California: Mar 21, 2021
Who turned out the lights? It’s dark in here. Don’t they pay their electricity? — Roger Rabbit
Ivo Welch
December 31, 2021
**************** call gas on demand $80-$200.
**************** add price graph from p.316
Clean.
High-quality.
Jack of all Trades:
Fossil fuels are one-trick ponies,
→ perfect economics only for heat.
Already know e-tech will improve greatly.
No moonshots necessary but welcome.
PS: Read up on primary vs. nameplate power.
Electrons are electrons, but
Time and location matter
Noon vs 8pm
Nevada vs. New York
Allocation problems are tremendous:
uncertainty (supply and demand);
short-term (which plants to switch on);
long-term (which plants to build where);
how to move e around (transmission).
Even more heterogeneous than USA.
Discuss US / California first:
Illustrative only, and
good to explain basic workings.
It is not enough to decarbonize US!
| U.S. | NatGas | Coal | Wind |
|---|---|---|---|
| Power | 45% | 20% | 10% |
| Energy | 40% | 20% | 9% |
| U.S. | Nuclear | Hydro | Solar |
|---|---|---|---|
| Power | 10% | 10% | 5% |
| Energy | 20% | 7% | 2% |
Plants are built primarily when
demand grows, and/or
old plants age out.
US demand is mostly stable:
For many years now, only Wind and Solar;
A few new natgas plants on East Coast.
Very rare nuclear plant here and there.
No new coal plants, few new dams, etc.
China, India, etc.:
Large new building programs.
Lots of new Wind and Solar plants.
Some nuclear plants, some others.
But huge coal plant building programs:
And new world record in 2021;
And coal employs millions of people!
Transmission grid is expensive.
Distance matters!
→ Supply is cheaper close to demand.
Low current transmission cost only because of generation proximity.
Complex regulation, tying in, etc.:
Regulators captured by utility companies,
with some good reason,
and a lot of lobbyists and lawyers,
and big hurdles for competitive clean engineering innovators inexperienced in politics.
Complex problems abound.
Only known good allocation system:
But free market must be shepherded!
alternative free-for-all cannot work either;
needs regulation, but is still somewhat manipulatable and indeed manipulated.
Compromises between scylla and charybdis.
Details are fascinating stuff.
In California (not elsewhere):
Capacity auctions for new plants (3 years);
Power auctions for delivery (daily).
Baseload Power.
Intermittent Power.
Storage (Dispatchable Power).
| Nuke | GeoT | Coal | Gas | |
|---|---|---|---|---|
| Calif | 9% | 5% | 3% | 34% |
| USA | 20% | 1% | 19% | 40% |
| World | 10% | 0% | 37% | 24% |
| Hydro | Wind | Solar | Other | |
|---|---|---|---|---|
| Calif | 18% | 10% | 12% | 9% |
| USA | 7% | 9% | 2% | 2% |
| World | 16% | 5% | 3% | 5% |
Electricity’s biggest cost is typically upfront plant building cost:
Construction / capital.
Plant scalability varies:
biggest scale needed for nuclear,
most flexible scale for Wind and Solar.
Cost of grid tie-in is high, too,
incl. regulatory costs,
… and does not scale linearly.
Idle (Standby) Minimum Running Cost:
Active Generation Cost:
Fuel and Extra Wear:
Tries to take all costs into account.
Projected over lifetime of plant:
disagreement over lifetime → different LCOEs.
If you got it wrong, …
… you may lose a lot of money.
Following are ballpark inflation-adjusted figures, differ by location, regulation, etc.
| Type | Today | est 2050 |
|---|---|---|
| Solar Panels, Roof | $100 | $30 |
| Solar Panels, Utility | $35 | $15 |
| Wind, onshore | $35 | $20 |
| — | — | — |
| Geothermal | $35 | |
| Nuclear | $70 | $60 |
| Gas, Always On | $40 | $45 |
| Coal | $75 | $65 |
| — | — | — |
| Hydro | $55 | |
| Gas, Dispatch | $200 |
Wind and Solar are the cheapest large-scale sources of energy that civilization has ever seen!
… and they will become even cheaper!
Unbelievably cheap, too cheap to meter!
≈ computers
… but they do not always work. ☹️
Clean intermittent generation for cheap is a solved problem:
$15/MWh or $5/MWh is relatively unimportant.
The silicon solar cell component will soon be negligible part of solar plant cost.
Costs will be primarily connections / running / maintenance.
Often best to colocate Solar, Wind, Batteries to share connection and operation costs.
Natgas is cheapest on-demand source of power in the USA:
Turn off, turn on.
Nearly infinite capacity,
but not as abundant everywhere else.
$40-$200/MWh — cheap, tough to beat.
Maybe should add $20/MWh for pollution.
Natgas often leaks on pipes and at EOL.
Makes natgas look cleaner than it is.
Think $100-$250/MWh
Requires large scale:
Environmental opposition to hydro.
PS: small-scale geothermal works well for home heating cooling and is economical!
Cheap running cost.
Very limited supply of good locations
PS: Also consider compressed air in caverns.
Primary nuclear problem today:
Fixed cost ($20 billion/plant).
10 years: potentially obsolete before open.
Honest Disagreements:
Put your money where your mouth is?
Gates, France, China are bullish.
UnionCScientists, Germany are bearish.
Hated and obsolete.
But kept alive by:
Abundant availability and e-needs now.
Fossil fuel subsidies and lobbies.
Large employment bases, pivotal voters.
Still, fortunately, plants are aging out now,
USA, 250 plants + 0 uc (construction)
Coal is still hated and obsolete.
But growing countries need power now.
Coal consumption and coal plants set new records in 2021!
The World’s Real Problem:
China, 1000 + 200uc .
India, 250 + 50uc .
World, 2,000 + 500uc .
Once built, sunk-cost equation changes:
30-50-year lasting impact!!
Any good ideas? Now is the time!
If demand is reasonably constant, choose
either basepower;
or intermittent + storage;
basepower + storage is not so great.
Power connection infrastructure is expensive,
more so for intermittent power.
Optimal solutions are often messy mixes.
One size ``no fit all.’’
In sum:
Problem is No Longer Cheap Generation!
Problem Now is Cheap Energy Storage!
Solve it, and wind/solar will take over.
Think $200/MWh.
Li comes in small manufactured cells.
Cells are finicky and small.
Not easily scalable:
Expensive packaging into small cells.
Twice the capacity is approx twice the cost.
Lots of mundane improvements, but
will likely always be finicky small-scale.
Problem: Wear out after 1,000 cycles.
Anode and cathode expand and contract.
But very soon 3,000 - 5,000 cycles!
→ Much cheaper for long-run use.
→ Very big deal for utility-scale storage.
PS: No scientific reason why not 50,000 cycles, either.
2 GW power out of 450 GW total.
10 GWh energy out of 4,000,000.
Problem is capacity, not power.
Batteries rapidly expanding, but still tiny:
Many specialty uses (e.g. 6-10pm).
Still too expensive for overnight,
much less for multi-day storage.
World Need:
Giant house-sized tubs with person-sized electrodes that scale according to whim,
with quad storage → only double cost,
with cheap chemistries, $50/MWh,
and then it will be lights-out for most power plants other than wind and solar in 20-30 years!
Possible, even likely.
But not here yet.
Better to spend lots of $$$s now on research and development than on deployment now.
I am optimistic. (10 years?)
But what if I am overoptimistic?
Think $250/MWh.
Categories: Flow, Dammed, Pumped.
Limited Potential (but still 2x today’s).
High capacity, lower efficiency than batteries:
Still cheaper for big energy needs.
Here now and can supply 8-12 hours!
Some even multi-day (Hoover).
Few suitable locations.
Large scale only.
Extremely high upfront fixed cost.
What if battery power becomes cheaper more quickly than expected?
Not crazy at all:
Maybe $100-$200/MWh ?
Perhaps not great long-run solutions,
but not uncompetitive either (for now).
Polar vortex-like wind/solar outages?!
Seasonal e-storage?
Long-term e-storage needs higher capacity at very low scaling cost, but lossier is ok.
Underground heat reservoirs?
Underground compressed air reservoirs?
Hydrogen?? (not as battery substitute!)
Always a bad idea.
Never store electricity for heat.
Store heat in insulated containers.
Heat storage can work on any scale:
Wind is always somewhere. Yet:
Grid transmission lines are expensive.
Transmission is lossy over long distance.
Transmission has other issues:
Power line breaks,
causing fires,
or even cyber attacks.
Very complex engineering and political problems:
It evolved naturally for local industrial needs.
Total regulatory mess in USA now.
Needs to allow more intermittent energy:
Intermittent adds potential wires overload.
→ Sometimes prices turn negative!
Even in California, often in Germany.
Secondarily, transmission grid
Feasibility/cost remain location-specific and complex.
Solar energy will cost $15/MWh. Wow!
If we could store and release for another $15/MWh (heck, $30-$50/MWh),
we could built 200+ GWh of storage,
plus electrify most fossil-fuel uses,
and it would be lights out for fossil fuels (for 2/3 of uses)!
But storage costs more like $200/MWh.
Reliability is hugely important.
When electricity goes out, the world stops.
All industry and jobs depend on it.
Inefficient to have too many decentralized backup generators everywhere.
| Coal | NatGas | Nuke | Hydro | |
|---|---|---|---|---|
| USA | 1,410 | 1,317 | 797 | 249 |
| China | 3,860 | 148 | 161 | 1,103 |
| World | 9,621 | 5,585 | 2,440 | 3,843 |
| Wind | Solar | (Oth) | Total | |
|---|---|---|---|---|
| USA | 191 | 39 | (2.2%) | 4,092 |
| China | 186 | 45 | (1.1%) | 5,562 |
| World | 828 | 263 | (2.6%) | 23,171 |
| Coal | NatGas | Nuke | Hydro | |
|---|---|---|---|---|
| USA | 593 | 1,953 | 594 | 294 |
| China | 3,556 | 803 | 1,002 | 1,448 |
| World | 8,115 | 7,306 | 3,025 | 5,548 |
| Wind | Solar | (Oth) | Total | |
|---|---|---|---|---|
| USA | 790 | 1,071 | (3.0%) | 5,458 |
| China | 1,001 | 3,379 | (0.4%) | 11,230 |
| World | 6,833 | 10,152 | (2.3%) | 41,953 |
Generation is basically a solved problem.
Storage is the unsolved problem.
Wish nuclear plants were a lesser conundrum.
Natgas is economic challenge for c.e.
Coal plants outside OECD remain vexing.