The California Power Mess

[kennyp2339]

Great thread, and yes, this does seem to be a trend or perhaps an omen of the future for many other states in the US. We're in the midst of exiting the oil age and bumps are now starting to show.
Most energy delivery companies are focused on sub-transmission upgrades, mainly changing the old school 34.5kv construction to 69kv, this is to create a web were substantial load from regions can be shifted onto one another, or create a looped scheme system rather then the traditional radial feed for better system reliability.
One of the other major tie ups that many don't realize when discussing peaker plants, large solar fields, battery storage ect... is that one area might have issue with overlapping and serving another area due to simple phase rotation issues between substations, I see this pretty frequently when discussing distribution load shifts, circuit ties, new loop tie schemes.
I think one of the most important thing to rationalize is that the US is entering a new age within electric delivery, we're at the very start of trying to scientifically work through the existing politics / laws all while upgrading our grids to include sustainable load sharing capacity to minimize the need to reduce energy. Me personally, all the other noise in the room, is noise in the room for now, we have got to make a reliable foundation first, before we can build out and bring in the newest of the new tech on board.

 

[woodgeek]

Yawn. Seems like a lot of sour grapes being milked around this.

I haven't read too closely but is seems that what is happening is called a very strong heat wave, like a once in a decade heat wave.

New England had a once in a decade cold spell a few years back, and its fossil powered energy infrastructure struggled, utilities called for conservation, prices went haywire, profiteering, etc.

How is this different?

In the end, if you want spare capacity in a system, be it peakers in CA, or nat gas storage (or more pipelines) in New England, someone has to pay for that. Or rather than pay for it, they can endure a short shortage every ten years or so.

 

[peakbagger]

I think the bigger issue is do the utilities spend the money on proven fossil infrastructure to back up the grid in high use periods or do they go with emerging non fossil backup along with load reduction. The cheapest KW is the one they dont need to generate during a peak load event. I have been a long term an advocate of real time pricing which is proven to help encourage load shifting. That is a cheap way to cut peaks and spread them out.

 

[woodgeek]

Yeah, a natgas peaker plant and a grid battery do not work the same. The former is good at providing relatively cost effective backup power over a more extended multi-day period (assuming nat gas is available), while the battery is (currently) cost effective only for providing a lot of power over a much shorter period for grid stabilization, like minutes to hours.

The vision that I always had of grid batteries providing diurnal (or few day) storage for PV power in a 100% renewable grid...nope, that's not what they're being used for now. Its the energy storage capacity that is expensive, storing a an hour of power is much more affordable. So I think they are used only in a limited way for duck curve management to flatten the ramps and the peaks.

But since even CAISO is still only getting a fraction of its total daily energy from wind and solar, the rolling blackouts are not due to the intermittency of Renewables, or there not being 'enough batteries'. Its about not enough baseload or regional grid capacity to meet a **one week per decade** event. IOW, a LOT like the polar vortex problem in New England.

 

[ABMax24]

I think the bigger issue is do the utilities spend the money on proven fossil infrastructure to back up the grid in high use periods or do they go with emerging non fossil backup along with load reduction. The cheapest KW is the one they dont need to generate during a peak load event. I have been a long term an advocate of real time pricing which is proven to help encourage load shifting. That is a cheap way to cut peaks and spread them out.

My understanding was that California did sell electricity on the spot market like we do here in Alberta, but maybe I'm wrong.

Alberta runs on a live real-time price market, sometime the prices are 0 or very close to it when renewables are producing at periods of low demand. Other times prices max out at $999/mwh when there is a limited supply of electricity during high demand. During these events it creates incentives for other generating units to power on, as well as high demand users to sell their cheaper contract power back to the market for a profit. For the most part this stabilizes the grid.

For Alberta it works great, but most of our electricity consumption is by industry and they both monitor and react to price changes for financial benefit within their business. In my experience though homeowners have trouble deciphering energy vs transmission charges on their bills, let alone trying to monitor energy rates. As such our residential energy rates are roughly based the average monthly energy prices.

California would require a different approach. I think smart home/grid technology is required to monitor and adjust consumption based on prices without homeowner input. At least for appliances like; AC, washer, dryer, dishwasher, and car chargers.

