How bitcoin makes internal combustion engines more energy efficient.
An essay exploring the sexy side of bitcoin mining — engines and power generation.
A bitcoin mine can be deployed to monetize electricity wherever it is generated and thus yields increased energy utilization efficiency for all energy sectors.
In recent years a lot of excitement has been generated over the potential to use bitcoin mining as a way to improve the capacity factor of intermittent energy sources by reducing associated curtailment that is inherent to intermittent sources such as wind and solar.
While I am keen to see how bitcoin mining plays out within the intermittent energy industry, I am personally far more interested in understanding how bitcoin mining can improve the application of internal combustion engines throughout the world. I have several reasons for this interest:
- Internal combustion engines are everywhere and are used by everybody.
- Internal combustions engines will continue to be everywhere for the foreseeable future.
- Internal combustion engines can have very low utilization and capacity factors in many applications resulting in wasted fuel.
- Internal combustion engines are just fucking awesome… brrraaapppp brrrraaaaaaaaapppp!
Bitcoin — the ultimate worker
If you don’t know what a bitcoin mine is that’s fine, for this essay all you need to know is that it is simply a bunch of computers doing fancy computational work that make the bitcoin network run properly and stay in sync. All bitcoin mining computers need is electricity and an internet connection.
These computers — called ASICs — do work over the internet and since the internet is accessible anywhere on the planet via cellular or satellite connection it means they can be plugged into any electricity source, anywhere. In this way a bitcoin mine is spatially unconstrained by its end user / owner, unlike almost every other electrical load.
For example, a bitcoin mine can still be productive for its owner even if it’s located on the other side of the planet because the owner is connected digitally, whereas you sort of need to have a lightbulb within eyesight for it to be any use to you.
A load having no spatial constraints allows a bitcoin mine to pair very effectively with engines and power plants which are operating with poor energy efficiency.
Engines and energy efficiency
I am here to talk specifically about internal combustion engines, but I’d like to first frame the discussion around thermodynamic heat engines in general to get you into a frame of mind of what engines are and how their efficiency is measured.
A heat engine is a device that extracts useful work from the flow of heat from a hot source to a cold sink. In the thermodynamic sense this also means from a higher pressure source to a lower pressure sink. A few examples of common heat engines follow:
- A wind turbine is an engine that extracts work from the wind to turn a turbine and generate electricity which is consumed by a downstream load.
- A living creature is an engine that extracts work from digesting food to propel itself and perform its many cellular functions like cell division and growth.
- An internal combustion engine extracts work from the heat and pressure generated by combusting fossil fuels in air, driving a rotating shaft which is consumed by a downstream load.
The thermodynamic efficiency of an engine is measured by the ratio of useful work to the flow of heat required to produce that work. Efficiency = W / Qₕ.
Now that we’ve established what engines are in general, for the rest of this essay when I refer to ‘engine’ I will be talking about internal combustion engines specifically.
Better loading yields better fuel efficiency
The mechanical performance of an internal combustion engine and its overall fuel efficiency is generally independent of whatever type of downstream load that it is working. The engine’s efficiency is a function of a variety of conditions such as fuel quality, ignition timing, fuel-to-air ratio, internal friction, rpm, loading and many other factors. The type of engine also matters as reciprocating, Otto cycle engines have different mechanical properties than rotating, Brayton cycle engines.
While loading an engine doesn’t change its inherent mechanical properties and performance, it does have a significant effect on how efficiently the engine converts the fuel into work. Below are a couple charts relating fuel efficiency vs engine loading for the two main types of internal combustion engines — reciprocating engines and turbines.
In both charts shown above for both reciprocating and turbine engines we see the electrical efficiency (the ratio of work output to fuel energy input) increase* as load is increased. When power plants are underloaded they consume more fuel for the output compared to when they are fully loaded, especially when the fuel is natural gas or petrol.
*Note: Diesel engines behave a little differently with efficiency peaking just before peak load, but we will not explore why here.
When it comes to distinguishing between turbines and reciprocating engines, there are a few notable differences:
- Reciprocating engines can operate relative efficiently at lower rpm, whereas turbines cannot
- Turbines have higher peak loading efficiency than reciprocating engines
- Turbines are usually better suited for very large power loads, 1 megawatt and above, whereas reciprocating engines are best suited below 1 megawatt.
Regardless of the kind of engine, when it comes to generating power efficiently it is very important to load engines as much as possible to maximize the value of the fuel consumed. In today’s world full of fiat hysteria over carbon emissions* it is critical that we continue to optimize how we optimize our engines.
*Using carbon equivalent emissions as a measure to mitigate pollution is actually quite ineffective, but I’ll tackle that mountain in a future post.
The challenges of variable loading
It is not just fuel efficiency and and carbon emissions we need to worry about when it comes to underloaded engines.
Variable loading forces engineers to design power systems that can handle peak loading periods while also turning down output during periods of low demand. This usually results in oversizing power systems or adding redundant, peaker power plants.
Utility grids supply electricity to residential, commercial and industrial sectors and will experience variable loads not just during the day but as well as across the seasons. At night when most people are asleep the load on the grid is generally low, while during the day it picks up as people conduct their business. Summer months may experience higher loads in warm climates due to the demand on power from air conditioning.
Power systems engineers have few options at their disposal to mitigate when and how people choose to consume electricity and thus must rely on incentivizing people to shift their load by how the electricity is priced.
But the problems caused by variable demand do not reside just with utility power systems. Power plants are used in a wide array of applications well beyond prime movers for utility grids, for example in large industrial facilities like midstream natural gas plants and upstream oil wells. Even large buildings, ships and small automobiles can have variable load problems as well.
Engines are everywhere and the opportunities to help optimize their loading by leveraging bitcoin mining are abundant. So let’s explore a couple interesting practical and speculative applications.
Let us start with a few examples in the oil and gas industry because oil and gas is what I know well as someone running a business focused on helping producers deploy bitcoin mines all across North America.
The oil and gas industry not only has cheap fuel for engines in the form of wasted natural gas, which makes it prime territory for mining bitcoin, but it often encounters scenarios where variable loading requires inefficient engineering solutions.
Load banks, of the dumb variety.
Load banks are effectively large banks of resistors with cooling fans that are built in compact packages to be used to load up engines. They are commonly used for start up load testing of new engines but in many cases, such as in midstream gas processing plants, load banks are used to keep the power plants above a minimum load in order to protect them from premature failure.
Reciprocating engines that are used as the prime movers to power midstream gas processing facilities are usually sized for the maximum facility throughput. The reality is that, especially in periods of low commodity pricing, these engines may end up largely oversized for the facility throughput. Running reciprocating engines at low load is not a good idea as low combustion temperatures can cause damage due to wet stacking:
Cummins Engine Co. defines wet stacking as a condition that can occur when running an engine at a no-load or less than 30% of the unit’s standby power rating, that manifests itself “in the accumulation of carbon particles, unburned fuel and condensed water and acids in the exhaust system due to incomplete combustion caused by low combustion temperatures.
Therefore these engines are often loaded up by these “dumb” load banks to protect them from costly maintenance events. This means that more fuel is burnt for no true productive reason.
So why not use a “smart” load bank instead, like a bitcoin mine? Not only can the engine be loaded up to prevent wet stacking but the extra fuel consumed is monetized and transmuted into beautiful, clean bitcoins.
Monetizing redundant and idle engines
Sometimes power system engineers choose to deploy multiple smaller engines instead of fewer, larger engines to power a facility, either in parallel or simply for standby conditions. There are many reasons they might go this route, either simply for redundancy back-up power if the prime engine goes down for maintenance or perhaps as a cost savings measure.
When engines are sitting idle they are doing nobody any good and the capital that was used to build them is not doing any productive work. From an emissions perspective, all of the carbon that was emitted upfront to manufacture and deploy that engine in the first place is not finding a return on investment. Building idle surplus equipment is not exactly great for the environment and I’m sure Mother Nature would have preferred we kept her precious minerals in the ground!
