Thread by @GeePawHill

Oxygen Not Included: I gave some bad advice on the how-to of the pour-in infinistore rig. I'm of the opinion one should correct such sins to the same audience one committed them upon.

This is the frame. The airlock door and the airflow tiles are required, to handle the pressure. (Both kinds of tile can handle any liquid pressure.) It must be either all doors or at least a door in that location: if it were airflow the gas we're gonna trap could leak.

As you can see, you can surround this bad boy with insulated tiles and stay on a 4-tall grid. In this case, I don't need insulation. You can widen that trough on the left side as much as you like. This one will leave just one column open to the air. How wide depends on sources.

Notice the gas pipes. A bridge in, a bridge out, and the three pipes to connect them all. The middle pipe between the bridges is over the three-deep part. Notice that, as long as that door is open or auto, the interior is accessible. Plumbing next.

Here's the plumbing. Nothing too special or clever. If you need non-pour input -- you will if you're positioned above one of your sources, the intake pipe is at the bottom. The output is at the top, heading out from the pump and going left to my base.

I'm using bridge priority mechanics & output-ordering. Incoming liquid will want to go right, and appears in sequence before the pump outlet, so will take priority. This keeps you from pumping liquid twice if you have a consumer. Otherwise, it goes left and pours in to the vent.

The wiring is nothing, just make sure you wire the pump. I always use conductive, so I can upgrade the grid later and not have to get inside. It'll take me a minute, cuz I'm still researching conductive. Then I'll show you the final how-to sequence.

Okay. 1) Lock that damned door. 2) Pour one container of your target liquid into the bottle-emptier. It will look like this. It's not obvious cuz of airflow tiles, but there's liquid all along the bottom of the chamber, not just under the door. That's important.

2) Check your gas view. Here's mine. See the two tiles of co2 inside? They're trapped, and that's the key. They are surrounded by liquid, solid, and door tiles. No access to the airflow, no where to go.

3) If the two tiles have different gases, go on to 4. (That does happen, just not usually.) If not, pump *any* other gas into that line with the two bridges and the one pipe between them. It doesn't matter what gas, as long as it's different.

Step 3A in progress here: the two tiles are still co2, but the pipe that crosses them is full of o2. I will deconstruct it, and then I'll have two tiles, one o2, one co2, and that's what step 3 does.

And after the destruction at 3b, here's the gas view again. Two gases now. One co2, one o2. Technically, the thing will now work. Practically, tho, we're still at risk. We have one required step and a couple of optional ones left.

Step 4: This is mandatory. See the two liquids at the top below the bottle-emptier? The one on the right must be capped. If not, and anything happens to sublimate it, the gases will escape. Capping it with any solid tile will work.

Step 5: You're good to go, but there's still some risk. If a non-target liquid falls into your pour-in area, it can cause problems. As far as I know, this guy's not safe for a multi-liquid infinistore. So I make sure I have a roof if it's anywhere risky.

Step 6: You can add more liquid vents over the pour-in column(s), really as many as you want. The speed with which it pressurizes and moves incoming liquid. Before: that's 65 tiles, 650kg of water. Before, and after. Elapsed time, about 3/4 of a cycle.

Alrighty then. I've made up for my bad how-to from before. This one is the correct one. My Friday Geek's Night has started. Chatcha in a couple hours. Feel free to ask questions or make mean-spirited comments. :)