We Supercharged the Ink Boiler 🦑🔥 | Wind Down Wednesday

Last week, we proved the ink boiler could work. This week, we found out just how far we could push it. The original design successfully converted Glo Squid ink and mucin using waste heat recovered from our cobalt volcano tamer—but its throughput was limited. Tonight, we combined that thermal-buffer design with the in-pipe boiling technique we learned from Stinger and turned the little prototype into an industrial machine. The core principle stayed the same: hot cobalt enters a crude-oil thermal buffer, while mechanized airlocks let us control how much heat reaches the boiler. But instead of slowly feeding ink into a conversion chamber, we ran it through radiant liquid pipes in 1 kg/s packets and allowed the phase change to happen directly inside the pipes. Getting there involved plenty of engineering nonsense. We broke pipes. We overheated things. We built far more capacity than we actually needed. And at one point we accidentally dumped hot metal into our cooling section and worked the aquatuner hard enough to overheat the steam chamber. But eventually, it worked. Along the way, the Workshop helped make the design substantially better: • Stinger introduced us to the in-pipe boiling approach and taught us a rocket-battery trick that completely changed how I look at battery stacks. • Xaeric joined The Workshop and shared a beautifully simple circulating conveyor setup. As new hot metal enters the loop, cooler material is automatically displaced and allowed to leave—eliminating some of the sensing and control hardware from our original design. • Redneck caught an important thermal-management lesson that I completely missed during the stream: more steam mass in the steam chamber would help absorb and damp the enormous temperature swing when fresh volcano metal enters the system. Then we finally stopped asking, “How many pipes can we build?” and asked the much better engineering question: *How many pipes do we actually need?* Eight happy Glo Squids produce roughly 400 kg of ink every cycle, or about 0.67 kg/s. That means: • One ranch ≈ 0.67 kg/s • Three full ranches ≈ 2 kg/s • Two 1 kg/s boiler lines can process the output of approximately 24 Glo Squids Our ridiculous ten-line test system therefore had enough theoretical capacity for roughly *15 full eight-squid ranches.* Slightly overbuilt. 😂 The good news is that the experiment gave us everything we needed. The next revision can be dramatically smaller—probably only two processing lines for the scale we're actually planning to use. We also started following the resource chain downstream. The slime produced by the boiler can be converted into phyto oil, but we discovered another important constraint: the slime must be cooled before that process or it changes state inside the pipes. That gives us our next challenge. Can we take this working boiler, recover its byproducts, produce phyto oil, generate the required bleach stone, and eventually close the loop all the way to biodiesel? The Workshop is starting to live up to its name. 🔧 Join The Workshop on Discord:   / discord   More Oxygen Not Included experiments, ranch designs, automation, and engineering coming soon. Take care of yourselves. Take care of each other. And leave things better than you found them.