Bitcoin Mining & The Grid (Part 1): Generators | Braiins

adminAugust 25, 2022

0 25 20 minutes read

YOU ARE READING: Bitcoin Mining & The Grid (Part 1): Generators | Braiins AT en.noithatvietphat.vn

This article is the first in a series on bitcoin mining and energy infrastructure. Each article offers an introductory level explanation of power grids and their relationship to mining to better educate miners and other bitcoin investors.

read part 2: transmission, reduction and behind the meter here.

Reading: Bitcoin mining tax our grid

‍

introduction‍

The most impressive feature of bitcoin is the mining algorithm. More beautiful than the math and theory behind the broadcast and difficulty settings themselves is the fact that it all actually works. that there are no death spirals in mining, that there is a global distribution of hashrate competing for energy infrastructure and cheap rates. it’s amazing. what were once cypherpunks competing with laptops in the late 2000s is now an incredible ecosystem of radically different business models: hosting and licensing, on-grid proprietary giants, generator joint ventures, off-grid natural gas and even pig manure!

Because of this incredible success and diversity of business models, there is no shortage of arguments about the effect of bitcoin mining on the network. Often bitcoin investors find themselves on social media (this author included) pointlessly arguing against a belief about bitcoin and energy, then turn around an hour later to argue with someone else on the opposite side of the coin. political spectrum about their beliefs on the same issue.

energy and bitcoin mining are two pretty complex topics, and many arguments online can be avoided (or at least made more polite) if all parties read a manual on how networks known as the electrical reliability council of texas (ercot) and how bitcoin mining can interact with these markets or specific pieces of electrical infrastructure from a power systems perspective.

Part 1 of this series will focus on generators and the relationship of bitcoin mining to its operation and role within the energy system. but readers should be aware that the concepts in this article are generalized and simplified to provide an introduction to grids, not an exhaustive study of them. networks are incredibly complex systems, and engineers spend their entire careers working on just one small aspect of how they work. the networks are also a marvel of technological coordination that carries with it all the mathematical complexity, rate law, and political baggage one can imagine. With all that in mind, let’s get started.

‍

what are generators?

A power system is a set of power “sources” (generators), “paths” (transmission), and “sinks” (loads) that must function with a high level of reliability and quality. within the limits of physics. generators are the power sources and thus generate power that needs to be delivered to the loads through the pathways available to them. the energy must be consumed immediately by the load at the time of generation, unless it is stored using energy storage technology. To manage this process of matching real-time generation to load across available routes, a central authority (grid operator) typically uses a mathematical algorithm to identify the most economical (cheapest) and most reliable (capable of continuing to operate in events where you lose an item(s)) way of operating the network. In layman’s terms, this means that sophisticated math is used to determine the best combination of generator output across the entire system to serve the load and prevent a single element from being overloaded in the event something fails or trips.

Below is a simple image illustrating the difference between energy sources (generators), energy paths (transmission lines and substations), and energy sinks (loads, residential load in the case of the image). transmission and distribution are generally differentiated by voltage level, but both are “paths” through power flows. ercot and other iso’s typically only work at transmission level voltages, and distribution companies manage those lower voltage levels.

Generation sources have traditionally been thermal generators. primarily coal, natural gas, and nuclear power plants that burn fuel to boil water & generate steam, which is used to drive a turbine to generate power. recently, solar & wind energy have become a growing percentage of the generation mix. these facilities operate passively depending on the weather and rely heavily on solar energy & wind forecast to predict your departure.

Today, generation developers are companies that specialize in the location and development of applications. interconnection of generators to the electrical system. These companies are focused on understanding what the future of the grid will look like and how they can develop generation infrastructure to capture value. generator owner/operators are companies that specialize in owning & operate the generators once built. Generation developers also sometimes own and operate the facilities they develop, and sometimes sell them to others.