You can actually see Alberta's current energy stats here: http://ets.aeso.ca/

Using the drop down tabs on the side you can go to current supply and demand and see every generating unit in the province.

 

[peakbagger]

California and many utilities do real time pricing at an industrial and commercial level but only a few utilities do real time consumer pricing. The "duck curve" phenomenon is driven by consumer demand so the way to flatten it out is push real time pricing down to the consumer.

 

[spirilis]

Keep the g** d*** nukes running and for the goodness of the environment, build more! Especially with Electrify Everything.

I can't wait until our society has dependable, reliable and consistently buildable nuclear power plants with a closed fuel cycle (likely fast-spectrum uranium or thermal-spectrum thorium+uranium) producing short-lived waste. These kinds of issues will look like the Dark Ages in an advanced "nuclear" powered society.

 

[peakbagger]

At this point in the US there is no private company that could afford build a nuclear power plant. No insurance company would bond it. Plant Vogtle in Georgia is probably the last large nuke to be built in the US for a long time. https://www.savannahnow.com/news/20200903/georgia-power-plant-vogtle-expansion-still-on-schedule.

Several projects have gone bust in the past few years and several companies associated with those projects are went bust also. Note the cost has doubled. The US long ago lost the ability to make several of the major components. last thing I knew, China and Russia were the only sources of the large forgings needed.

There are a couple of Small Modular Nuclear Reactors (SMRs) designs that are winding their way through the regulatory process. If and when one gets built is anyone's guess. They inevitably are smaller production line based designs that are designed to be gravity cooled in event of a loss of plant power. If the plug is pulled they just keep circulating by convection until the fuel if burned out. No matter what NIMBY still applies.

 

[spirilis]

Yes I know this. We're in the dark ages with low-carbon energy as I see it. Solar and wind are great but not-so-mysteriously disappears without control, the esteemed batteries may come but the price is sky high today. My guess is we will make incremental progress until we can reliably build nukes again.

 

[spirilis]

Yeah, a natgas peaker plant and a grid battery do not work the same. The former is good at providing relatively cost effective backup power over a more extended multi-day period (assuming nat gas is available), while the battery is (currently) cost effective only for providing a lot of power over a much shorter period for grid stabilization, like minutes to hours.

The vision that I always had of grid batteries providing diurnal (or few day) storage for PV power in a 100% renewable grid...nope, that's not what they're being used for now. Its the energy storage capacity that is expensive, storing a an hour of power is much more affordable. So I think they are used only in a limited way for duck curve management to flatten the ramps and the peaks.

But since even CAISO is still only getting a fraction of its total daily energy from wind and solar, the rolling blackouts are not due to the intermittency of Renewables, or there not being 'enough batteries'. Its about not enough baseload or regional grid capacity to meet a **one week per decade** event. IOW, a LOT like the polar vortex problem in New England.

I don't know, those polar vortex events happened a few years in a row. Oregon and WA are seeing abundant forest fires now. Could be once a decade but I suspect that will shock folks when it happens again next year or the year after.

 

[spirilis]

PS- Anyone else look at those forest fires and think "Man I wish I'd had a chance to fell & buck some of that for myself before Mother Nature took it back?"

 

[begreen]

At this point in the US there is no private company that could afford build a nuclear power plant. No insurance company would bond it. Plant Vogtle in Georgia is probably the last large nuke to be built in the US for a long time. https://www.savannahnow.com/news/20200903/georgia-power-plant-vogtle-expansion-still-on-schedule.

Several projects have gone bust in the past few years and several companies associated with those projects are went bust also. Note the cost has doubled. The US long ago lost the ability to make several of the major components. last thing I knew, China and Russia were the only sources of the large forgings needed.

There are a couple of Small Modular Nuclear Reactors (SMRs) designs that are winding their way through the regulatory process. If and when one gets built is anyone's guess. They inevitably are smaller production line based designs that are designed to be gravity cooled in event of a loss of plant power. If the plug is pulled they just keep circulating by convection until the fuel if burned out. No matter what NIMBY still applies.

So, assuming a sea change in the next few years as the need to get more nuclear online for baseload replacement of fossil fuel power, what has the most reasonable chance at success.? Yang proposed a rapid development of TSMR, but even that had an operational horizon that was quite far out. 2050?

 

[begreen]

I don't know, those polar vortex events happened a few years in a row. Oregon and WA are seeing abundant forest fires now. Could be once a decade but I suspect that will shock folks when it happens again next year or the year after.

California too.
I think we are going to need better solutions quicker. The fastest still seems to be conservation and improved efficiencies.

 

[kennyp2339]

California too.
I think we are going to need better solutions quicker. The fastest still seems to be conservation and improved efficiencies.

Its such a balancing act, I'm a proponent of business and commerce, but I think that all generation at this point needs to be regulated, deregulation has opened up Pandora's box so to speak with to much competition, that competition is what gave rise to the fossil market, mainly natural gas peakers, it also drove the actual cost per kw down, far below the minimum of what a nuclear plant can produce, many companies were and are looking for subsidies to keep nuclear online, basically it costs them money to produce electric, but at the same time, there forced to keep things running due to licensing issues, and the amount of cash they have tied up in infrastructure .
Also by shutting down nuclear, we essentially stopped advancing the tech aspect of it, there are theories and computer programs in place to mimic plant operation, but since we haven't advanced since the 70's layout, we are essentially behind the 8 ball, just lots to consider here.

 

[peakbagger]

So, assuming a sea change in the next few years as the need to get more nuclear online for baseload replacement of fossil fuel power, what has the most reasonable chance at success.? Yang proposed a rapid development of TSMR, but even that had an operational horizon that was quite far out. 2050?

B&W has been building effectively SMRs for the US nuclear program for 50 years plus. The government has a few billion in the budget to fund some trial designs in this decade. I think the optimistic schedules I have seen is 5 to 10 years for deployment. Rosatom in Russia is producing barge based small reactors and are accepting orders. I do not know if they are natural circulation design boilers. There are several other foreign designs but doubt they can get through US licensing hurdles quickly. Bill Gates is also invested in a somewhat mature SMR design.

 

[begreen]

B&W has been building effectively SMRs for the US nuclear program for 50 years plus. The government has a few billion in the budget to fund some trial designs in this decade. I think the optimistic schedules I have seen is 5 to 10 years for deployment. Rosatom in Russia is producing barge based small reactors and are accepting orders. I do not know if they are natural circulation design boilers. There are several other foreign designs but doubt they can get through US licensing hurdles quickly. Bill Gates is also invested in a somewhat mature SMR design.

Are most of these still plutonium-uranium reactors?

 

[peakbagger]

Are most of these still plutonium-uranium reactors?

My understanding is yes they are conventional fuel blends. Meaning lots of nasty waste. Lots of claims out there about various reactor technologies that use different fuel mixes that result in higher efficiency and lower toxicity of the resultant waste but above my limited understanding on how much is hype and reality. The thorium fuel cycles that India was working on were intriguing.

 

[begreen]

We have to do better and smarter this time around. Efficiency may be one thing, but there are many other costs involved in the lifetime of a nuke power plant that need to be accounted for. I haven't heard of the thorium fuel cycles in India. Off to Google land to find out more.

 

[spirilis]

I think folks' fears about the nuclear waste are vastly overblown and preventing them from taking a critical look at its value. The tech used right now, dry casks after cooling in a spent fuel pool for several years, appears to be a pretty stable technology. The high density of the fuel also means high density, i.e. compactness, of the waste. What entices me most about this stuff is that some 95% of its fuel value is still intact - when it becomes cost-effective to pull the stuff out of the casks and reprocess it (most of the fuel value would require fast-neutron spectrum turning the U238 into Pu, but there are designs intended to do this while keeping things on-site). As this will probably be decades away, the nasties radionuclides will have already decayed away and what remains has longer half-lives (and comparatively less danger/risk).

Most of what we use today employs straight uranium, enriched to 5%, some Pu is produced in the midst of the fuel cycle but most of it is fissioned away (some 1/3rd of the heat produced in the life of a fuel rod comes from the Pu produced by stray fast-neutrons hitting the U238).

The big problem here is speed. Renewables are going to have to cut it for now 'cause it's all we can build fast. I don't have high hopes that it's going to get us as far as the proponents hope. We'll fail to decarbonize fully by 2035 or 2040 or whatever the targets are.

The lowest hanging fruit IMO, is keeping existing nukes online. I keep hearing pro-renewables folks on twitter complain that new renewables are cheaper than existing nukes, but I've yet to hear them actually claim to deploy that much renewable energy fast enough to trade off when the nukes go offline. Nukes going offline results mostly in natural gas being burned more. It's a hideous scam for the environment.

 

[spirilis]

It's like everything else in the decarbonization story - Need breakthrough technology and needed it yesterday.
Second best, though, is building nuclear that we know how to do - and do it on wide scale with identical designs done repeatedly to improve upon construction costs/practices. USA never did this with civilian power generation, most of our nukes are bespoke designs altered in one way or another. France and S. Korea pulled this off though in the past few decades.

One of my concerns is the Vogtle AP1000's will probably be the first and LAST ones to get built in the USA. All that manpower & construction experience gained during the saga will be... a one-shot deal.

 

[spirilis]

We have to do better and smarter this time around. Efficiency may be one thing, but there are many other costs involved in the lifetime of a nuke power plant that need to be accounted for. I haven't heard of the thorium fuel cycles in India. Off to Google land to find out more.

FYI- Thorium is basically a breedable fuel that produces U-233, a thermal-spectrum (slow neutron) fissionable fuel (similar to U-235 and Pu-239). It's good stuff, the critical mass is a little smaller than U-235 and a little bigger than Pu-239. What's special about thorium though, is it can be "bred" at thermal spectrum, i.e. slow neutrons slowed down by a moderator. That's interesting since most conventional reactors use thermal-spectrum with moderation (water or graphite).

By contrast Uranium breeders depend on neutrons moving around much faster, no moderator inside the reactor, and usually require several orders of magnitude higher neutron density/flux to attain "critical" operation, but the fast neutrons slam U-238's and the resulting U-239 transmutes into Pu-239 which then fissions at fast or slow (thermal) spectrum.
I'm talking about specialized reactors, like what Bill Gates' TerraPower original Traveling Wave Reactor (never found to work well IIRC), and of course the DOE's Integral Fast Reactor (which DID work well) accomplished. Those run at high neutron speed all the time typically using a liquid metal coolant that won't slow down neutrons flying through them. They depend on producing gobloads of Pu-239 and fissioning it, but do require occasional shutdown to reprocess the fuel and get rid of the fission products - which hamper the nuclear chain reaction by acting as neutron poisons (sponges; isotopes that can capture several neutrons without decaying or re-emitting neutrons, acting like the control rods).

Thorium is a neat idea and we've played with it before- the last nuclear fuel load of the Shippingport reactor, the USA's first civilian nuclear power generator, incorporated some thorium fuel rods so they could characterize how well it "breeds" (it does so adequately).

The big thing with Thorium is some folks - Kirk Sorensen was one of the recent pioneers talking this up - are talking about taking a design tested at Oak Ridge Natl. Labs decades back, the Molten Salt reactor, and designing it around a U-233 fissile fuel seed, Thorium "blanket" reactor design, probably with graphite tubes throughout the reactor to act as the moderator. As there's no water involved, the use of graphite shouldn't risk causing a "Chernobyl" style failure mode. Much safer design, basically. IIRC there are still materials issues yet to be resolved with them for the metallic alloys compatible with the hot salt and neutron embrittlement.

The India connection has to do with the fact that they have a lot of Th deposits - from black sand beaches. Good link: https://atomicinsights.com/lftr-in-...-better-than-a-silver-bullet-energy-solution/

One point I keep hearing the Thorium "guys" ramble on about is how safe it is from producing nuclear weapons. I call BS on this because the USA tested U-233 nuclear bombs before.

 

[woodgeek]

Spirilis, you're on fire! LOL.

I agree with you about keeping existing nukes spinning, even doing major overhauls if we need to extend their lives. And that dry casks are FINE for storage of any amount of foreseeable waste.

