Welcome back everyone, I'm Jordan Geesege, and this is The Limiting Factor. This is video number one of my Grid Storage series.
大家欢迎回来,我是乔丹·吉斯奇,这是《限制因素》节目。这是我《储能网络》系列的第一集视频。
One of the first things you'll notice about any grid storage product is that they're always marketed with a duration figure that's usually listed in hours. Some products like Tesla's Mega Pack are listed as 2 to 4 hours, and others like form energy are listed as 100 hours.
The question is, what exactly does duration mean, and how is it determined? Furthermore, why can't a grid storage project take, for example, a 2-hour grid storage product and buy 5 of them to make a 10-hour solution? To answer those questions, today I'll walk you through what duration means as a technical term and why it matters, and then get into some cost calculations to work through duration as an economic term.
Before we begin, a special thanks to my Patreon supporters, YouTube members and Twitter subscribers, as well as RebellionAir.com. They specialize in helping investors manage concentrated positions. RebellionAir can help with covered calls, risk management, and creating a money master plan from your financial first principles.
I'd also like to credit the book, Monetizing Energy Storage, by Oliver Schmidt and Ian Stoffel, which provided the inspiration for this video series. All images in this video that don't have a credit listed were taken from the book. If you'd like to read the book yourself, which I recommend, it's available for free as a PDF and it's also for sale as a book. I'll provide a link to those in the description.
Lastly, thanks to Matt Smith of RebellionAir, who has a background in energy for reviewing the video.
最后,特别感谢在能源领域具有背景的RebellionAir的Matt Smith对本视频进行评论。
Let's start by answering the question that appeals the most to common sense. Why can't a grid storage project just stack two-hour grid storage units to create a 10-hour storage solution? The simple answer is that it comes down to design. Two-hour energy storage is most cost efficient on a two-hour time frame, because that's the use case it was designed for. And 10-hour energy storage is most cost efficient on a 10-hour time frame for the same reason. That means if a grid storage product stacks several two-hour energy storage units, rather than making the project cost efficient at longer durations, it just provides more storage that's cost efficient at a two-hour duration.
Note that I keep focusing on cost here rather than revenue. That's because one of the underlying assumptions of this video is that if a grid storage product is the cheapest option at a specific duration, that's the duration where it'll most likely be used, regardless of what other revenue streams it can pursue with that capacity. We'll get more into the revenue side of the equation in other videos.
So why is the cost of a grid storage product so tightly tied to duration? To understand that, we first need to know what duration means as a technical term. All energy storage technologies, whether that be flywheels, batteries, or hydrogen, can all be assessed from the perspective of nominal power capacity and nominal energy capacity. Dual just means the maximum safe power and energy capacity under normal operating conditions.
So I'll use the term's power capacity and energy capacity for short. For the purposes of this video, power capacity is measured in any increment of watts, like kilowatts or megawatts, and it's a point in time measurement of how quickly some other type of energy is converted to electricity. That's as opposed to energy capacity that's measured in watt hours, which is energy stored in some other form before it's converted to electricity, like chemical or mechanical energy.
Let's look at an example. With pumped hydro, the gravitational energy of the water in the upper reservoir is converted into electrical energy when gravity pulls the water through the turbine. Bear in mind that for the rest of the video, I'm going to use the word turbine for short, even though it doubles as a pump and is part of the generator slash motor. Moving along, the size of the reservoir determines the energy capacity and the size of the turbine determines the power capacity.
Because all grid storage technologies have an energy and power capacity, they also have what's called an energy to power ratio. For example, a large pumped hydro station will have about 20 gigawatt hours of energy capacity in the upper reservoir and turbines that provide about 2 gigawatts of power capacity. So in this case, that results in an energy to power ratio of 10 to 1.
You can probably see where this is going. The 10 to 1 energy to power ratio means that a large pumped hydro station can provide power for about 10 hours, which is called discharge duration. Discharge duration is generally abbreviated to just duration in grid storage products.
Now that we know the technical definition of duration, we can look at why it's so tightly tied to the economics of grid storage. In short, it's because each technology has strengths and weaknesses for converting energy to power that's dictated by physics and engineering. That in turn results in a duration where each technology is most cost efficient.