High energy prices are the main signal for generation developers or investors to build more generators. recent tax legislation has become another sign that has stimulated the development of solar, wind and & storage, as investors seek to capitalize on the large tax windfalls from building such facilities.

ercot is the non-profit entity that operates the texas grid, which accounts for 90% of the electrical load in texas (the grid’s name is also ercot, I know, it’s confusing). ercot does not own cables, substations or generators, but rather is the “manager” of the system. this means that ercot has special responsibilities for planning the system, handling the actual transactions and operating the market in real time.

ercot has an “unregulated” market, which means that different entities must own different portions of the industry. no transmission owner can also own generation, etc. Power generators are privately funded at Ercot, and Ercot generators operate in one of two distinct categories (markets): power and ancillary services. ‍

These are the basics of the energy market.

All generators produce and submit to ERCOT (hereinafter “the grid operator”) a “supply curve”, which serves as an indication of how much money they would need to spend to produce one megawatt (MW). of power. for example, an offline coal facility needs the grid price to be at a certain value in dollars per megawatt hour ($/mwh) before it comes online (i.e. starts generating power) so that they can break even on your costs associated with turning on . once online, the plant needs a little more money to pay for a little more fuel to increase output a little more, and so on. the resulting supply curve represents the marginal cost, or the dollar value required to produce the next megawatt of power. this ends up looking like a curve that starts at the minimum output of the facility and goes up and to the right, as thermals need more fuel to produce more power.

‍

Unlike thermal generators, wind and solar power are passive energy sources with a production cost of $0, meaning that their marginal cost to produce the next megawatt is always $0 (a flat line at $0, instead of an upward curve). With the tax incentives for renewables referenced in the introduction to this article, some renewables can be priced negatively, meaning that a generator receiving certain tax benefits could get a credit of $26/mwh only for being online and therefore will bid -$25/mwh instead of $0/mwh in the market, and capture $1/mwh even though the grid price reflects its negative bid (-$25/mwh). Since renewables trade $0 or even negative on the market, they are often referred to as “cheaper” than thermal generators.

still not sure? let’s continue…

Grid operators typically dispatch generation (i.e. order a generator’s output at a specific level) starting with the lowest marginal cost generators and then moving in the direction of higher cost generations working through the “offer stack”. operators also optimize (and thus set the price of the network) for something called “constraints”, not just the cheapest generation.

but typically, in any given local area, the cheapest generation is dispatched first, and the supply stack is incrementally worked up to dispatch enough generation to match the amount of predicted load on the system. eventually, the operators will move up the list of offers until they reach a total number of generators that, together, will provide the amount of energy necessary to satisfy the demand in the given time interval.

It is crucial to note that a generator’s marginal cost data (the values they submit as part of their supply curve) do not contain earnings or information about the capital invested to build the generation facility. it only communicates the variable cost, such as the cost of fuel and regular maintenance.

The entity at the top of this bid stack during any interval that provides the final and most expensive megawatt hour sets the grid price for all generators for that interval, so they are sometimes called “price fixers”. Because both wind and solar energy bid the market at $0, in contrast, they are often called “price takers.” This is because these energy sources offering $0 never plan to be the marginal cost unit (the one at the top of the bid stack). if they did, everyone would be paid $0/mwh for their power…and no one would make any money! there is a whole field of study on how the electricity system markets will work when renewable energy begins to “fix the price”, which this author will obediently ignore in this article for the sake of time. we’re omitting the “constraints” here for the sake of simplicity, but know that this is generally how it works.

This dispatch process typically ends up looking like the graph below, where power generators below the market clearing price (gray line) and to the left of the demand curve (blue line) are turned on and paid the compensation price (grey line). line) for its energy. generators above the market settlement price and to the right of the demand curve remain offline. this process runs continuously throughout the day, redistributing generation based on expected load levels.

