A simple Gunk > NatGas boiler

[6Havok9]

Hello dupes and dupettes,

in my current colony there's a volcano one screen away from the printing pod, with an aetn directly under it. Since there are no natural gas geysers, I devised this little contraption whose main purpose is to provide gas for the kitchen.

It's probably nothing new for the most seasoned printing pod AIs around here, but I had great fun while designing it, and I've never seen anything similar.

 

So, here's the thing. I had to deal with severe space constraints, but I'm quite happy with the result.

It uses flaking mechanics. The heat source, a rockgas spewing volcano, is on the top right. A wolframite door injects heat into 3 tons of steam, currently it keeps it at above 950 °C to process 400 grams per second of gunk, which is 12 bionic dupes. Cobalt tempshift plates (diamond should work too) inject heat into two insulated igneous rock tiles: the bottom one flakes gunk, the top one flakes petroleum. The top tile has to do much more work than the bottom tile, but conveniently it interacts with all 3 tempshift plates. Flaking gunk avoids creating sulfur, and we get 5 kg of petroleum.

The curious thing is that while the gunk tile must equal or exceed 5010 g to flake, the newly created petroleum does not, and gets flaked instantly by the upper insulated igneous tile. While this screenshot captures the exact moment petroleum is created, petroleum does not linger. All of the mass is converted with no losses.

 

It does not use beads: since the drop is very short, sometimes they cause minor mass deletion when petroleum and gunk tiles swap, which is undesirable. But more importantly, in this case they cause MASSIVE heat deletion, which is unacceptable. This puts too much strain on the tiles in charge of flaking, draining their energy too fast and wasting precious volcano heat.

Also, using beads causes petroleum to flake one tile above the one I'm using, due to fluid swapping tomfoolery. That tile is in contact with just 2 tempshift plates, greatly reducing its flaking capacity. Furthermore, mass deletion inevitably leads to some residual petroleum in the system.

 

The gold amalgam door below the mafic rock tile cools down the gunk in case it gets too close to the boiling temperature, I set it to 430 c. While it sometimes kicked in when processing 200 g/sec, I've never seen it activate at 400 g/sec. I don't know why for sure, but it seems quite self regulating and the gunk sits at 414/426 °C.

It uses very little heat energy: heating gunk from 37 °C to petroleum then sour gas is very, very heat multiplying.

Sour gas then goes through an heat exchanger and is condensed by the aluminium spike, kept cool by the aetn entombed in brine ice.

I didn't add a conveyor bridge by the condensing tile because it condenses the sour gas too quickly for rounding errors to not matter, but not quickly enough to avoid sulfur creation altogether. This results in a lower natural gas (and sulfur) output than expected. Edit: due to the small quantity of gas involved, I think It might be possible to completely avoid sulfur creation by using an extremely long heat exchanger/sour gas tube. While not possible here, it would definitely not be space efficient.

Sulfur is extracted at a rate of 132 g/sec to help with cooling and comes out at around 47 °C from the middle tile of the heat exchanger, the same spot where gunk enters the system. A quite reliable average of 268 g/sec of natural gas goes up to be pumped out.

If processing 200 g/sec max, granite insulated tiles and aluminium tempshift plates work, with steam at around 610 °C. However I wanted to future-proof it and allow for some freedom of experimentation.

Some (not so) useful overlays in the spoilers.

Spoiler

Conduction panels in the steam room are cobalt, rest are aluminium. I should change the input pipes to granite, to reclaim some aluminium. The pipe segment in the aluminium tile on the left should be a radiant pipe, not an insulated one... oh well. Luckily everything can be accessed by dupes.

The timer sensor has no use atm and is gonna go. The autosweeper activates when there is no sulfur on the rail just before the conveyor meter.

The upper thermo sensor prevents the door from closing if the magma gets too cold. If that infinite storage breaks, it's over. I really should link it to the gunk valve too...

The pixel pack looks cool and helps distribute the heat from the sour gas.

Much love and cya!

[psusi]

Wait, what?  I could have sworn that flaking only happens to SOLID tiles and makes them melt off a 5 kg drop of liquid.  Not for any other phase transitions.

[Zarquan]

Any upward phase transition can flake.  Liquid to gas and liquid to liquid are quite useful.

[6Havok9]

On 5/2/2025 at 12:53 AM, psusi said:

Wait, what?  I could have sworn that flaking only happens to SOLID tiles and makes them melt off a 5 kg drop of liquid.  Not for any other phase transitions.

A liquid tile with mass above 5010 g can flake.

What's weird here is that the resulting petroleum tile is exactly 5 kg, but it still flakes.

I tried to break the system with a mechanized airlock above the gunk tile, to accumulate gunk. Then I dropped the gunk blob. It still manages to very quickly flake multiple 5kg petroleum packets from the gunk tile below, and the petroleum tile flakes again.

It finally breaks when the top tile doesn't have enough heat energy to flake the petroleum tile almost instantly. This causes the petroleum tile to stay for long enough that the gunk tile below won't flake anymore, since it now considers itself "blocked" by the petroleum above. I guess. And the process does not restart once the topmost insulated tile has enough energy, because the petroleum tile is only 5 kg, and not 5010 g or more. Why? I don't know.

Ironically, the mechanized airlock used to break the system also allows it to restart, by closing it and pushing petroleum up. The petroleum gets deleted until there's only gunk, which can be dropped and the process starts again.

It's weird. 

Riedit: I've seen all sorts of outcomes now and they don't make sense, or better, I can't figure out why.

- 5 kg of petroleum remain and never flake (ok, but why did it work before)

- 5 kg of petroleum remain, but flake once the tile is hot enough (ok, but it's not supposed to. Still, it resumes "normal" operation)

- 5/25 kg of petroleum appear, flake but some grams remain. (Why?) Gunk tile then refuses to flake. Forever.

- 5/25 kg of petroleum appear, flake but some grams remain. Gunk tile goes on flaking like nothing matters, with petroleum on top. Residual petroleum mass and residual gunk mass start to vary, not by much. Noticeable after some operations.

I can only attribute the last two to rounding errors and simulation lag, while the first two.. it's just lag? Maybe?

Whatever. I'll stick with "if It works, don't try to break it".. for now.