To further explore cost efficiency, let's again use pumped hydro as an example. Pumped hydro reservoirs can be made larger to increase energy capacity and more or larger turbines can be used to increase power capacity. So why not just use pumped hydro for every duration by either adding turbines or increasing the size of the reservoir. In the case of short durations in the range of minutes, emptying and converting a leg-sized reservoir to power that quickly wouldn't be cost efficient. That's because turbines are expensive and they need to run as much as possible to lower the capital cost per unit of power generated. So usually pumped hydro runs during off-peak periods to fill the upper reservoir over several hours and then discharges over several hours during peak periods. If it discharged in a few minutes, that would require more turbines, which are the primary expense and then the turbines would sit idle for hours a day not generating revenue. That is, for shorter durations, it's better to use a technology with a lower energy to power ratio that can convert energy to power more cost efficiently. Whereas one kilowatt of power capacity for pumped hydro costs about $1,000, it's only about $40 for supercapacitors.
What about longer durations beyond about 20 hours for pumped hydro? You could use a smaller turbine or a smaller number of turbines and a very large reservoir, but this time the reservoir would be underutilized. It would spend most of its time being slowly emptied and filled over the course of days or weeks. Even though excavating a reservoir to create energy capacity is relatively cheap, it's still expensive compared to other options like hydrogen storage. It's because the energy density of pumped hydro is 1 watt hour per cubic meter. Whereas for hydrogen it can be hundreds of watt hours per cubic meter and in some cases no storage tank is needed and the hydrogen can be stored in rock formations underground. That means hydrogen energy storage can be profitable at longer durations because the cost of storage is low.
So in the most basic of terms, some technologies produce power more cheaply and other technologies store energy more cheaply. Water ration tells us what the sweet spot is for the ratio between power and energy and therefore where it's most cost efficient and most likely to be profitable. As a side note, for some technologies like batteries, energy and power are coupled. That's because a battery both stores energy and converts it to power. For other technologies like pumped hydro, energy and power are decoupled. That's because the energy and power, that is, the reservoir and turbines, can be engineered and built separately. Coupled versus decoupled is a useful concept because it can give hints as to how a technology will scale. Decoupled technologies tend to be better suited to longer duration use cases because they use tanks, reservoirs or geologic formations to store the energy, which is cheap. Whereas the storage mechanism in a battery, for example, uses a cathode and anode to store energy, which is by comparison a complex and expensive way to store energy.
Now that we know what duration is from a technical perspective and why it matters, let's take a closer look at the relationship between duration and cost. An important building block for assessing the cost of power and energy capacity is the concept of specific investment cost. It measures how much each incremental unit of power or energy capacity costs but doesn't factor in any other variables. I'll add more variables in a moment.
The image on screen shows that turbines cost $90 per kilowatt, which is specific power cost, and the reservoir costs $1 per kilowatt hour, which is the specific energy cost. Bear in mind that these numbers aren't real world costs, but rather numbers that we're selected to simplify the calculations that we'll walk through in a moment. The bar chart on screen, which we touched on earlier to compare the cost of power for pumped hydro and supercapacitors, shows more realistic numbers for specific power and energy cost.
Although specific cost is useful for a quick assessment, it's handicapped by the fact that it looks at power and energy costs independently. In reality, you can't have a pumped hydro station with just a reservoir and no turbines. And you can't have a grid storage battery pack with no power regulation and conversion equipment. To take those factors into account, we need to use the total investment cost calculation.
For total investment cost, the specific cost information is combined with an energy to power ratio to look at the economics of a grid storage technology at a specific duration. Any energy to power ratio can be used, but I'll stick with the 10 to 1 ratio we've been using for pumped hydro, which is a duration of 10 hours. For the 10 to 1 ratio, the energy cost is 10 units of energy capacity at $1 per kilowatt hour each for a total of $10. And the power cost is 1 unit of power capacity at $90 per kilowatt. So, $90. That adds up to a total investment cost of $100 for the pumped hydro station.
From there, we can use that $100 figure to find the average power and average energy cost, which, confusingly, can also be called the total investment cost of power and the total investment cost of energy. Let's run through the calculation. Average power cost is calculated by taking the $100 total investment cost of the hydro station we just modeled and dividing it by its 1 kilowatt power capacity. The result is an average power cost of $100 per kilowatt. For average energy cost, we again take the $100 total investment cost, but this time divide it by the 10 kilowatt hours of energy capacity to arrive at $10 per kilowatt hour.
Now that we've finished all the calculations, let's do a quick recap of the results and what they tell us. Specific investment cost told us that independent of any other variables, each unit of incremental power and energy cost $90 per kilowatt and $1 per kilowatt hour, respectively. Average energy and power cost, on the other hand, factor in both the energy and power-related equipment costs for each incremental unit of capacity. And the result was $100 per kilowatt and $10 per kilowatt hour, respectively.
