The founders of Bitheat.io seek to build a new kind of bitcoin miner that brings back profitability to home-based mining. The water heating miner, PipeMiner, also could also make bitcoin mining more environmentally sustainable. AllCoinsNews got the opportunity to talk tothe two inventors of the PipeMiner, Iowan brothers Julian and Harrison Stahl.
ACN: How long have the co-founders/brothers been involved in bitcoin ecosystem?
(Julian) We got into Bitcoin during the $30 spike back in 2011. I (Julian) had stumbled upon an article about some guy using his university computers to make $20/day in Bitcoin. I sent it to Harrison and asked if we should get in on this. The next day we did some cursory research, read the white paper, and aired in 8 AMD HD6870. We continued to mine for about a year or so and kept up with Bitcoin news. Towards the end of the profitable life of the GPUs we looked for ways to extend our investment by reusing the waste heat to offset the electricity cost. We had several ideas but the cost and engineering of implementing those heat recovery solutions would have never returned on that investment, so we had to shut them down.
Once the price started taking off in early 2013 we knew Bitcoin was here to stay and invested what money we could into it. Since then we have been very watchful of developments, waiting for an opportunity to contribute back to the system.
With debates around the block size and other potential threats all centering around the common thread of mining centralization, and with no easy software solution in sight, we looked at a way to engineer around the problem. And here we are today. Doing what we can to help the system.
ACN: What other types of services or applications have you been involved in?
(Harrison) Well before we got into Bitcoin, we used to have pretty regular LAN parties, so we had extra computers around and decided to contribute some of those spare cycles to the BOINC distributed science computing platform. I (Harrison) even convinced a few friends to join our computing team to help get us higher on the leaderboards. Doing this for a few years and being able to see the difference in hardware efficiency for different computing gave us an intuitive understanding of the general concept of distributed computing. This understanding is what allowed us to only take a day to decide to throw money at some GPUs and get involved with Bitcoin.
ACN: What is the background of both founders?
(Julian) We are brothers, and we have worked alongside each other in our family business for our entire lives, immersed in a world of customers, business, financing, and manufacturing. We are primarily self-taught; between the internet, good old fashioned textbooks, and years of hands-on experimentation, we have usually been able to learn what we need when we need it.
My background is mostly mechanical. I taught myself how to TIG weld when I was 13, and have since learned the basics of most types of manufacturing. I put these skills to work in our family’s machine shop, where I was allowed the freedom to experiment and try new ways of doing things. Having a background in a small business setting, you end up having to learn a little bit of everything, from cost accounting to customer support.
(Harrison) Harrison- My background is on more the digital side of things, but still know my way around the shop from the years working alongside my brother. I cut my tech skill teeth when I was 12 modifying my first gaming computer to accommodate the needs of our video production business. We made instructional and motorsport event DVDs, using our parent’s mail order catalog business as our distribution channel, as we were both still in middle school. In the 11 years since I’ve done everything from designing circuit boards and messing with 8bit microcontrollers up to web design and using Docker to deploy services and everything between those extremes.
ACN: Can you describe the Bitheat miner/water heater, its makeup and how it works for our readers?
(Julian) The PipeMiner is intended to replace an electric heating element in a standard (US, for now) electric water heater. This means you won’t have to go and buy a new water heater if you already have an electric one and want to use our miner. We are using 3M’s Novec fluid to transfer the heat from the circuit board out to the pipe, and then into the water. The fluid allows the ultra high power densities needed to get the size small enough to allow for a retrofit. Being a sealed unit, we can purge the atmosphere from the pipe before sealing it, which helps with reliability. The ASICs, their 12v->Vcore regulators, and required sensors are what is actually in the tube. The head unit, a BeagleBone Green as of now, and an ATX compatible 12V connector will be outside of the unit. This is where you’ll need to plug in your own PSU. The only heat that isn’t recovered is that produced by the PSU (and head unit), usually around 10% loss.
The target wattage for this miner is 150-400w inside the pipe. A normal heating element is around 4500w. We are going to make up the difference for the lack of recovery time by increasing the thermal capacity of the tank. We plan to increase the tank temperature to around 70C, whereas a normal tank is around 40-50C (this will also require the installation of a mixer valve for safety). Our miner will not interfere with the top element in the water heater, so it can be used during a high demand period to avoid running out of hot water. By doing this we can get as close to 100% hashing time as possible. If you don’t use much hot water on a particular day, the miner will shut off when it is up to temp, which I guess is a good incentive for taking a daily shower. The ratio of energy storage and hot water demand variability will determine how well the miner is utilized. We intend for it to be a set-it-and-forget-it device that will try and optimize itself to usage patterns.
The aim is that the user can monetize energy they will be spending for heat anyway. Therefore, the effective electricity cost for mining is the percent you are not recapturing. Using a platinum supply, you would get a ~93% discount on your electricity rate to run the miner, as that electricity would have gone to heat even if you didn’t have our device.
ACN: I saw Julian on Linkedin that you have separate machinist/CNC and CAD qualifications, as well as experience. How has this played a role in designing your prototype?
I have designed many products ranging from complete small internal combustion engines, to converting a vehicle from gas to electric. The CAD tools are very powerful for prototyping, as they can let you see inside the design and make internal measurements you can’t do in real life. The CNC machining made for quick work in making the prototype parts. By being the designer and manufacturer you can quickly iterate, and eventually end up with a functional design that is easy to make.
ACN: Do you have any background in computer circuity, chips or heatsink design?
(Harrison) I did all of the electronics design, circuit layout, and firmware on our first electronics product, the TEC Tach, a solid state RPM gauge that uses LEDs instead of a mechanical needle. I had started on another version of the tach aimed at a larger market, but our initial results from the PipeMiner thermal test rig were so promising that we have since shelved the tach to focus on fully developing the prototype PipeMiner.
ACN: How much testing have you done with your prototype?
(Harrison) We’ve verified the heat transfer characteristics of the Novec fluid and the heatsink wand with our test rig. We made a second rig to do pressure testing of the fluid at different power levels and of the fluid’s effects on standard components to make sure we chose compatible components for the circuit prototypes. We are now testing the power converter and mechanical assembly that you see in the pictures. After verifying our converter design for reliability we will be stuck waiting for ASICs to be able to have a completely functional prototype. Until then we’ll just be using resistive loads for all of our testing.
ACN: Have you applied for a patent for you design?
(Julian) We do not intend to, unless it is required to get investment to make this project a reality. We would like to be as open with the community as possible on our product, because our goal is the decentralization of mining. The PipeMiner is something very different and we want to make sure people fully understand it. If a mining company that already has (current gen) chips would like to partner with us, we would be more than happy to share information to see this project come to fruition.
ACN: Are you looking for angel investors or venture capital to help fund your proof of concept and implementation?
(Julian) We have applied to Boost VC, an accelerator focused on blockchain and VR companies, and are hoping to get some traction from that to help complete our design. We have gotten as far as we can on our own, and are now primarily looking for community support and feedback on our idea. We see investment as an unfortunate necessity when companies like 21inc have a $1 million minimum order just get them to talk to you — gone are the days of pre-order funded mining companies.