Oxygen not included. Problem with pipes

[Veuxxx]

Hello, I’d like to start by saying that I’m new to this game and would really appreciate some help with my problem. I’m trying to build a cooling system based on heat exchange between a tank of cold salty water and pipes carrying water at 95 degrees. The thing is, I’ve been struggling with this setup for several days—constantly modifying it or adding parts. Some versions work fine in the sandbox, but when I transfer them to my asteroid, nothing functions, even though I believe I’m building it exactly the same way as in the sandbox.

My main issue is this: the tank on the left (where the water is almost boiling) has a pump that sends water directly to the central pipeline, flowing into a liquid shutoff valve. Each shutoff valve is connected by a separate automation wire and linked to a pipe temperature sensor set to trigger above 21 degrees Celsius. However, the radiant pipes start to clog, and my hot water bypasses these clogged pipes, going straight into the right tank. Why do my pipes get clogged? How can I improve or redesign my system?

I know about the aquatuner, but I specifically want to build this kind of system and understand how it works. Sorry for the phone photo—I can’t take screenshots or upload additional files right now. I hope you can help. Also, thanks in advance to everyone participating in the discussion!

[myxal]

The way ONI pipes work is that every connection between 2 pipe segments has a flow direction which doesn't change unless pipe segments and/or ports are added/removed. Presence, or lack of liquid blobs has no effect on flow direction. At a high level, the game sets up the direction so that liquids flow from outputs (green) to inputs (white) on any continuous network of pipes.

The golden rule of building functioning pipe setups is to avoid non-deterministic flow direction. In your setups, you have a continuous pipe which runs from one outpout, through 2 inputs, and to another output. This is non-deterministic, and for gameplay purposes, the game will pick randomly which way packets flow between the 2 inputs.

[HarryMcCleave]

Exactly what myxal said, from the users perspective it is random, for the technical side it is non-random but reliant on information (the order in which the conduit outputs are stored in memory) that the user doesn't generally have. Consider the simplification of the problem area of your pipes: Either set of flow directions (red arrows) would make sense and which will happen will be determined by which bridges output is considered first as a flow source when that segment of conduit has its flow directions generated (only happens on save/load or when a conduit network is modified).

[asurendra]

To precisely control flow, use liquid bridges. Put bridge on problematic pipe so game will clearly understand what you want from it.

For better understanding I advise to use mod "pipe flow overlay". It literally draw arrows on pipes so its obvious where you flow goes

Edited yesterday at 11:56 AM by asurendra

[Simonova]

I will also say for applications like yours (really cool design!), I've become fond of looping the liquid that fails the temperature check for the shutoff back around to the input (or, in the case of this screenshot, cooling the air around the system to -66 degrees with nectar), so that I have a single simple loop, but still achieve my goal of "always motion under the thermo-sensor", which solves a similar problem I'm guessing your setup is designed to solve (I know you don't want to use an aquatuner, but if you take it out, "hot loop" and "cold loop", split by the liquid shutoff, can be applied anywhere where you're trying to actively move heat).
One thing to note is the liquid bridge at the bottom of the screenshot feeding the nectar into the insulated line coming off the aquatuner: input slots control priority, so in this case, the liquid coming off the aquatuner will have priority over the liquid coming through the bridge, so if the aquatuner pipe is moving liquid full speed, the "return loop" on the bottom will fully stop (and stay stopped until the aquatuner pipe is clear, so still some breaks to the "always motion under the thermo-sensor" goal, but I called close enough good enough).

Final note for what your trying to achieve, running the piped liquid through a conduction panel, that has a conductive metal tile on the center tile of the conductive panel, and that tile surrounded by tempshift plates that are flooded by the liquid you're trying to exchange temperate with, will move head between materials much faster than just conductive pipes in liquid; something like this:

I forgot entirely, I use the exact same sort of "hot loop, cold loop" to regulate the temperature of the living area of my colony (I'm playing on the lab from the frosty planet pack, so lots of cold)

Here's the full loop: