A Year in the life of a Bitcoin Miner：How to Make Money by Mining Bitcoin at a Power Station

From this:

To this:

What’s it like to build and operate a bitcoin mine?

Perhaps you’ve put together your first NerdMiner, a larger GPU mining rig, a Bitaxe or you may have spun up an old S9 or S19 ASIC in your garage.

Maybe you’ve put together your first few racks of miners and you feel ready to graduate to the next level of scale and complexity.

Now that you’re hooked on becoming a full time bitcoin miner, what comes next?

Let’s use a recent project we’ve built at mineracks to show you through the steps you can expect: from finding a business problem to solve, the first concept and design through to manufacturing, installation on site, practical completion and then the daily challenges of operating a bitcoin mine efficiently and profitably.

The Project

We partnered with an energy company to demonstrate the viability of generating additional revenue from bitcoin mining at a power station during times when the spot price for power goes negative or when bitcoin mining pays better than selling power on the national energy market (NEM).

This solves a business problem that many power stations that use gas fuelled generators face. You can’t easily turn gas generators on or off — there is a whole start up and shutdown procedure that takes a lot of time and expertise to do well without shortening the life of your equipment.

This means potentially running your generator at a loss during some times of the day unless you can quickly redirect the energy you produce away from the grid and towards a customer who will pay more for the power when no one else wants it.

The example above shows the Queensland energy grid spot price during a sunny afternoon weekday. With wind blowing on our turbines and sun shining on our solar panels, the price the grid will pay for additional power you produce at this time is only $0.002 per kWh — that’s not much! Meanwhile, an energy consumer on the same grid is paying $0.25 per kWh or 125x more.

The forecast shows that 30 minutes later, there’s so much power being generated relative to demand that you’ll actually get charged $0.026 per kWh to put your power onto the grid, creating a strong incentive to turn the generator off or to find another customer willing to pay more.

Fortunately, as we know, Bitcoin mines will happily always pay something for power. It was our job to find out how much more and to prove the business case by deploying a container load of bitcoin mines at the power station.

Selecting a Container

We started off by sizing up the different types of containers and mines we could put on the site.

We looked at whether it made more sense to go with air cooled, liquid immersion or water cooled mines.

In the end, the liquid immersion cooling container turned out to be the best option for the site given that we wanted minimal maintenance (air cooled and water cooling both require a lot more hands on attention to keep running).

Fabrication

With drawings accepted, we got underway. We selected a container manufacturer with an excellent track record（FogHashing）.

Simulataneously, we started work on preparing the site and realised that we did not have sufficient access to town water for the design of our intended cooling unit.

So we quickly made the decision to change over from a water cooling tower unit and instead selected a dry cooler. Dry coolers can cool down a container without using any external supply of water, it just needs additional electricity to power pumps and fans.

It took about 25 days for the factory to complete construction — right on schedule.

Testing

Because flying to the factory in China for progress inspections wasn’t practical at the time, we figured out a way we could run remote acceptance testing of the container’s components over HD video calls.

During these calls, we realised that the internal wiring of our switchboard — though not visible to anyone using the container — was not cabled with the correct coloured sheathing for Australian wiring standards.

We had to make the difficult call to open up the switchboard and re-do all the wiring with the correct colours — fortunately this didn’t delay our overall schedule and we made it onto the Brisbane bound ship in port on time for the month long sea journey.

Shipping & Groundworks

Meanwhile back at the power station, works were underway to prepare the ground, cable ducts, foundations and footings for the container and dry cooler.

We hoped to have all these works completed in time for the container’s arrival.

After a long but safe 26-day sea voyage, our shipment cleared customs and we received delivery of our equipment at our staging area in Brisbane.

Unpacking the tight fitting components inside the 40” container our shipment came in required a variety of heavy forklift equipment and some experienced logistics.

Once transported to site, we needed to coordinate the hire of a mobile crane to lift the 6 ton of equipment onto the concrete footings and precisely control the final placement.

Install, Commissioning, Energising

With all the power and internet cabling now hooked up and connected to the container, the next step was to check those connections with a test mine.

Now that the container was in place and all services connected and validated, it was time to purchase the mining machines and get them installed.

Of course on the day of installation, the weather decided to make life more interesting. With a liquid immersion setup, we needed absolutely no water contamination, not even a drop, on any of the fluid lines or immersion tanks.

Fortunately nothing could dampen our enthusiasm for getting these machines hashing just before a bitcoin halving — so with an abundance of tarps, spare hair dryers and wet weather gear, we got to work unpacking the mines.

The first step was to turn on and test each mine while also configuring each mine to our mining pool, then power off.

The mines we purchased were normally designed to be run in air rather than in liquid immersion. So we had to strip away all the fans that would not be needed and prepare the mines for going into the immersion tanks. Next we had to cable up each mine with power and network cables.

After the weather cleared up, we were finally able to connect the fluid lines between the container and dry cooler and then bleed off the excess air in the dry cooler side fluid lines (for safety these are pressurised during transport). We were at last ready to fill the immersion tanks.

With everything setup, we turned on the pumps and fans for the dry cooler, and pumped a bit more fluid in for the extra volume needed inside the dry cooler. We knew we got this right when the fluid levels stabilised and the pumps had been running for a while.

Next we turned on banks of mines one by one and started seeing our first hash rate come online. We were hashing!

Transition to Operations

With the container running and all setup, it was time to put everything in place to make it as smooth as possible to remotely manage and monitor the operations.

While waiting for an industrial PC to arrive, we setup temporary monitoring and remote administration software on a laptop installed inside the container. Now from anywhere in the world, we can make sure the container is still producing uninterrupted hash rate while also doing routine maintenance tasks.

We can update the firmware running on each mine, reboot mines, change mining pools, move up or down our power consumption and hash rate in response to changing energy market signals, etc.

Here’s a glimpse inside the completed container:

We also needed to carefully look after the security of the bitcoin we mined. Fortunately at mineracks we take this really seriously and were able to leverage our stock of SeedSigner, SeedSleeve and SeedHammer products to secure our bitcoin in multisig offline cold storage.

A multisig key inside the SeedSleeve tamper evident protective enclosure

Mission Accomplished?

So with the container operational at the power station for several months, what conclusion did we reach?

Had we succeeded in proving the project objective that adding bitcoin mining to a power station could improve the bottom line?

We compared 3 different scenarios. Scenario 1 was using the power station to power our bitcoin miners 100% of the time, essentially disconnecting from the energy grid. Scenario 2 was to ignore the option of mining bitcoin altogether and just sell on the spot market, including at times when grid prices go negative or less than can otherwise be earned by mining bitcoin.

In the 3rd scenario, we only power on our bitcoin mines when the spot price is less than what we can make with bitcoin and here we go: an 18% boost to the power station’s normal revenue.

Also we found that on average 61% of the time the power station could earn more from bitcoin mining than from spot prices.

But what about the environment? By running bitcoin mines instead of powering homes, wouldn’t the power station be harming our planet? If you want to have this misconception thoroughly challenged and conclusively debunked, my friend Daniel Batten has you covered.

Our Call to Your Action:

Do you know of a power station who’s struggling to sell 100% of their power 100% of the time?

If you do, please consider asking them to learn more about bitcoin mining and how it can change the economics of their business and in turn contribute to lowering power costs for all of us.

Do you need help with your bitcoin mining project? Fog Hashing is a leading liquid-cooling solution provider for the computing-intensive applications. By utilizing advanced immersion cooling technology to achieve high-density thermal management for Blockchain and AI computing, it makes HPC infrastructure more efficient, robust, and sustainable.

Reach out to us today and we’ll be happy to help.