Featured Post

This is the Kodak Moment for the Auto Industry

Plug-In Drivers Not Missin' the Piston Electric vehicles are here to stay. Their market acceptance is currently small but growing...

Saturday, April 24, 2021

40 Thousand Miles In A Tesla Model X

My daily driver is a 2016 Tesla Model X 90D. Although the term 'daily driver' doesn't mean as much as it used to since I work from home nowadays #pandemic. I've done annual reviews of the vehicle each year since I purchased it, you can see them here: 1, 2, 3, 4. 

Painted Hills biking adventure with our Model X

This is not an annual review, it's just a quick post, not for an annual milestone, but for an actual mileage milestone. The odometer just rolled over 40 thousand miles. A record by no means; but these miles have been fun. We drove to Grants Pass in the winter, the Painted Hills in the desert of eastern Oregon in the summer, brought home an Xmas tree on the roof, we pulled our camper, went to Comic-Con in San Diego, to Crater Lake, the Oregon Caves, through a wildlife safari, to a butterfly pavilion, and to Thor's Well on the Oregon coast. We used biohazard mode during the Oregon fires. And it was all fueled by free Superchargers and the solar panels on our roof.

I've continued to track the battery degradation. As you can see in the graph below, it has started to level off. 

The other degradation charts that I've shared were time-based. This one is mileage-based. Here's to 40 thousand more miles of fun before I upgrade to a 2025 Model X with full self-driving

Disclosure: I'm long TSLA

Saturday, April 17, 2021

Moore's Law, Wright's Law, Swanson's Law, & Jevons Paradox: How these axioms will impact EVs and our future energy system

Original scaling predictions by Gordon Moore
This became the basis for Moore's Law

In this post, we'll look at a few "laws*" that have (and will continue to) transform our world. We'll look at how these laws (or more accurately, axioms) came about, how they've impacted society, and how these laws apply to Electric Vehicles (EVs) --  since that's kind of what we're about here. 

Moore's Law 

Moore's Law is the most well-known axiom we'll examine. It's named after Gordon Moore, the co-founder of Fairchild Semiconductor and Intel, as well as the former CEO of the latter.

In 1965, Moore observed that the number of transistors in integrated circuits had been doubling about every two years. Moore projected that this exponential growth would continue for at least the next decade.

Now, 5+ decades later, the trend has continued and has had huge implications on the cost of computing power. This exponential growth means that the phone you likely have in your pocket is more powerful than the computers used to land the first humans on Luna. Compared to 1965 (when Moore made the observation), computers are now everywhere and have changed the way we live. Moore's Law has resulted in lower cost, lower power, and better performance computing. It has allowed chips to be used in nearly everything we own from toys and cars to appliances. This has been a springboard for innovation: the internet, smartphones, the cloud, console & computer gaming, and even the current streaming wars are all the result of Moore's law.

How does this apply to EVs?

With all of their high-tech features, Tesla's vehicles have been called computers on wheels. The impact of computing on personal transportation is most obvious in a tech-forward car like a Tesla, but computer chips are used throughout all modern vehicles for functions like antilock breaks and airbag deployment. This became painfully apparent in early 2020 when several auto manufacturers had to shut down production due to a worldwide chip shortage.

Looking forward, the increase in computing performance and cost reduction will enable better in-car entertainment, communications, and (eventually) autonomous driving.

Wright’s Law

Next, we'll look at Wright's Law. Theodore Paul Wright, also known as T. P. Wright, was a U.S. aeronautical engineer. He had a storied career at Curtiss-Wright Aeroplane Company, where he started out as a Naval Aircraft Inspector and moved up to Chief Engineer. From there Wright became a member of the National Defense Advisory Committee under President Franklin D. Roosevelt. In this administration, he had several roles and titles primarily focused on the production of military aircraft.

It was during this time that Wright determined that for every doubling of production, the labor requirement per airplane was reduced by 10-15%. In 1936, he detailed his findings in a paper titled “Factors Affecting the Costs of Airplanes.” The paper described that the cost of each unit produced decreases as more units are produced.

This is the "economy of scale" axiom. As you make more of something, you can expect the per-item cost of manufacturing to drop. This follows that the manufacturing equipment would have higher utilization; you'd be able to buy materials in bulk and receive better pricing from suppliers, etc. As production grows, optimizations for scalability allow manufacturing costs to be further amortized.  Thereby, creating a positive feedback loop where more production leads to lower labor and supplies costs, allowing for more production.

As production costs drop, prices can be reduced. As prices are reduced, it becomes easier to displace older (stagnated) technologies. Additionally, with lower prices, new uses are found, thereby creating growth opportunities (TAM expansion), further allowing production to expand, further reducing costs.

One of the lesser-known aspects of Wright's paper is that it lays out a “we learn by doing” principle. The lesson is that it's better to go forward with an 80% plan rather than waiting for a 100% complete plan. Until you start, you don't know all the issues you'll encounter, so rather planning out the final 20%, you could instead start and find out where you really need to focus your attention. This paper was published 64 years before the Agile Manifesto, yet they share this start-early core principle. You cannot know the unknown unknowns without 'doing'. “Ready, Fire, Aim” is a similar concept.

How does this apply to EVs?

Batteries are currently the largest cost factor for EVs. The battery pack in a vehicle is a significant factor in its range, performance, and cost. Luckily, batteries have been following Wright's Law. Production has been steadily increasing and costs have been on a steady decline since the start of this generation of EVs began a decade ago. EVs started on the high end of the price range. To date, affordable EVs have been ranged limited (sub 100 miles). By mid-decade, this will change. EVs will be as affordable as similarly equipped (and ranged) gas cars. By 2030, EVs will be notably cheaper than their gas-powered cousins.

Swanson's Law

Swanson's Law is named after Richard Swanson, the founder of SunPower Corporation, a solar panel manufacturer. Swanson regularly gave talks and wrote papers and articles that showed the cost decline trend of solar photovoltaics (PV). Specifically, his data showed the price of PV modules dropped ~20% for every doubling of production.

This is an industry-specific application of Wright's Law unit cost curve or economies of scale.

How will this impact EVs? 

As solar prices decrease, it will be more affordable to cover your rooftop in solar and to have solar canopies over parking lots. These solar canopies both provide shade and generate electricity. It's incredibly rewarding to know that your car can be powered by the sunlight hitting your roof. Each 1kW of solar on your roof is enough to power an EV for about 4000 miles.

Solar production scales well with electricity needs in many parts of the world. Solar generates electricity during the day when we tend to be more active and grid demand is higher. PV also generates more electricity in the summer when the air conditioning units are running. 

Combine the price reductions for batteries (see Wright's Law above) with the price reduction of solar panels and you soon have the ability to have a home battery and industrial-scale battery backup. This turns energy "digital." Today's energy grid has to match supply and demand in real-time. This makes for a fragile system. To compensate for this, utilities have to over-provision and/or have wasteful, polluting spinning reserves to avoid rolling blackouts or other outages.

Batteries, on the other hand, can respond in milliseconds and can provide hours of backup. If there's a short-term increase in demand, batteries can easily cover it. If it's a longer-term issue, the batteries give the grid operators the time they need to ramp up energy generation, bring more systems online, or recruit supply from neighboring systems. This makes for a far more robust energy grid.

For us, we use our home battery system to time-shift our solar production from mid-day to peak hours. This reduces the stress on the grid and reduces our electricity bill. Additionally, we charge our EVs overnight when there is surplus energy on the grid (this is when the wind mix is typically at its highest) and electricity prices are at their lowest.

Jevons' Paradox

The Jevons' paradox is named after the English economist, William Stanley Jevons. It was described in his 1865 book, The Coal Question

Jevons observed that England's consumption of coal soared after James Watt introduced a more efficient coal-fired steam engine. A more efficient system uses less fuel, yet consumption greatly increased; hence, a paradox. This has also been called the Rebound Effect or the Backfire Effect. 

This paradox is often misunderstood or even maliciously misconstrued to attack efficiency efforts so we'll spend a little more time on this one.

In retrospect, the cause of the soaring coal use in steam engines is apparent. In the 1800s, most labor was done with muscle power (either human or livestock). When the steam engine became more efficient, the cost of using coal became affordable and it started to displace workhorses, draft horses, mules, oxen, and the like. The society of the time was starved for horsepower to achieve the burdensome labors of the day. 

They were early in the S-curve of work energy (see below). Before Watt's engine, few people were using coal-powered machines. So coal, engine parts, and repairs were all expensive. The improved steam engine overcame the cost of coal, this cost was the barrier to entry for many applications of the technology. Clearing this barrier allowed steam engines to move up the slope, from the initial slow growth phase to the exponential phase. 

Technology Lifecycle

The decline phase is generally kicked off when a better alternative arrives, rather than the underlying need disappearing. The steam engine displaced animal labor. The steam engine was later displaced by a combination of internal combustion and electric motors, thereby completing the lifecycle.

Coal, on the other hand, was not as quickly dismissed as the steam engine. It found additional uses in electricity generation. Coal use hit its peak in 2007 and was generally displaced by natural gas (methane). Advancements in hydraulic fracturing of shale caused a "fracking boom." This left coal at a major cost disadvantage just as coal had done to ox power more than 100 years before. 

It's Not A Backfire Effect 

Now, with some perspective, you can see that this was not a Rebound Effect or Backfire Effect. It was a Breakthrough Effect! When an underlying demand is poorly served, a new technology that makes it easier to access will greatly increase demand.

The Jevons Paradox was not a Backfire Effect; it was a Breakthrough Effect.

Unpriced Externalities Causes A Tragedy of the Commons

Coal was cheaper than draft horses because of the more efficient steam engine but also because there was no direct cost to pollute. When you purchased coal (then and now), there was no additional cost for the air pollution that it would cause. That price would be paid collectively by everyone that had to breathe polluted air. 

How does Jevons Paradox apply to EVs? 

Our need to travel is not disappearing, it just might be served in a new and better way. 

Just as Watt's steam engine was more efficient than the ones that came before it, the electric vehicle motor is far more efficient than the internal combustion engine that it is replacing. Electricity is far less expensive to use as a fuel than gasoline or diesel. Electric motors require far less maintenance. EVs now have the ability to have similar ranges as gas-powered vehicles and the growing plug-in infrastructure makes it nearly as convenient to plug in as it is to fill up. Plugging in an EV in your garage to charge up overnight is even more convenient.

Given the cost, maintenance, and convenience advantages, once the initial cost reduces, EVs will have a "Jevons' Breakthrough."

Tying It All Together

These Laws are closely related and help us understand how things change with exponential growth. Wright's Law is about the reduction in labor cost as production scales. Swanson's Law is about the reduction in unit cost with scale. And Jevons Paradox is about the adoption of technologies, thereby enabling the prior. All the while, with Moore's Law continuing to increase compute horsepower to underpin all this growth.

How will these impact our renewable future? Wright's Law will result in battery prices dropping. This will result in EV prices continuing to drop. Since EVs are cheaper to operate, Jevons' Paradox means that new uses will emerge (ridesharing, delivery, autonomy, underground transport...). Moore's Law will enable more computing power, allowing EVs to become autonomous and connected with more in-car entertainment options.

Swanson's Law will continue to lower the price of solar panels. This along with the reduced price of batteries will allow for more renewable energy on the grid. This will further reduce the cost of energy and energy storage thereby making EVs cheaper to operate and again cheaper to build. This will further reduce our use of fossil fuels, moving them down the decline curve while providing us with cleaner air and water. 

These Laws create a self-reinforcing positive feedback loop that will accelerate EVs, energy storage, and renewable energy into multi-trillion-dollar markets.

Murphy's Law

Of course, you shouldn't forget Murphy's Law, which could bring all of this crashing down ☘️

and one more bonus axiom.

Hofstadter's Law

If you thought these changes were going to happen overnight, remember this recursive law:

It always takes longer than you expect, even when you take into account Hofstadter's Law.
— Douglas Hofstadter, Gödel, Escher, Bach: An Eternal Golden Braid

Sidebar: "Laws*"

Many of the things we looked at are called Laws. However, they are not Laws of Physics (like Boyle's law), nor do they fall into the legal category of laws (like Megan's Law). There's a long list of better ways to describe these (e.g., observations, dictums, principles, effects, axiom, rules of thumb, razors, corollaries, heuristics, hypotheses, parables, emergent properties...). However, the English language (as far as I know) does not have an overarching word for this category. Calling something a "Law" is far more catchy than calling it a "Heuristic." Moore's Observation just doesn't have the same ring to it, so here we are. 

If you know of a better term for these types of "Laws," please let me know in the comments below. There may be some obscure German word that's perfect for this; until I learn it, I'll use the term axiom or law.

Meta Sidebar About Sidebars    

If you've read this blog for any length of time, you'll know that I cannot say something like Moore's Law, without a pedantic sidebar about it's not really a "Law." I recently did this in another post about the term "AC Batteries." The AC Battery sidebar occurred relatively early in its post and it was relatively long. This broke the flow of the post before it even had much of a chance to get started. So now, I'm trying something new and putting all the sidebar(s) at the end of the posts. This way I can still note and clarify the turbid occurrences without breaking up the flow of the primary story.

Disclosure: I am long Tesla