Heating With Bitcoin Miners: What It Actually Costs

Tyler Stevens quit his job at a Lockheed and Boeing joint venture six hours after watching a rocket he had worked on for two years leave the pad. He drove straight to Denver International, flew to Nashville, and turned up at Bitcoin Park with no plan beyond a conviction that Bitcoin miners belonged in people's houses.

That was January 2024. He now runs Exergy, and he has spent the eighteen months since arguing a single point: a Bitcoin miner is an electric heater that happens to pay you.

I have been in mining for seven years, starting with off-grid flare gas mitigation, and I still feel like I understand a fraction of it. The part I keep coming back to is hash rate ownership.

The dominant conversation right now is treasury companies and nation state adoption and what the administration thinks. That is interesting, and to a point it matters. But Bitcoin is a distributed protocol that has to stay sufficiently distributed for any of the rest of it to mean anything. Watching someone do the unglamorous work on that problem is more encouraging to me than another balance sheet announcement.

Key takeaways

• Every watt a miner draws becomes heat. A miner is resistive heating with a protocol attached that pays you for the electrons you were going to spend anyway.
• The arbitrage is against stored fuel, not natural gas. Propane and heating oil run three to six times the cost per unit of energy that piped gas does. That gap is where the economics live.
• The addressable load dwarfs Bitcoin mining. Stevens puts US furnaces and boilers alone at five zettahash if it all ran on hash rate.
• The bottleneck moved from engineering to plumbing. The hardware works. What is missing is tradesmen who know how to size and install it.
• You are paying for heat regardless. That is what makes this different from buying a miner to run a business.

A miner is a heater with a protocol attached

Stevens credits his friend Rob Warren with the cleanest framing of proof of work: Satoshi invented a way to make a computer sweat.

Energy cannot be created or destroyed. Push electricity through an ASIC and effectively all of it comes out as heat, because there is no screen and nowhere else for it to go. Physically this is identical to a resistive heater running current through nickel chromium wire. The only difference is the path the electrons take, and the fact that a network of other computers pays you when your path happens to find a block.

The misconception Stevens hears most is that a mining heater splits its power between two jobs, some to heat and some to mine. It does not. It is the same energy doing both.

What makes Bitcoin specifically suited to this is granularity. You cannot break an AI data center into pieces and make one of them your water heater, because those workloads need uptime and coordination. Mining rewards are proportional to the electricity you put in and the machine does exactly one thing, so you can switch it on when the room is cold and off when it is not. It behaves like a heater because you can treat it like one.

Where the money actually is

The economics do not work against cheap piped natural gas, and Stevens does not pretend otherwise. Gas runs around two cents per kilowatt hour equivalent in the US.

They work against stored fuel. Propane and heating oil land between eleven and thirteen cents on the same basis. If your electricity is fifteen cents and you are also earning Bitcoin against it, the arithmetic changes. In the high altitude Colorado towns near him, where there is no gas line through the granite and everyone burns propane, Stevens says savings of eighty to a hundred percent on heating are available once the Bitcoin rebate is counted.

He calls it a rebate deliberately. Heat reuse, to him, describes a mining operation that sells warmth to a greenhouse when it suits them. This is the inverse: a heating operation whose one commandment is heat, where Bitcoin arrives as an offset.

The clearest example he gave was a hot springs resort in the mountains, 40,000 gallons of pool kept at hot tub temperature, geothermally warmed and topped up with propane. They spend $22,000 a year on propane and have thirteen cent power. Switching the supplemental load to four Auradine hydro units would raise the electricity bill to roughly $24,000, and those machines would mine about $27,000 of Bitcoin a year running only when heat is actually called for. A business paying $22,000 a year to stay warm ends up roughly $3,000 ahead, with the hardware paying for itself in about thirty months.

The scale is the part that surprises people

Roughly half the world's energy goes to heat. About half of that is what the industry calls comfort heat, meaning space heating, water heating, keeping buildings and pools warm. Comfort heat matters because it sits in the temperature band a miner can actually deliver. These machines have no flames in them.

Stevens ran the numbers for the US specifically. Furnaces and boilers alone would be five zettahash if they ran on hash rate. Water heaters would add 2.3. Propane and heating oil, the segment where the savings are largest, would be another 1.5.

