power, and an optimistic average of 100 GW. The US has an average consumption of 450 GW (2014). Windmills can help a bit, but not a factor 5. With a trillion dollars per year it would get more interesting.
  22. May 21, 2017 #122


    User Avatar

    Staff: Mentor

    My understanding is that you only sell back the generation to the utility (not the transmission or distribution). So the power you draw costs double or triple what the power you sell back to the utility costs. So you would need to generate much more than you use in order to have a power bill of zero.
  23. May 21, 2017 #123
    I'm sure that's true in some/many cases, but the link provided they said

    Now maybe there is fine print in there somewhere that breaks that down into production cost versus total cost including distribution, but I get the impression that consumers expect a near zero bill if the generate as much as they use.
  24. May 21, 2017 #124


    User Avatar

    Staff: Mentor

    That link doesn't say anything useful to the point we're discussing, but I've looked into it some and it does seem you are correct that at least for the time being, most net metering schemes involve selling back the power at retail rates. But that is likely to change as it constitutes as subsidy for solar users (and as solar has grown, subsidies are reduced):
    http://www.eei.org/issuesandpolicy/...ocuments/Straight Talk About Net Metering.pdf
  25. May 21, 2017 #125

    It turns out that we can get that statement to work based on the technicality that only 18% of our oil actually comes from the middle east. (We get 38% from Canada.) So far from eliminating all oil use, we only need to switch 18% of it to electrical.

    So let's see if I did this right:

    We get 1.74 million barrels of oil from the persian gulf per day.
    One barrel has 70,000 kW/day of energy.

    We get 121.8 GW of energy from the persian gulf. Can a trillion dollars in renewables generate this? MFB's estimate from just solar comes close, and I'm assuming that it takes 11 years to build all this stuff, so efficiency will likely improve.

    Now for replacing cars with electrics: Remember that we replace half our cars every 11 years anyway. (The average age of all cars on the road is more than 11 years. Did I interpret that correctly?) So the cost of replacing the cars is not the cost of the cars, but the ADDED cost of electrics vs gas cars. If we assume that we are doing the $1.5 trillion plan over 11 years, then half of our 253 million car fleet will be replaced. Divide $500 billion into that and you have $4098 per car, assuming all cars sold are electric. That many electric cars is well beyond what is needed, so let's assume only half the buyers take that money to by their electric cars, and put the rest into new powerlines.

    It sounds like it could work, unless I have made a mistake in the math somewhere.

    Edit: I just found another source that says we only get 12.9% of our oil from the persian gulf. I win this game.
    Last edited: May 22, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook