Electric Vehicles Shall Prevail

US policy toward electric vehicles abruptly changed in 2025, rapidly winding down the sales incentives which had previously been in place and resulting in blunting what had been healthy growth in the sales of electric vehicles. Rising gasoline prices in 2026 has restored some of that interest, but primarily in used electric vehicles now as supplies of new vehicles have shrunk.

Nonetheless the fact remains: sales of internal combustion vehicles peaked about a decade ago and have remained at best flat since, while electric vehicle sales are growing rapidly. Don't believe it? Sales peaked in the US in 2015, and 2017 world wide.

China is the dominant market for electric vehicle sales, larger than the US and the European Union put together.

 

In 2026 electric vehicles have achieved purchase price parity with gasoline vehicles, and with vastly lower costs per mile. US policy shifts will slow vehicle electrification in the US market, but globally it is accelerating.

EVs have won, all that is left is the circus.

Renewable power generation in Texas

Texas has long been a leader in the deployment of solar and wind, though conflicted about it. Power outages in Texas are often attributed to wind and solar, with the sage observation that the wind doesn't always blow and the sun doesn't always shine. With the Texas grid mostly isolated from the rest of the US, gaps in generation mostly cannot be ameliorated by importing power from elsewhere.

For example a few months after the February 2021 extended winter power outage in Texas, FERC reported that, "Natural gas-fired units represented 58 percent of all generating units experiencing unplanned outages, derates or failures to start. The remaining portion was comprised of wind (27 percent), coal (6 percent), solar (2 percent) and other generation types (7 percent), with four nuclear units making up less than 1 percent."

But during the outage the administration in Texas called out Wind as the primary culprit. Later corrections never achieve the same reach as the initial blame does.

Yet nonetheless, in 2026 Texas has reached several big milestones in its conversion to renewable power. This is amazing, commendable news.

---

Recent Milestone #1: Solar Surpasses Coal

In a report published on May 13, 2026, the US Energy Information Administration said, "In our most recent Short-Term Energy Outlook, we forecast that annual electric power generation from utility-scale solar will surpass that from coal for the first time in 2026 within the electricity grid that covers most of Texas."

So far, solar is not displacing coal so much as it is shutting out coal from new capacity being built. This is almost guaranteed at this point: not only does solar have no fuel cost, it is cheaper per Megawatt to build. It is difficult for coal to compete with a technology which is both less expensive to operate and less expensive to deploy.

Unfortunately though, as shown in the chart, electricity from methane (I decline to use the term natural gas) is also growing strongly.

---

Recent Milestone #2: Texas Battery Capacity Surpassed California

In the longer term, the most pivotal technology deployment isn't the megawatts of capacity of wind or solar generation, it is the megawatt-hours of battery storage deployed. Grid-scale batteries are how renewable power can address "the sun doesn't always shine." argument often advanced in favor of fossil generation. For example the Heritage Foundation in April 2025 called for policies to discourage deployment of renewables and battery storage in favor of fossil sources.

At some point in the last several months, Texas passed California in total battery capacity deployed. This is a big milestone, and a wonderful accomplishment.

Some day soon I hope to write about renewable generation in Texas passing Methane. It is, at this point, inevitable.

Twenty Eight Kilowatts

Our 2016 Kia Soul EV is able to charge at about 7 KWatts using our charger at home, a level2 EV charger built into the SolarEdge solar inverter. It has a CCS plug.

 

Today I charged the car while out and about, using the other charging port on the Kia: CHAdeMO, a DC high power charging standard. I should put "high power" in scare quotes as it is quite pedestrian by today's standards, it is a standard from years ago.

• CHAdeMO managed to deliver 28 KWatts to this car.
• Twenty. Eight. Kilowatts.
• Four times more power than the CCS charger at home.

It charged from 62% to 90% in 22 minutes, when it began to slow as the battery approached full. I unplugged it at 90% and continued on my merry way.

Electric short-haul vehicles

The market for forklifts and similar lift trucks electrified early. The market for Class I lift trucks intended mainly for use inside warehouses crossed 50% electric in approximately 2010. In indoor environments there is a substantial advantage for a vehicle which produces no exhaust fumes. The early models used lead-acid batteries, gradually shifting to Lithium chemistries as batteries aimed at electric vehicles improved.

More recently, other classes of forklift which handle heavier loads and outdoor use have been electrifying, as batteries have reached a capacity to operate all day without charging and electricity is less expensive than the equivalent propane for internal combustion forklifts.

Electric lift trucks have several inherent advantages over propane:

• Efficiency: electric vehicles excel with frequent starts and stops, which a forklift spends all day doing.
• Regenerative lowering: it takes energy to lift a heavy load, but regenerative braking techniques can recover energy while lowering a heavy load. Anything lifted up will (eventually) be lowered back down, over time the fleet of forklifts will recover a good portion of the energy spent lifting.

Short haul trucking

Drayage trucks, heavy vehicles intended for short distance duty such as to and from transport hubs, are at the start of their electrification process now. They also have several inherent advantages:

• Idling: even with a decade of effort in schedule optimization and just-in-time arrival, such vehicles spend a substantial amount of time waiting for loading and unloading. Internal combustion vehicles consume fuel while idling, electric trucks do not.
• Emissions: air quality near ports and transportation hubs has been a concern for decades, and have resulted in ever more strict limits on emissions in the vicinity of the port. A zero emission electric drayage vehicle more easily meets these requirements.

Vehicle to Grid?

To me, one of the interesting potential developments for electric vehicles is to take advantage of the battery capacity available in fleets of electric vehicles with one common owner and fairly stable usage patterns. Warehouses and ports mostly do not operate at 100% capacity 24/7. Noise ordinances and the economics of three shift work means they will often not be fully staffed overnight. There will be hours where some of the equipment is plugged into chargers and mostly sitting idle.

These are the same hours where solar production is not available. Might the fleet owner be able to make some amount of revenue while the equipment sits idle? The equipment does need to end the night with a mostly full battery for the first shift's work, but there may be an opportunity to charge while power is cheap and, knowing usage patterns in advance, participate in virtual power plants to bid into the day-ahead market.

School buses remain the best example of Fleet-V2G potential. They can charge after dropping off children after school, at a time when solar power is still generally feeding the grid. They will sit all night, and can supply some amount of power in the late evening hours.

Virtual Power Plants and the California Grid

The California Independent System Operator (CAISO) has published a report regarding electricity generation in the summer of 2025, which includes data through the end of September 2025. They provide a nifty interactive chart at that link, and the underlying data is easy to find in the source of the web page. I've included the data for July and September at the bottom of this post, as we're going to dive into those details in a moment.

Below is the graph of the sources of energy on the California electrical grid in September of 2025, the period ending about two weeks ago at the time of this writing. Importantly, the Solar resources shown are only utility connected solar arrays. Rooftop solar is behind the meter, reducing demand for electricity rather than adding to the supply measured here. There is approximately another 19 Gigawatts of solar capacity installed behind the meter.

---

This is the same data, shown as a percentage of the total. Because total electricity consumption varies throughout the day, the contribution of baseload sources like Methane-fired generators varies as a percentage of the total even though their output is constant.

A few observations:

• The contribution of wind power is smaller than I expected. I may get a skewed view of the prevalence of wind generation in California as I pass through the Altamont Pass wind farm regularly, but I expected wind to be a larger percentage.
• It is too small to be visible in the graph, but Solar power never actually drops to zero. It continues to supply about 4 Megawatts all night. I believe this might be Ivanpah, a solar thermal generation plant in the desert, which continues generating power from stored heat even after the sun has set.

---

I tried to incorporate behind-the-meter solar into a similar graph, below. This is a crude estimate:

• In 2024, 19 Gigawatts of rooftop solar was installed versus 21 GWatts of utility-scale. I made the assumption that the production from rooftop solar would be approximately 19/21 of the utility number.
• Prior studies show that rooftop solar is not installed in ideal locations nor properly angled toward the sun, I made the additional assumption that it would be 80% as productive as utility solar.

The result hews considerably closer to the 67% renewable result announced by the state last year.

---

Virtual Power Plants

I have combined the categories which CAISO reported separately as Batteries and Demand Response, because until 4/2024 Virtual Power Plants formed via aggregation of residential batteries like Powerwalls were contracted as Demand Response. Only utility-scale battery installations like Megapacks were accounted for as Batteries. Splitting into two categories obscures and substantially minimizes the true contribution of battery power to the grid.

We participate in a Virtual Power Plant in northern California, allocating about half of the 27 KWh of capacity installed at the house. For a number of days this summer in the late afternoon and early evening, the house supplied about 6 kilowatts of power back to the grid. It is quite smooth, we don't even notice unless we look at the app to see what it is doing.

---

Raw data from the CAISO report web page.

"September": [
  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1392.000662, 1319.840318,            // Demand Response
   1239.827827, 1177.653666, 1023.942241, 287.33, 0, 0],                                // Demand Response

  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2264.012689, 6312.981683,            // Battery Storage
   10219.53932, 9740.80126, 7382.450552, 5789.963039, 2869.891625, 0],                  // Battery Storage

  [11665, 11665, 11665, 11665, 11665, 11665, 11665, 11665, 11665, 11665, 11665,         // Imports
   11665, 11665, 11665, 11665, 11665, 5500, 5500, 5500, 5500, 5500, 5500, 5500,         // Imports
   11665],                                                                              // Imports

  [742.4458, 725.114696, 719.58188, 591.859728, 356.992952, 328.293048, 330.9584,       // Wind
   323.6145657, 277.992424, 205.762648, 221.590544, 199.863504, 216.38616,              // Wind
   311.644072, 316.5584821, 324.313968, 396.1921449, 535.622064, 785.470392,            // Wind
   905.120696, 1081.627632, 1164.064064, 1096.58392, 1058.725816],                      // Wind

  [3.7224, 3.7224, 3.95928, 4.43304, 5.02524, 7.1064, 155.30868, 3417.63696,            // Solar
   8622.78732, 11304.55656, 12390.19452, 12923.98668, 13035.8448, 12592.86717,          // Solar
   12166.76592, 11529.86232, 8779.161639, 5405.92308, 1148.5296, 12.79152,              // Solar
   5.34672, 4.01004, 3.8916, 3.84084],                                                  // Solar

  [1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861,        // Other renewables
   1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861,        // Other renewables
   1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861,        // Other renewables
   1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861, 1755.632861],       // Other renewables

  [1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682,        // Other
   1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682, 1682],                   // Other

  [7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014,        // Hydro
   7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014, 7014],                   // Hydro

  [2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280,        // Nuclear
   2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280, 2280],                   // Nuclear

  [26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188,         // Methane
   26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188, 26188,         // Methane
   26188, 26188],                                                                       // Methane

  ...
]

One Year of Electrified Caltrain

In November 2020 California voters passed Measure RR to fund electrification of Caltrain down the San Francisco Peninsula. After several years of construction, the new electrified trains entered service on September 21, 2024: exactly one year ago at the time of this writing.

The electrification measure was an audacious plan for Caltrain to recover from the Covid-driven disruption in travel patterns by radically improving the service. The electrified fleet would provide more frequent service because the new engines would accelerate and decelerate far more strongly than the diesel locomotives did. The rolling stock would additionally be refreshed with new passenger cars.

So: did it work? Based on ridership data from the last year, to me it certainly appears so.

The large drop in ridership in 2020 is due to Covid. I tried to show the slope by drawing red lines in the few years between Covid and electrification, and between electrification and now. The rate of increase in ridership changed markedly for the better in almost exactly 9/2024 when the lines were electrified.

Ridership has not yet returned to pre-Covid levels, but now appears to be on track to do so if trends continue.

Iceland Carbfix Tour

In July 2025 we took a tour of the Geothermal Exhibition at Hellisheiðarvirkjun in Iceland, all about Geothermal power. My spouse is a Professional Geologist, for whom this was an especially interesting tour.

We took a slightly more extensive version of the tour which included the CarbFix plant, a carbon capture and sequestration project where carbon dioxide is injected deep underground to mineralize.

Near the Carbfix injection site is the Climeworks Mammoth plant, a direct air carbon capture facility. We didn't get to go inside, we could only see it from a distance.

Steam pipes from the geothermal vents back to the power plant.

Steam pipes within the power plant.

Turbine within the power plant.

Climeworks Direct Air Capture facility.

Carbfix H2S+CO2 pumping facility.

Carbfix H2S+CO2 meters.

Personal View of NYC Congestion Pricing

In the 2010s I managed an engineering organization with teams in California and New York. I travelled to NYC a number of times, typically staying near Chelsea Market.

The Maritime on 16th street was my usual lodging, next to Google's NYC office and with the 14th Ave subway station nearby. I recall the blaring of car horns being ever-present, continuing late into the night.

We brought the whole family to New York City in July, the first time I have been there in almost 10 years. We stayed in Manhattan in the Financial District, and went for pizza very near the Google building. The streets were very clear, nowhere near the level of traffic I remember.

(view from the Empire State Building)

(on the way to pizza)

In January 2025, New York City implemented a congestion pricing mechanism to increase tolls for cars entering the city. It had an almost immediate impact in reducing traffic:

• February, 1 million cars reduced
• March, large increase in public transit utilization
• April, large reduction in congestion
• Manhattan foot traffic finally tops pre-pandemic times

The Federal government, always eager to increase fossil fuel consumption, has revoked the needed authorizations and demanded that NYC end the congestion pricing mechanism. The two parties will present their arguments in court in October 2025.

I hope congestion pricing stays. The city is better for having it in place.

National Climate Assessment team disbanded

By Ed Hawkins, climate scientist.
CarlinMack created this version.

Three weeks ago contracts for the National Climate Assessment were defunded and work stopped.

Today the 400 people working on it were disbanded.

Production of the report is funded and mandated by law. Presumably in 2028, AI will write something.

Finding a Role in Climate

Climate Week is drawing to an end, not yet done but one can see the close approaching.

I have spent a bit over a year now on my own, doing some consulting work while looking for longer-term opportunities but also taking downtime away from the industry. I’m very motivated to work on climate, building on earlier efforts:

• two years as a Senior Fellow at Project Drawdown
• several years coaching climate community members starting their careers
• Cohort 5 of the ClimateBase Fellowship
• all of that coming after several decades in the Tech industry, at three startups (Dominet Systems, ConSentry Networks, Tailscale Inc) and two large companies (Sun Microsystems, Google). I held roles from ASIC designer to software manager to VP of Engineering.

I’m starting to focus again on finding the right long term opportunity, not just consulting. What I’d request of those whom I’ve worked with or had the pleasure to meet along the way is introductions at the right stage, for roles with:

• a focus on climate as the primary mission. Energy is the best match for my skillset, but I believe that land use and agricultural tech need more effort and I have relevant experience with satellite imagery.
• a position which is substantially leadership, from Director at a large company to Founder / Founding Engineer or VP at an earlier stage. I can help hire, evolve organizations, and build a product.
• a technical component which is not zero. Managers should manage, but I believe managers who completely lose touch with the reality of the engineering work become less effective as leaders. I would seek an opportunity where there would be suitable opportunities to contribute technically, and believe it is important that the team see those contributions.
• an organization with a European connection. We enjoy Europe, have travelled in Germany several times, and have substantial family connections there.

These sorts of opportunities are mostly not posted publicly. I have responded to a few public postings over the past year, that is not an effective way to proceed. I’d ask for warm introductions you may be aware of, early in the process, perhaps when founders are talking about a new venture or considering a new project which needs leadership.

Thank you so much for any connections you can provide.

SF Climate Week Opening Keynote

As I did last year, I took the train to get to SF Climate Week. In this area that means taking Caltrain up the Peninsula before switching to the Bay Area Rapid Transit (BART) to the Embarcadero, then walking to Climate Week at the Exploratorium.

Both of those train systems have been substantially improved since last year:

• Caltrain completed a years-long electrification project, replacing all of the diesel trains.
• BART finished deployment of a new generation of cars, retiring all of the 25 year old rolling stock.

From this one might infer a renaissance of mass transit deployment in urban areas... but one would be wrong. Indeed, in nearly every area of climate action where the Inflation Reduction Act had spurred progress, the new administration of the last three months has attempted to roll it all back.

---

Former Vice President Al Gore presented the opening keynote speech, fiery and powerful.

SF Climate Week 2025

This is SF Climate Week! The opening keynote with former Vice President Al Gore, long-time Speaker of the House Nancy Pelosi, & SF Mayor Daniel Lurie is this afternoon at The Exploratorium in San Francisco.

I'll be in SF this week as a volunteer helping keep things running, hope to see you there.

Coal Mining Policies

New coal policies are invariably announced in front of a group of workers wearing hard hats with lights affixed, and often in Pennsylvania for good measure. One might assume the mining profession is a huge economic force and under constant threat which must be fended off to preserve families and livelihoods.

As a profession, coal mining employs about 40,000 people in the US.

Source: FRED (Federal Reserve Economic Data).

---

Construction Management requires similar levels of education and experience and according to employment statistics enjoys a similar pay scale. There are 10x to 20x more Contruction Managers in the US.

Source: FRED (Federal Reserve Economic Data).

---

Coal policies are not driven by concern for workers. Coal policies are driven by concern for fossil fuel profits, which have only been made possible by externalizing the cost of the damage to human health and acceleration of global warming.

Electrified USPS Fleet

In its initial proposal for a refreshed fleet of delivery trucks in 2022, the US Postal Service and its contractor Oshkosh Defense proposal listed a combined vehicle + cargo weight of 8501 pounds. This is a remarkably specific number, exactly one pound heavier than emissions rules would constrain. Were the vehicle 1 pound lighter it would be required to be considerably cleaner — and likely electric, to meet those requirements.

With more effort, and additional $3 billion in funding from the Biden administration, by the following year this proposal was revised and proposed that most of the fleet be electric. A gasoline version would be used for long routes in rural areas without sufficient charging infrastructure. Initial electric vehicles were delivered in September of 2024, apparently to excellent reviews by postal workers using them.

Oshkosh has delivered only 100 electric trucks thus far, production was expected to ramp up from the initial deliveries a few months ago. The Oshkosh Defense CEO now says they'd be happy deliver the rest of the order as solely gasoline vehicles. It is difficult to see this in any positive way.

A Tale of Two Crises: Y2K and O₃

"Y2K," or 1/1/2000, was 25 years ago today. Dire predictions of how bad the Y2K Bug might be, with the failure of computing systems leading to widespread disruption, did not manifest. NPR chose what has become the dominant framing, a cynical take that Y2K was overblown and a delusional over-reaction. A nothingburger.

It is easy to see why one might believe this. Since 1/1/2000 we have lived through a seemingly neverending series of grift bubbles: the dot-com bust, subprime mortgages, cryptocurrencies, etc. It is easy to assume that Y2K was surely similar, a cynical hype cycle enabling some kind of profiteering.

 

Y2K Spending

To be clear: money was spent. Y2K remediation wasn't just some developers combing through COBOL, as is often depicted. It was more cost effective to simply replace a lot of computing systems from the 1970s and 1980s with something more modern.

Development of the modern Internet was accelerated by Y2K spending. The new systems were usually Windows Server or some form of Unix, with TCP/IP and robust networking built in. Businesses in many industries, their upgrade cycle moved up to meet Y2K demands, could make their service available on the Internet years earlier than they otherwise would have. I think we can even see it in the oft-cited productivity gains of the late 1990s.

Yet all of that effort and all of that spending wasn't in service to a fake grift. It worked. We fixed it. We actually fixed it. It is perhaps difficult to comprehend from our vantage point in 2025, but we faced a large problem and we solved it with a correspondingly large effort.

 

The Ozone Hole

We will digress for a moment to a different topic which might not seem related, but is: the Ozone Hole of the 1970s and 1980s. It is another formerly big problem which seems to have gone away — not entirely solved as the hole is still there, but the ozone layer is recovering. A common reaction is to question whether it was overhyped.

A lot of people put in a lot of effort for a lot of years replacing chemicals which caused most of the damage to the ozone. Money and political capital were spent: every nation on Earth ratified the Montreal Protocol mandating the phasing out of CFC manufacture.

It worked. We fixed it.

 

Why Not Now

The important discussion is not whether large challenges of the past were somehow not large challenges. The important discussion is why we have been unable to rise to similar challenges now.

• Climate change is everywhere but we're still debating whether it will be so bad and equivocating on what to do.
• Covid-19 should have led to HVAC retrofits to improve indoor air quality but it instead empowered antivaxers to rip people's masks off.

Within living memory we have risen to challenges requiring the whole world to cooperate, a feat which seems impossible now.

1. Then, the forces uniting us had the most effective means of coordination and of broadcasting their message: the UN and governmental coordination, and a mass media which created a shared reality.
2. Today, the forces dividing us have the most effective means of coordination and broadcasting their message: online social media and an entirely separate infosphere.

Wind Turbines and Bird Deaths

The practice of painting one blade of a wind turbine in order to increase its visibility to birds is becoming commonplace. Avian brains are not wired to detect such enormous structures as a threat, but increasing the contrast and making the movement more visible helps them see it as something dangerous to be avoided.

To be clear: this is a good thing. Harming birds, even unintentionally and without malice, is bad and taking steps to avoid it is good.

But this is another example of bad faith attacks, launched carelessly and repeated endlessly, turning into a multi-year effort by the clean energy industry to respond to. The number of birds involved was tiny, less than 0.02% of bird deaths. A study in South Africa found 848 birds killed, over the course of four years, across 20 wind energy sites. This is not a large number.

The next set of allegations about wind power are harder to refute because they are completely fabricated: that wind turbines cause cancer, or that offshore turbines kill whales.

The people making this stuff up do not want to challenge clean energy to be better. They are not interested in seeing improvement to address whatever harm they cite. They are only interested in deterring the deployment of clean energy because money from fossil fuels funds their movement. It is all in bad faith, and the disproportionate cost falls on clean energy to address.

Caltrain Electrification

Caltrain is a large publicly owned train service in California. For the past few years Caltrain has been installing electrical lines above the tracks running up the San Francisco peninsula. Train electrification was proposed thirty years ago, began work seven years ago and completed in September 2024 with the retirement of diesel and replacement by electric engines.

The electric train engines have considerably better acceleration, important given the number of stops along the peninsula, and can complete the journey from San Jose to San Francisco in 30% less time than the diesel engines could. This allows trains to be scheduled more frequently.

 

Induced Demand

Studies have previously shown that service frequency has a significant impact on ridership, and indeed, Caltrain ridership is up 54%. On weekends it is up even more, by more than double.

We talk about the Braess paradox, where adding a lane to a busy roadway will further increase traffic and result in more congestion not less. Building massive highways doesn't resolve traffic congestion for car travel.

Yet increased ridership does not trigger the Braess paradox in train networks. Inherent in the paradox is that moving entities choose their own route, and will optimize for their individual outcome even at the expense of the whole. The trains are not making individual optimization choices, they just run faster.

 

Electrified Trains are Better Neighbors

As a family we enjoy a restaurant right next to the San Carlos Caltrain station, one with outdoor seating. The most recent trip there was a revelation: the electric train is quiet while getting underway, and with no diesel fumes. Electric trains fit into the cityscape more cleanly, quietly whisking passengers on their way.

 

Emergency vs Loadshifting Batteries

Our house has 8 kilowatts of solar generation and 27 kilowatt-hours of battery storage, and is on a time-of-day electricity plan where power is cheaper in the early part of the day and more expensive from 3pm until midnight. We use 70% of the battery capacity to power the house during the peak hours each day, reserving 30% in case of outage.

This does work, but one battery is filling two roles with requirements at least somewhat in conflict.

• For load shifting we would prefer to use more of the battery to reduce our energy bill.
• If there is a power outage, we'd prefer to have more of the battery held in reserve than 30%.

Yet the batteries wired in to the house are not the only substantial energy storage on site. We also have an electric vehicle, with as much battery capacity as the house. A more modern EV than ours would have even more battery capacity.

 

Vehicle to Grid

That the substantial battery power in EVs could be useful in grid emergencies has long been recognized. Vehicle-to-Grid (V2G) is an idea to make the charging port of the EV bidirectional: charging the EV from the grid as normal, but also able to supply to the grid by draining the EV. In very large numbers, EVs could provide enough power to stabilize grid operation while still having enough power to be used for transportation.

V2G has been slow to catch on, owing mainly to the number of parties involved. Because it entails connecting a new generation source to the grid, it must follow similar processes and permits as a rooftop solar installation. The timeline of a solar install is long: it can be months with roofers / installers / electricians in several waves, and having time enough to apply for the needed permits and interconnection agreement.

Yet when people buy an EV they need a way to charge it almost immediately. Waiting to install an EV charger isn't realistic, so it will be installed to only draw from the grid not supply back to the grid. Designing V2G capabilities into an EV charger in hopes the homeowner will followup with additional permitting? Most consumers wouldn't ever do so.

 

Vehicle to Grid Vehicle to LOAD

Vehicle-to-Load (V2L) is an idea with fewer stakeholders involved: allow the vehicle to supply power to locally connected loads, most commonly by having electrical outlets and an inverter built into the car. This is useful in many situations, like an EV at a job site powering tools or while camping or picnicking.

V2L also allows the EV's stored power to be used at home during emergencies, albeit somewhat awkwardly. It doesn't power the whole house, it can power appliances which are unplugged from the house and plugged into an extension cord from the vehicle. Keeping cell phones charged, running medical equipment, or powering freezers full of stored food is quite feasible. Running central air conditioning is not.

This concept works well for emergencies in that it can power essential needs from a very large battery, especially because the battery is mobile and could go somewhere to charge itself if needed. Natural disasters in the last several years have demonstrated this, notably Hurricane Helene in the US Southeast just last week. EVs are helping supply power in damaged areas.

California, my home state, has a large enough market for automobiles that it has often been able to influence the auto industry throughout the US such as via fuel efficiency standards. A law being considered would allow the state to mandate Vehicle-to-Load capabilties for vehicles sold as of some future date.

 

Transfer switch

We'd very much want our next EV to have this capability. Our use of the batteries built into the home would change to entirely load shifting to further reduce our demand on the grid during peak hours. During a power outage of appreciable length, we'd rely on the much larger batteries in the vehicle to keep food frozen and cell phones charged.

For this to work we'd need to plug appliances into an extension cord from the vehicle, and some of them are hard-wired into the house. The traditional way to handle this is a transfer switch, intended for a generator. This is expensive, and difficult to retrofit. There needs to be an easier way.

Powerwalls and Time Based Controls

We are on a Time-of-Day electricity plan, the PG&E EV2 plan. Our cost for electricity is:

		Summer	Winter
Off-Peak	12am - 3pm	$0.31/kWh	$0.31/kWh
Mid-Peak	3pm - 4pm	$0.51/kWh	$0.48/kWh
Peak	4pm - 9pm	$0.62/kWh	$0.49/kWh
Mid-Peak	9pm - 12am	$0.51/kWh	$0.48/kWh

We are incentivized to keep the power-hungry activities like charging the electric vehicle to the daylight hours where solar is plentiful or overnight where demand is low. Unfortunately however, there are several confounding factors to our energy use which make things more complicated.

1. The house is on a hill, the crest of which blocks direct sunlight for longer and longer periods each winter day.
2. We have two Powerwalls, able to store 27 kWh to supply the house at other times.
3. BUT, when we installed the powerwalls in 2019 we claimed the Residential Clean Energy tax credit which requires the batteries be charged using only solar power for five years.

Combining all of these things, we ended up with an unfortunate confluence in the winter months when the hill allows only a few hours of direct sunlight: there is not enough excess solar to charge the batteries while also powering the house.

We would head into the higher priced times of day with little ability to time shift stored solar production. The batteries were never able to charge. Our power bills rose substantially, calling into question why we paid for this stuff in the first place.

The Tesla app has a "Time Based Control" mode, where it takes the time of day and rate plan into account. However its main focus is in exporting solar production during high value hours by running the home from battery. Lacking sufficient production to charge the battery, this resulted in poor outcomes with Time Based Control. I didn't look at it again for the next few years.

 

Hacking Around It

Instead we've come up with techniques to get things working acceptably:

1. On winter mornings set the Powerwall to reserve 100% of its capacity for power outages. All solar production during the day charges the batteries, trying to reach 100%. The home's needs are met from the grid during off-peak hours.
2. When peak hours start, change the Powerwall to 30%. It then discharges to power the home, allowing the evening load to be partially met using stored solar power.
3. Do this every day. Change the Powerwall setting in morning and afternoon, every day, all winter.

I of course wrote software to automate this, but Tesla has only recently decided to offer an actual API to control Powerwalls. For the first few years I was instead using authentication mechanisms and APIs which a community on GitHub would reverse engineer, and which Tesla kept deliberately breaking. I had to watch for errors from my software or, maddeningly, when it would run without error but Tesla ignored its commands.