Decisions about whether to build a new generation asset are made after careful research and modeling of the power system by engineers (such as this author) to understand what the expected future price will be at a specific generation location in the net. These pricing models do not provide certainty, so in some ways they are akin to using a very powerful and precise, but inevitably inaccurate, crystal ball to forecast market conditions, streaming updates, future recalls, growth of load and other generation competition.

for example, if a company expects natural gas to become the “price setter” for the local area and is planning a solar park, it now has to analyze whether the price set by the natural gas generator will be or not enough to recover the capital cost of the solar farm. therefore, while solar generators may bid $0 in the market, that does not mean that the capital expenditures can be paid back at a price of $0/mwh.

Enough of power generators.

let’s move on to the second category: ancillary service providers.

In addition to producing enough power to meet demand, grid operators must be able to maintain a constant frequency of 60 hertz (hz). In real power systems frequency is actually a dependent variable of many things in a network and therefore maintaining it and keeping the system afloat is of paramount importance. frequency shocks can cause blackouts, which can mean weeks or months without power. Readers can view the real-time frequency of ERCOT and inspect the different types of ancillary services and their deployment here.

To maintain frequency, network operators buy & pre-book “ancillary services” of generators and loads that can rapidly increase or decrease your output. By rapidly increasing output, a generator can increase the frequency of the system. By rapidly decreasing its output, a generator can decrease the system frequency. Grid operators have a certain “amount” of mw they need to buy (reserve) in advance to make sure they have enough “firepower” online to deploy (order up or down) during real time to maintain the frequency during the day. when real time arrives, network operators deploy these purchased reserves to manage frequency fluctuations throughout the day.

This frequency market is separate from the power generation market, there are two markets. the energy market, in which generation is switched on or “matched” with the expected load (or demand). and complementary services, in which the frequency is maintained at around 60 Hz to maintain the operating system and avoid blackouts. the energy market is the big, heavy dial, the ancillary services market is the small, precise dial.

Certain types of loads can also provide ancillary services. And this is where bitcoin mining, often cited as a “network balancing agent,” comes into the picture.

‍

what do generators have to do with bitcoin?

After the basic explanation of power generators and ancillary services above, it is appropriate to revisit the same topics and add bitcoin in the explanation to understand the role that mining can play in the network.

This section describes miners who are exposed to nodal prices or wholesale prices, the same type of price that generators receive. this price goes through a couple of layers of coverage, aggregation and abstraction before reaching industrial loads or residential consumers. (These layers are why residential customers can pay flat rates.) But ercot has special rules on the horizon that would allow some broadcast-level payloads, like bitcoin miners, to receive nodal prices directly. however, for now, most bitcoin miners are exposed to nodal prices through ppas and some sophisticated arbitrage. All of these details can quickly get complicated, so for educational purposes, readers can assume that everything described here about miners getting variable prices is (or could be, if miners acted rationally) true.

‍

quick math for miners

A fun exercise for bitcoin miners is to calculate your $/mwh rate. Just as all generators have a “supply curve”, which they must send to network operators, bitcoin miners can be abstracted into a $/mwh value into which they convert energy. this number is also sometimes called your “break-even price” because, if the machines were being powered by electricity that costs that amount, they would become unprofitable.

The following tables show some back-of-the-napkin math using a hash price of $0.22 to determine what the $/mwh rate would be for various models of miners. this math includes some extra steps to find how many machines would be needed to constitute one mw of load.

The math for determining the equilibrium price without counting the machines is a bit simpler.

so what does this have to do with the grid? There are two key takeaways.

‍

bitcoin enters the list of offers.

remember that generators are shipped in order of marginal cost. the cheapest generators are shipped first and the most expensive generators are shipped last (if shipped at all). upload resources can also register with a network operator, get a small payment and be part of this offer stack as well, not just generation. but instead of adding to the generation side of the market, this load functions as a tool to lower the expected load target.

From the tables above, antminer s9s have a break-even cost of around $90/mwh, or 9 cents/kwh. By participating in the market as load resources, bitcoin miners coordinate with network operators in a similar way to generators, but instead of increasing generation, they reduce load demand in response to wholesale prices. the result is that miners flexibly push the marginal price of power down for the network. Put another way, they don’t push the marginal price higher than its break-even point.