His aside on this is worth repeating: the energy spent making heat for people is something like 250 times what Bitcoin mining consumes. The people calling mining wasteful are, on his framing, ignoring an energy sink hundreds of times larger that produces nothing but warmth.

The hardware finally works, which was not true two years ago

Ask Stevens about this eighteen months ago and he would have told you the hardware was bad and the firmware was worse. Everyone doing it was strapping S19s into ductwork, and everyone doing it was already a Bitcoiner who enjoyed tinkering.

What changed is that efficiency gains between ASIC generations have flattened. During the race from 2019 to 2022, every new machine was far enough ahead that nothing else was worth optimizing for, which is a decent explanation for a decade of indifference to customers from the major manufacturers. With that race slowing, attention has moved to form factor and control.

At The Space, the Denver community building Stevens runs, the demo system is a single MicroBT M64 hydro unit. He describes it as the only single phase hydro miner available, which matters because single phase is what US homes and small businesses have. European installers have had a dozen three phase options for years.

The unit pushes 80°C water into PEX tubing a plumber ran through the exposed floor joists, and it heats roughly three thousand square feet on its own. People underestimate how power dense these machines are.

The install also produced the kind of detail you only get from doing it. Stevens asked the plumber for a mixing valve and got it fitted backwards, because gas boilers need protection from water that is too cold and the plumber had never encountered the opposite problem. Miner output at 80°C will cook the adhesive under laminate flooring. It has to come down to about 120°F.

On the software side the gap is control. Traditional heating is on or off, because you cannot run a flame at thirty percent. A miner is a computer and can sit at any throttle you like, which should make it a better heater rather than merely a cheaper one.

Braiins has dynamic performance scaling and Luxor has automatic thermal management, but both were built to stop machines cooking themselves rather than to hold a room at a set temperature. As Stevens puts it, you end up commanding your heater by how close the furnace is to blowing itself up. Auradine's machines retune inside thirty seconds, which is the responsiveness this needs. What is still missing is a miner that will talk to the thermostat on your wall.

The real bottleneck is the trades

Nobody chooses their own boiler. You call the heating guy and he tells you what you need.

That single fact is why Stevens thinks this has stalled short of its potential. Space heaters from Heatbit, Canaan and 21 Energy have proved people want hash rate in their homes, but a space heater is a gadget you buy. A furnace is a system somebody installs. Going from one to the other turns an engineering problem into a distribution problem, and Exergy is currently the translation layer, taking a customer's bills and building size and working out which machines fit.

His competitive set is not other Bitcoin companies. It is Rheem, Carrier, Bosch, and the tens of thousands of distributors who sell their equipment.

The incentive he keeps circling is hash rate splits for installers. Give a plumber one or two percent of the hash rate from every system they put in, and a trade that currently waits for your boiler to break so it can charge you to fix it starts monitoring a fleet it wants running perfectly.

These are computers with sensors and firmware, so they can flag a fault before the homeowner notices. That inverts the entire service relationship, and it is doable today.

Commercial work throws up questions the industrial mining world never had to answer. Are these machines UL certified? Some are not.

Do you insure the installation as electric heating or as a Bitcoin mining operation? The premium is eight times higher if you call it mining, for hardware that is physically doing the same thing either way.

Where to start, based on your situation

Your situation	Worth exploring?	What to look at
Propane or heating oil, cold climate	Strongest case	The fuel cost gap is the whole argument. Get a heat audit before buying anything.
Cheap piped natural gas	Marginal on cost alone	Solar generation or a sovereignty motive has to carry it.
Curious but not ready to replace a system	Start small	A space heater or a Bitaxe teaches you the behaviour without touching your plumbing.
Milder climate, low altitude	Look at a heat pump first	Around 300% efficiency by moving heat rather than making it. A miner can serve as the resistive backup.
Running a pool, hot tub or small commercial load	Very strong case	Large steady heat demand is what makes the payback maths work.

Why not just buy Bitcoin

This is the pushback Stevens gets constantly, and it deserves a straight answer rather than a dodge.

Mining hardware is expensive and the network gets harder every year, so a machine bought as an investment is a depreciating bet. But he is not running an operation that needs to be profitable. He is offsetting a cost he was going to pay anyway.

That is the entire distinction. Your heating bill is not optional, and last generation hardware is perfectly adequate when your margin requirement is zero. Modular hashboards make the upgrade path cheaper still.

There is a secondary argument he makes that I think is underrated: your utility becomes your exchange. No KYC, no account, just sats arriving because you kept your house warm.

I run Bitaxes at home and a piece of that is wanting to manifest a solo block. That is a lottery ticket I enjoy holding rather than an economic case.

What Stevens is describing stands up without any appeal to altruism or luck. You cannot scale a network on people plugging things in out of goodwill. You scale it on people saving money.

Why this matters more than another treasury company

I am skeptical that most of the publicly traded Bitcoin treasury companies work out. It has the feel of something too good to be true, and one of them recently published a release conceding that publicly traded mining is not a capital accretive way to grow sats per share. It is capital destructive.

Set that against what Stevens is describing. A heating company that happens to mine, with hash rate splits aligning everyone involved, running on an electricity bill the customer was always going to pay. No investor relations, no marketing department, no exposure to hash price. Heating systems last thirty years and you swap the hashboard every five.

There is a version of this where a meaningful share of Bitcoin transactions get processed quietly inside heating systems managed by plumbers and HVAC technicians who never think about the price. Stevens calls hash rate heating the perfect Trojan horse, hiding network security inside a sunk cost nobody stops paying. Cypherpunk to heat punk, as he puts it, and mining coming back to where it started.

The New York bathhouse got written up in Time and various local outlets for heating its pools with miners, mostly disapprovingly. They lowered their costs and used their electricity twice. That story will read very differently in five years.

What to do this week

If you are burning propane or heating oil somewhere cold, get a heat audit and find out what the gap actually is for your building. Exergy charges a service fee for this today and wants to make it free once the products stop needing a translator.

If you want to see what other people have built before spending anything, heatpunks.org is the forum Stevens set up for exactly that, and people post real projects with real numbers.

If you are in the trades, this is the least crowded edge available to you. Thousands of HVAC and plumbing companies sell identical services and compete on who calls you about your air filter.

Related reading

• Bitcoin Home Mining Playbook with Exergy. Stevens and Dylan Seib on candidate buildings, controls and hardware selection.
• Bitcoin Mining Revolutionizing Home Heating with Alex Busarov. The Heatbit founder on the consumer appliance route.
• The Bitaxe, Explained. The cheapest way to get hash rate into your house.
• Is Bitcoin Mining Worth It in 2026?. Home versus hosted, and whether you want mining exposure at all.
• Best Bitcoin Mining Hosting Companies. If you want hash rate without hardware in your house.

About Tyler Stevens

Tyler Stevens is the founder of Exergy, a Bitcoin heating company he started in April with Mike Clear and Dylan Seib, and CEO of The Space, an 82 member Bitcoin community building in Denver. He trained as a mechanical engineer and worked on thermal systems at United Launch Alliance, the Lockheed and Boeing joint venture, before leaving in January 2024. He wrote Bitcoin Mining Heat Reuse with Braiins and founded the heat punk community forum.

Sources

• Exergy, exergyheat.com
• Heat Punks community forum, heatpunks.org
• Braiins, dynamic performance scaling
• The 256 Foundation, open source mining stack

Watch the conversation

Timestamps

• 0:00 - Intro
• 0:30 - Tyler's background
• 7:01 - ASIC race and the return to consumer applications
• 11:28 - Making computers sweat
• 17:52 - Target markets for hash rate heating
• 20:10 - Total addressable market
• 22:37 - Demo of the heat system at The Space
• 29:48 - Bitcoin rebate mechanics
• 35:48 - Pool distribution and market timing
• 50:48 - Smart heaters
• 56:20 - Exergy's vision
• 1:05:01 - How to get started

Sponsors

• Bitkey: Bitcoin self-custody hardware that natively embeds into a two-of-three multisig, one key on the device, one on your phone, one held by Block. Code TFTC20 for 20% off. bitkey.world
• Unchained: Collaborative custody and Bitcoin financial services. Code TFTC for $50 off any of their services. unchained.com/tftc
• Obscura: A VPN that cannot log your traffic by design, with Lightning payments and no personal information required to sign up. Code TFTC25 for 25% off an annual subscription. obscura.net
• Opportunity Cost: Our open-source browser extension that converts prices you browse into Bitcoin and sats. opportunitycost.app