But I remain a skeptic about breeding and thorium. It has not been demonstrated at a decent scale and price point. It is easy to draw up paper designs, hard to solve the materials and chemical separation problems. If we needed to at any cost? Sure. But there is a current cost to renewables + lithium storage. And I think breeder reactors are well above that.

My idea for cheap nuclear power is fusion, BTW. It turns out that thermonuclear bomb yields can be scaled up much faster than costs. That is, a 1000 MTon bomb costs like 20X as much as a 1 MTon bomb, but it does weigh 100x more. These gigaton devices are massive, and basically a linear assembly of fusible fuel with a 'little' nuclear bomb igniter on one end. So my plan is build one of these, put it in a deep borehole in dry rock in a low population area, and set it off. All the energy goes into heat in the rock, and will remain there for thousands of years. Most of the radioactive reaction products (from the fusion) have short lives. Then drill other boreholes and extract the heat like its high temp (i.e. high quality) geothermal. Done. Fusion power. The bigger you make the device, the cheaper the power gets.

Have a nice day.

 

[spirilis]

I've been spending too much time the past few years reading about nuclear I think :-D

 

[spirilis]

Spirilis, you're on fire! LOL.

I agree with you about keeping existing nukes spinning, even doing major overhauls if we need to extend their lives. And that dry casks are FINE for storage of any amount of foreseeable waste.

But I remain a skeptic about breeding and thorium. It has not been demonstrated at a decent scale and price point. It is easy to draw up paper designs, hard to solve the materials and chemical separation problems. If we needed to at any cost? Sure. But there is a current cost to renewables + lithium storage. And I think breeder reactors are well above that.

My idea for cheap nuclear power is fusion, BTW. It turns out that thermonuclear bomb yields can be scaled up much faster than costs. That is, a 1000 MTon bomb costs like 20X as much as a 1 MTon bomb, but it does weigh 100x more. These gigaton devices are massive, and basically a linear assembly of fusible fuel with a 'little' nuclear bomb igniter on one end. So my plan is build one of these, put it in a deep borehole in dry rock in a low population area, and set it off. All the energy goes into heat in the rock, and will remain there for thousands of years. Most of the radioactive reaction products (from the fusion) have short lives. Then drill other boreholes and extract the heat like its high temp (i.e. high quality) geothermal. Done. Fusion power. The bigger you make the device, the cheaper the power gets.

Have a nice day.

Hah that's a fascinating take on Plowshares or Atoms For Peace. Nuke deep in the earth and set up some geothermal extraction. I wonder how viable that really would be...

 

[spirilis]

But I remain a skeptic about breeding and thorium. It has not been demonstrated at a decent scale and price point. It is easy to draw up paper designs, hard to solve the materials and chemical separation problems. If we needed to at any cost? Sure. But there is a current cost to renewables + lithium storage. And I think breeder reactors are well above that.

I do think this is a valid point and probably going to be for some time. There is one use-case where I've seen the fast-spectrum breeding make sense - the catch being that the cost is high, but the reactor is intended to be used in remote communities (Alaska, Hawaii, etc) where electricity may already be north of 35 cents/kwh.

The Oklo startup's Aurora reactor - https://en.wikipedia.org/wiki/Aurora_nuclear_reactor - a tiny 1.5MW unit - employs fuel rods vaguely similar to the IFR, using a little bit of molten sodium metal inside but most of the cooling happens with heat pipes and supercritical CO2 rankine cycle generators. The use of fast-spectrum self-breeding fuel enables the reactor to run continuously at full power with no refueling or reprocessing for 20 years straight. Like a "nuclear battery" for a remote community. That is very cool IMO. They're moving along pretty well with the regulatory process but they are breaking new ground, as the NRC has never really looked at non-water cooled reactors. They plan to develop larger, more ambitious reactors afterward. But still.... time is not on our side.

For the rest of us in cheaper electricity markets, breeders have an uphill battle to prove. I am curious if Elysium turns up anything groundbreaking though. They're using ordinary sodium chloride salt, in a fast-spectrum breeder reactor where the design can scale from small to large with the main "reactor" portion of the plant stays the same size. Ed Pheil talks about it here- https://www.titansofnuclear.com/experts/EdPheil