 

Changing the Game

At the beginning of this month, our five years was finally reached. I set the Powerwall to be able to charge itself using grid power. This didn't immediately change much behavior, until I toggled it to Time Based Control again.

Now, suddenly, things are much improved. At 3pm every day the Powerwall begins supplying the house energy demand, and all remaining solar production is sent to the grid. At midnight when rates drop, the Powerwall charges itself to 100% using grid power, to be ready for the next day.

This is already better than the system I had cobbled together:

• There were many days when the total solar production could not fully charge the batteries. Now, no matter what, the batteries are 100% charged every day.
• We are on the Net Energy Metering plan from 2019. Sending solar to the grid offsets our use at other times, so long as the house can be powered from battery.

 

Futures

I do want to change the current behavior in one way: instead of charging from the grid overnight, I'd prefer the Powerwall try to charge from excess solar. In summer it will usually be able to do so, and in winter it can try and then start charging itself mid-morning from the grid if it isn't going to make it to full. I'm looking into the Fleet API which Tesla published this year for what might be possible.

However, fundamentally, this stuff needs to be easier for the homeowner. I've been writing custom software and manually intervening for years, just to get a decent result out of it. I should not need to do that, after having paid so much for the system install.

One small provision in the Inflation Reduction Act was to remove the five year solar charging mandate for batteries installed after its passage in 2023; the goal is to incentivize more batteries on the grid. With freedom to charge the battery, it should be able to figure out how and when to charge itself. The behavior of the system over time should inform future operation, deciding when to charge from the grid and when to trust that solar power will provide. Next-day solar forecasts can inform this decisionmaking.

Surplus Energy Response

The phrase "too cheap to meter" entered the energy discourse in the 20th century, referring to the potential of nuclear power. Though originally coined to refer to fusion power, the label instead became associated with all nuclear energy. Seventy years later fusion is not here yet, and fission power has been a solid source of baseload power but could never be described as cheap.

The idea of energy too cheap to meter is a compelling one, we just had to wait for an entirely different technology to deliver it: solar photovoltaic. Panels installed on residences and commercial buildings are typically installed "behind the meter," where it directly supplies the energy demand of the building. The proof that solar power is too cheap to meter is that is not, in fact, metered.

Solar deployment has grown incredibly rapidly, faster than the distribution grid would be ready to accept it all. Deployment behind the meter has been essential because the grid in front of the meter hasn't been able to deploy new capacity so rapidly, and solar deployment continues to accelerate. A post by Ben James argues that solar energy can be deployed so inexpensively that using it completely off the grid, for economic activities which can be economical with free energy so long as it can handle being run intermittently only when the sun is shining, is compelling.

• hydrogen production, via electrolysis of water
• fertilizer production, producing ammonia via air capture and energy
• kerosene production, also via air capture
• ... and other chemical processes made possible by prolific free energy

 

Surplus Energy Response

The electric grid has a notion of Demand Response, when there is heavy demand which stresses available generation — for example, by air conditioning on a hot afternoon. We have reached the point where we also have the opposite situation: we need a surplus response. Many builings now produce substantial excess behind-the-meter power during the day, so much so that the grid cannot absorb it all. We need our buildings to become smarter about putting the excess energy to useful work:

• pre-heat or pre-cool HVAC, somewhat overshooting the temperature setpoint while energy is free
• store hotter water, with a smart water system to mix scalding with cold to get the desired water temperature
• charge electric vehicles for free, with knowledge of when the vehicle is likely to be needed
• charge up batteries in appliances throughout the building, allowing high peak load appliances to be installed in buildings not originally built for them

Yet we can take it even further. Limitless free energy, albeit at limited time ranges within a day, allows us to make choices we would never have otherwise considered.

• heat a pool or hot-tub to be ready for impromptu human use at any time during summer months
• run heat pumps in a sunroom, open to fresh air yet maintained at a comfortable temperature
• indoor urban hydroponics, pre-engineered gardens which are never too hot nor cold and provide generous fresh produce

In building enthusiasm for the energy transition, providing services which seem impossibly luxurious yet are provided entirely by surplus energy would be a compelling outcome.