Continuing to use miners with s9 machines as an example, this is how the network affects your operations. As Ercot operators work through the bid stack and crank generation until it matches the expected load, they face two choices when they reach the $90 marginal cost range.

option 1: mandate that bitcoin miners participate as load resources with a marginal cost of $90 to shut down.
option 2: order generators with a marginal cost of $91 to turn them on, so that load resources with a marginal cost of less than $91 (ie miners running s9s) can continue to operate.

This scenario comes with some caveats.

First, generators must respond to orders from grid operators or face fines. if the grid doesn’t work with consistent mathematical precision and transparent coordination, bad things happen. The ability to materially modify the network as a generator or load balancer carries great responsibility.

Secondly, for bitcoin miners, load adjustment in response to price signals is currently an “opt-in” situation, where miners are paid to participate, regardless of whether they are sent or not in real time. this means that the example bitcoin miner will be paid with the s9s, even if they don’t have to be turned off. but if a miner is exposed to nodal prices (for example, faced with the choice of paying $91 to win $90 or going out of business), cutting power when prices are uneconomic is simply a rational way to participate in the market. as are generators that don’t turn on unless it’s economical to do so.

okay, caveats aside, let’s go back to the two options listed above.

what is the best offer to minimize the total cost of the system? effectively, any type of load that is exposed to wholesale prices and that can be switched off during times of high prices, will be switched off. for example, steelmakers have been shutting down in response to market prices for a while. and this same behavior applies to bitcoin miners. any miner who doesn’t want to be online when the price isn’t cheap won’t be. then..

‍

Why is bitcoin mining special?

Bitcoin miners are unique because of how fast, transparent and flexible their response to price fluctuations can be.

riot’s whinstone mining farm in texas is a perfect example of how miners shut down during tough conditions at ercot. during winter storms in February 2022, the site shut down 99% of its operations to reduce load. this flexibility made headlines.

now imagine if ercot has visibility into every miner over 10mw that is connected to the transmission grid. the price at which the machines break even would become an integral aspect of the wholesale energy price because bitcoin miners themselves can set the wholesale price.

Consider a high price situation where marginal power prices rise to $90/mwh, and then bitcoin miners in houston running some s9 are ordered to shut down. (Remember: They wouldn’t want to be online anyway, and Ercot pays them just for being available to turn off at higher prices.) By shutting down, let’s say these miners actually allow load to match generation, and thus become price setters, setting the marginal price for the entire ercot system. All generators get $90/MWh for that interval, assuming there are no restrictions, because that price is how much Bitcoin could have produced one MW of S9 with that marginal MW. bitcoin became a price setter!

Further imagine that bitcoin miners with the s9s would send something like this graph below to ercot, where instead of a supply curve, they offer to reduce their mw load by a certain amount based on the network price. this would be the opposite of the generator supply curve mentioned above where marginal load shedding is offered instead of marginal power generation. And of course it’s not hard to imagine a miner that is not only running s9s, but also has a diverse portfolio of machines with a variety of breakeven prices.

‍

this kind of paradigm will certainly make asics cheaper and thermal cycling will become a real concern for machines. but what else? this still adds load to a system, right? sort of “push you off a ledge but then catch you before you fall” and claim it will save your life. so in a place like ercot where the market is already tight, $90/mwh is incredibly expensive. but bitcoin mining can do something else extremely well.

‍

miners solidify generation revenue forecast.‍

Attentive readers will remember what we saw earlier in this article about how generation developers rely on engineers like this author to run sophisticated and not-so-accurate models to forecast the prices that determine whether a location is lucrative enough to build a generator. Also, remember that Ercot prices are considered “price signals” for new generation, making them inherently lagging indicators of cargo growth that could support new generation builds.

So what if next generation development could be partially subscribed with bitcoin as a collocated buyer (buyer of a certain amount of energy)?

Instead of having to rely solely on network prices for revenue, new generators could purchase “buy insurance” that allows them to hire a bitcoin miner if their network price forecasts turn out to be a fiasco this would present an incredibly novel tool to de-risk generation development, allowing generators to take their buyer (their energy buyer, a bitcoin miner) with them to a new site.

For renewable generators, this is especially attractive. a 200mw solar or wind farm could see huge profits by collocating it with a much smaller (eg 30 – 40mw) bitcoin mine that simply draws power from the grid when the solar or wind farm doesn’t generate enough to satisfy your needs. This co-located mine could generate enough revenue for the generator to bolster financing while allowing 160-170 MW of nominal capacity to the grid during peak hours, all while still bidding $0.

The economics are slightly different for thermal generation (e.g., natural gas), as under the current “marginal cost” paradigm, the marginal cost of the thermally produced megawatt supplied to the grid after serving the placed mine would technically be more expensive than those megawatts first supplied to the mine. Remember how the normal supply curve is up and to the right, as thermals require more fuel to produce more power. this is not to say that renewables are better, since all types of generation have their advantages and disadvantages, but rather a highlight of the implications of generation with a fixed marginal cost of $0.

Current financial paradigms for developing generation assets are not set up for this type of underwriting: most companies developing generation have a rigid and conservative set of requirements, and bitcoin is too new a technology to fit into this hole from now on. however, the paradigm shift is here, and generation developers who are willing to be the first to take advantage of this model will certainly reap the rewards. I hope this placement model to improve funding hits existing generators first, as they have no problem trying something new if they no longer meet revenue targets.

‍

advanced: playing around with bitcoin mining

‍This section explores more advanced implications of associating generators with miners. in certain situations in which the network operator knows that the real-time conditions in the immediate future will be tight (there is not much margin between the available generation and the expected demand), the “ruc” network operator (commitment of reliability unit) thermal generation units, whereby the network operator orders certain thermal generators to be online, running at minimum and therefore available to be dispatched if necessary. Since generators take time to come online, having them up and running is a more conservative approach to operations planning.

this is different from the standard procedure, since normally those generators decide themselves when to be available through their supply curves & your police (current operational plan). Earlier in this article, grid operators were described as generation dispatchers, but that is only half true. generators actually tell the grid operator when they will be online and available. then, once the generators are online and available, the generator must follow the operator’s dispatch orders.

when the unit is run by the grid operator, the operator has to pay a premium to the generator to be online and available, a premium depending on the anticipated grid economic conditions (otherwise the generator would have decided to turn on itself ) . with bitcoin mining settled, a generator would always have an incentive to be online & available and thus would make rucing a thing of the past and shift the cost burden of rucing from the consumer to the collocated bitcoin miner. environmentalists probably wouldn’t like the idea of running more thermal units, but they couldn’t deny that moving ruc costs away from the consumer and into the bitcoin miner would reduce the costs consumers face associated with rucs while also would improve reliability.

‍

mining as an auxiliary service

the potential of bitcoin mining ancillary services is a bit more straightforward than the generation aspect, as most readers are probably familiar with the narrative of bitcoin mining providing “network balance” . To quickly summarize ancillary services: After grid operators ensure through the power market that they have enough generation to meet demand in any interval, they also need to regulate the frequency on the grid, as it ranges from 59.5+ and 60.5-hz. (for context, the frequency of a network is always oscillating within this band as operators in a real system work to match generation and load). traditionally keeping the frequency within this range bands is achieved by taking advantage of generators that can rapidly ramp up or ramp down their output, and also, for certain acute (rapid) events, by large loads that are connected to relays so that they drop immediately if the relay detects a frequency event.

The graph below is an example of a generator-forced outage (i.e., a generator tripped or was forced off for some other unplanned reason) that caused a drop in frequency (blue line), causing which resulted in a load of resources (eg potential bitcoin miners). , green line) responding immediately by tripping and restoring the network frequency.

‍

Bitcoin’s place as a network resource is pretty clear. Large flexible loads that have the ability to pay for themselves to come online and respond immediately to frequency events are a new asset class for the power system. so what exactly does this look like? In addition to simply responding to acute events by dropping cargo, at Ercot, ancillary services are currently sold “day ahead,” that is, the day before operations. miners who can qualify to provide these types of services (by showing that they can scale up or down and follow instructions quickly) will sell their capacity at auction. the network operator will buy the services, forcing the miner to reserve those megawatts during its operation the next day. during real time, the miner may be asked to increase or decrease to fulfill his obligation, depending on the type of service he sold.

firmware that allows mining machines to ramp up and down while minimizing long-term damage to the machines would be an amazing tool for the power system. but it is unclear to what extent asic miners can or will incorporate this kind of ramping ability.

As briefly mentioned above, thermal cycling will likely be a limit to how often miners can increase or decrease frequency to change the frequency of operators. this author hopes that mining companies will eventually emerge specifically to perform this type of service, using extremely old machines and combining them with generators that cannot perform this service by themselves (renewable or nuclear), in this way they can take advantage of the existing electrical infrastructure and split the revenue with a generator that is already well versed in the electrical markets.

‍

Frequency regulation is a double-edged sword.

Although this author is very optimistic about integrating bitcoin miners into the power system, placing such concentrated rampant loads on the system also creates risks.

One risk is unclear operating hours. if bitcoin miners do not transparently share their operating schedule or supply curve information with a network operator, the operator has no knowledge of when and how much bitcoin miners will turn on or off throughout the day . and because large miners connected at the transmission level are not treated with the same scrutiny as generators, they are not currently required to share this operating schedule or present a supply curve to grid operators.

changing the load drastically affects the frequency, which can cause blackouts. large miners increasing or decreasing their production has a huge effect on the network, as their production can drastically affect the frequency of the system and the rate at which the frequency changes. with more loading ramps, more ancillary services will need to be purchased to ensure the grid operator has enough firepower to manage the frequency during real-time operations.

but how many ancillary services should be reserved to handle turning bitcoin miners off and on? the network operators won’t know unless the miners give them an idea of how they plan to operate. network operators already have to acquire more reserves due to the penetration of renewable energy, bitcoin miners not operating transparently present another reason for the network to obtain (buy) more resources from ancillary services to regulate frequency in real time.

this problem will probably be solved with new rules and additional incentives. bitcoin miners operating at a significant size that interconnect on the transmission network will likely face interconnection rules and responsibilities that more closely resemble the existing rules for generators, not the rules for uploads. for example, bitcoin miners will have to prove that they have redundancies in their network connections, so that one faulty cable doesn’t drop all of their load.

miners may also have to submit operating schedules or supply curves, and could face penalties for not behaving according to their schedule or supply curve, penalties which in turn will be used to pay for more ancillary services. For their service, miners are likely to be allowed to take nodal prices and avoid annoying streaming, delivery, and matching peak charges that would add to their total cost.

Miners connected to a distribution tier may not face the same scrutiny, but they also won’t be exposed to the same pricing and fee incentives as miners connected to streaming.

‍

conclusion

Congratulations on making it this far! The goal of this article (and the entire network series) is to develop among bitcoiners a better understanding of how modern networks are managed and operated, so that creative miners and enthusiasts can think of new ways to bring these two together. industries.

Hopefully this article will provide useful information on how ercot works and increase the reader’s appreciation of the potential bitcoin mining has to change the energy industry. Many of the topics discussed in this article are generalized for the sake of simplicity, and most of the sections above are probably worth 50 pages of their own analysis.

Issues such as restrictions, negative prices and others are also important pieces of the puzzle that bitcoin mining presents, and all of these will be addressed in later articles.

‍

read part 2: transmission, reduction and behind the meter here.

Reading: Bitcoin mining tax our grid

‍

This article was written for the braiins blog by blake king. Blake is an energy engineer who builds and analyzes software models of electrical networks. his views here do not reflect those of any of his past, present or future employers. follow blake on twitter.

‍