TL;DR: I built a new pipeless counterflow heat exchanger that can handle at least 100 kg/s crude oil in a more compact way. It is very powerful, but very fragile, build with care.
I like my pipeless counterflow heat exchangers. Every colony I make has one. I have been using version 2 with minor improvements until now. Recently, I learned about 2 tile wide escher waterfalls, so now I can make them smaller, which calls for a new version. And this version is completely duplicant serviceable, so I can add more thermal conductivity to improve the design as materials become more plentiful.
The major idea behind this design is that crude oil flows up the counterflow heat exchanger up to the boiler (which is currently a 50 kg/s flaker boiler with space for a second one) and then the petroleum flows down the same path the crude oil flowed up. Liquid to liquid thermal conductivity as a 625 times multiplier, so direct contact is helpful.
I built and started up this system in survival mode. This time, I will include the priming process, as this is significantly more involved than the old version.
Spoiler
First, I built a structure like this:
The metal tiles are aluminum. On this map, I was fortunate enough to start with an iron volcano early, so I made the airflow tiles out of my most plentiful metal, steel. The insulated tiles are igneous rock, but they could be replaced with a better material (that I don’t currently have). But none of the internal insulated tiles will be in contact with more than one liquid, so it shouldn’t matter.
A few things to note that can be hard to see: The crude oil is coming in to start priming on the inlet on the far left. This is where all the crude oil will enter from. The petroleum will gather in the large airflow tile chamber next to it. To the right of that is the boiler, which has two boiler plates connected with 2000 kg steam inside heated to (currently) 530 C. Once hot crude oil starts flowing in, I will turn it down to a temperature closer to 420-430 C. This is connected with conveyor bridges and temp shift plate to a steel door (actually 2 mini-doors to negate the door heat bug) to a volcano heated heat source set to 900 C. This heat source could be replaces with thermium aquatuners.
There must be a large amount (30,625 kg or 35 tiles) of crude oil to prime this machine, as it must fill half of each pit and enough to push the crude oil up in to the boiler.
This is a repeating modular design, and each module looks like this (at various stages of priming on the crude oil side).
Every bridge has a pipe on its inlet, and it will be given 100 mg carbon dioxide and 100 mg natural gas from the contraption on the far right of the first picture. I use 100 mg because it is the smallest quantity a valve can produce and it negates any backward heat flow (as <1 g gas does not conduct heat). Ensure on the lower pipe on upper escher waterfall has carbon dioxide in it.
I use a pipe to deliver the gas. When one of the pits becomes full of crude oil (meaning all 4 tiles have crude oil and it is pouring to the right), I break the top pipe then the bottom pipe, which causes an escher waterfall, making the crude oil to flow to the right.
While I waited for my honestly at this point slow crude oil source to prime the system, I added an important drop of naphtha to the petroleum dropoff at the top, that will stop the crude oil from flowing down the petroleum’s path and will enable a minor benefit for the petroleum later.
The bottom layer has been primed, and the crude oil is being forces up to the top layer. I also added a tile over the naphtha for now so that the crude oil can’t overflow and pour down that I will get rid of once crude oil can’t overflow there.
At this point, you will need to put in airflow tiles in certain locations so that the petroleum will not liquid-bypass the escher waterfalls.
One thing to note is that when the crude oil poured over in to the first upper module, it did something weird to the naphtha that clogged the system. If this happens to you, you will have to remove the naphtha through mopping without removing all the crude oil or it will happen again. You could also brick off the naphtha where the mesh tile is.
Once all the crude oil escher waterfalls are done, the setup should look something like this:
I put in a door to block the crude oil so I could have time to add a bit of petroleum to the ends of the boiler to prevent non-crude oil from leaving. Unfortunately, the crude oil appears to be too cold to flake initially, so I have to wait until it heats up.
I got impatient and increased the heat plate to a dangerous 580 C and added an 4x automation bridge to increase heat flow. Eventually, it began flaking at about 165 C.
Note that there is a corner bypass mechanic involving the same gases going across corners. When making the petroleum escher waterfalls, break CO2 first, even if it is on top. If it is, rebuild pipe to move natural gas to top or you will break the escher waterfall. This could be solved by using a different gas between the top and bottom waterfalls. Once the gases are in place, you can remove the airflow tiles and the petroleum should flow without incident.
I also added petroleum manually to the end turnaround area at the top right.
I had to redo the end of the system, as there was a bypass issue.
After that, I scrounged up some more aluminum and built temp shift plates and cleaned it up.
When it was first completed, it looked like this:
The flow is as follows:
Crude oil is poured in on the left.
It then falls through a series of escher waterfalls to the right, heating up as it comes across hotter and hotter petroleum.
When it gets to the far left, the escher waterfall pushes the crude oil up and it goes left until it hits the vertical passage directly to the right of the petroleum tank.
The crude oil is then pushed up in to the boiler, where it becomes petroleum and flows down.
The petroleumthen follows the reverse of the path of the crude oil, being pushed along by the escher waterfalls.
On the right, it hits naphtha, which stops the crude oil from falling the wrong way and allows the petroleum to bead down.
Then it moves left until it hits the petroleum tank, which is an infinite storage tank.
The system appears to be stable unless gas enters the room or any of the liquids are disturbed. If gas enters the room, it breaks the escher waterfalls and the system breaks.
This version has repeating modules 4 tiles wide, which is an improvement over the 8 tiles wide and it is one tile shorter, allowing for more separated thermal contact points, which should improve the efficiency of the boiler in the same space. Its new width means they can also be stacked more efficiently, furthering the compactness. This system's modular design allows it to be build in any number of configurations with as many contact points as you want.
I currently don't have the crude oil production to thoroughly test this, so I entered debug to conduct a stress test. I started dumping 100 kg/s crude oil, which is the maximum amount my boilers can handle (when I build the second one). Note a different boiler design could boil more, but they would likely use more of the heat source. I found that my dupe access at the top right needed to be taller and that the crude oil spout needed to be airflow tiles or doors to prevent pressure damage. Due to unlimited resources, I added as much thermal conductivity as I can, using conveyor bridges and 4x automation wires.
The crude oil enters at approximately 77 C, enters the boiler at about 388 C. The petroleum leaves the boiler at around 399.5 C and leaves the exchanger at 110.7 C. The thermosensors in the boilers is set to 420 C for this flow rate, but it can be lower with lower flow rates.
As I ramp up my oil production, I will redesign my survival version to look more like this latest one.