So while we have so much idle infrastructure why not load it up with a bitcoin mine? If the fuel is cheap enough then it makes sense to put this engine to work so it can slowly pay off its capital investment. Even if it’s a critical back-up engine it can still be engineered to transfer over to utility power by shutting off the bitcoin mine in the case of the prime engine failure.
I could offer more examples as I’m sure there’s many more practical applications out there but I think you get the drift.
Whereas the practical applications I described above generally have low cost fuel and surplus power available, here I will speculate on some interesting (but probably impractical) applications. Further research is certainly required, I just think this is fun to ponder about, so don’t be hatin’!
Commercial and military marine ships are huge, lumbering beasts that need a ton of energy to accelerate their mass, overcome drag and displace water. In some cases, a combination of reciprocating and turbine engines are used:
Gas turbines are commonly used in combination with other types of engine. Most recently, RMS Queen Mary 2 has had gas turbines installed in addition to diesel engines. Because of their poor thermal efficiency at low power (cruising) output, it is common for ships using them to have diesel engines for cruising, with gas turbines reserved for when higher speeds are needed. However, in the case of passenger ships the main reason for installing gas turbines has been to allow a reduction of emissions in sensitive environmental areas or while in port.
Having to engineer a combination of diesel reciprocating and gas turbines engines complicates the system design and increases the cost, so could the diesel engine be displaced by using a bitcoin mine to load up the turbine to its practical load limitations while its at low cruising speed?
Of course we would expect to use more fuel that would not be put towards the main purpose of the engine — to propel the ship — however it is possible that the cost savings in not building redundancy and secondary power systems, each with their own maintenance challenges, could prove feasible. Many other considerations would need to be analyzed in a thorough feasibility study, like the added fuel cost of the weight of a bitcoin mine, but it could be a fun exercise for the marine power systems engineer.
The world is slowly moving towards electric (EV) and hybrid vehicles, but the significant limitations of today’s batteries mean internal combustion engines will continue to be a large part of the equation for the foreseeable future.
So while we are stuck with fossil fueled internal combustion, how can bitcoin mining make automobile applications more efficient?
Similar to the marine propulsion example above, the point of the engine is to propel the vehicle, not to mine bitcoin. We certainly do not want to consume a single ounce of expensive gasoline or diesel to power a bitcoin mine, so what is the opportunity here?
Braking of course! Every time an automobile brakes heat is generated and energy is dissipated to atmosphere. EV’s and hybrids have the ability to regenerate that brake energy by instead recharging a battery or hydraulic accumulator, but the old fashioned ICE vehicles do not. Could it be feasible to brake the vehicle by loading the engine with an ASIC instead of applying the brakes?
Once again, I won’t attempt to do a feasibility on this crazy application and will instead maybe reserve that for a future post. Conceptually anyway, the bitcoin industry is trending towards ever greater ASIC power density (power consumed per unit weight) and cost efficiency, so perhaps it could make sense. It could be as simple as installing a higher capacity 12V alternator off the engine to load up a few kilowatts worth of ASICs during braking — the ASIC power supplies bypassed and everything!
In this essay we learned that:
- Running at low % engine loading is generally a bad thing for engines and bitcoin mining provides a new opportunity to optimize engine loading and improve their productivity per unit of fuel spent.
- Human beings are literally engines, so if you’ve been hating on internal combustion engines just stop it! :P
- There are many practical opportunities for applying bitcoin mining engines in industry, in both the fossil fuel and intermittent energy industries.
- Interesting (and farfetched) applications for bitcoin mining could be feasible such as vehicle braking and marine propulsion.
My final words that you can take home to your coldcard are this:
Internal combustion engines are not going anywhere and while we continue to be blessed with their presence (or cursed, depending on how much fiat you put in your breakfast) it is worthwhile finding new ways to improve their fuel efficiency!
Onwards and upwards, to the moon we go!