Besides getting a more accurate view of incremental cost at the station level at one specific duration, what else can we do with these calculations? If the total investment cost calculations are run for multiple durations for multiple technologies and graphed out, the result is on screen. As I said earlier, confusingly, average power and energy costs are also known as the total investment cost of power and the total investment cost of energy, which is why the Y-axis says total cost followed by kilowatts and kilowatt hours rather than average cost. For each point along the X-axis, from 1 minute to 100 hours, the energy to power ratio or duration was entered into the total cost calculation to arrive at a cost for that technology at that duration for both power and energy.
As for the technologies that were assessed, pumped hydro is in blue, flow batteries in yellow, lithium ion batteries in red, and supercapacitors in orange. Using power as an example, supercaps are the cheapest option up to 12 minute bursts of power. Lithium ion and flow batteries are the cheapest option from 12 minutes to 10 hours and beyond 10 hours pumped hydro is the cheapest. And of course for the total cost of energy, it's the same result at the same durations. Supercaps are cheapest up to about 12 minutes, lithium ion and flow batteries from 12 minutes to 10 hours and beyond 10 hours pumped hydro is the cheapest. Each of these technologies could of course be used at other durations but it wouldn't be as profitable or could result in a loss. There are of course exceptions to that, but I'll get into that in later videos.
As a final note, once again, bear in mind these estimates don't include factors like operating costs, finance costs, and power costs. The purpose here is to add another layer of nuance to duration. Although a technology might technically have a discharge duration of, for example, two hours, the duration where its most cost efficient actually falls into a range rather than a specific point in time. Furthermore, that range isn't just shaped by the physics and engineering limits of a specific technology but also competition from other technologies that may be more cost efficient for a given duration.
In summary, today we covered several aspects of duration. The first is that duration refers to discharge duration, which is a technical term and it's another way to express the energy to power ratio of a technology. A 10 to 1 energy to power ratio means a 10 hour discharge duration.
Second, duration matters because at a physics and engineering level, every technology has first principles characteristics that lend it to providing energy capacity or power capacity. Those characteristics result in a maximum rate that it can or should be discharged before damage or safety issues arise. Because those characteristics are fundamental to a technology, duration tends to be the point where a technology is the most cost efficient.
Third, we looked at specific cost, which tells us how much each unit of power or energy costs for each technology independent of other variables.
接下来,我们关注的是具体成本,它告诉我们每种技术每单位电力或能源的成本,独立于其他变量。
Fourth, we looked at total investment cost, which takes into account that power and energy aren't independent variables. There's always an energy to power ratio or duration to factor in. That's because to be useful, power capacity requires energy capacity and energy capacity requires power capacity.
Fifth, if we use the total investment cost formula and run a series of calculations using a range of ratios for energy to power capacity and plot the results on a graph, then we can see how the cost of each grid storage technology changes with duration. This illustrates that rather than duration being something that's fixed at a point in time like one hour, each technology has a relatively wide sweet spot for cost, where it has advantages or disadvantages compared to other technologies.
What all this means from an investor and consumer perspective is that when a company that sells grid storage hardware gives a duration figure, they can be referring to either the technical term for duration or they could be trying to market their product where they believe its cost competitive with other technologies.
For example, Tesla's Mega Pack uses the technical definition of duration and offers two options. A two hour duration option and a four hour duration option. Both battery packs have roughly the same energy capacity at about 3.9 megawatt hours, but the two hour duration version has a power capacity of 1.9 megawatts and the four hour duration version has a power capacity of 1 megawatt. That is, an energy to power ratio of 2 to 1 and 4 to 1 for a duration of 2 hours and 4 hours. This is a side note, the two hour version has a price tag that's about 13% higher than the four hour duration version. Why the cost difference if both packs store the same amount of energy? It's because just like pumped hydro needs an extra turbine to generate more power, the two hour Mega Pack needs additional inverter capacity to supply extra power. That is, each product has different economics to cater to different grid services. I'll talk more about grid services in another video.
EOS Energy on the other hand lists their grid storage technology as good for durations of 3 to 12 hours. In this case, rather than using a technical definition of duration, EOS believes their product will be competitive for 3 to 12 hour grid services. EOS has an interesting product that has potential, probably the most interesting new battery technology I've seen for grid storage outside of sodium ion. As part of this grid storage series or after the series is complete, I'll do a deep dive into their technology.
If you enjoyed this video, please consider supporting the channel by using the links in the description. Also consider following me on X. I often use X as a test bed for sharing ideas, and X subscribers like my Patreon supporters generally get access to my videos a week early.
A special thanks to Phil Roberts for your generous support of the channel, my YouTube members, X subscribers, and all the patrons listed in the credits. I appreciate all of your support, and thanks for tuning in.