Glitch pump question

[myxal]

Does the glitch pump mechanic (teleporting liquid cells through walls using airlocks) work with horizontally-placed doors? I'm actually trying to avoid it in a cryofuel machine design - I had the doors vertically and liquefied gases were teleporting outside.

Spoiler

 

[Gurgel]

I don't think I understand what you are talking about ...

[ghkbrew]

I think you're talking about droplet teleportation? If so, then the orientation of the airlock door doesn't matter.  If a droplet finds itself inside a solid tile (such as a door that just closed) it will search upwards until it finds a non-solid tile for the liquid to appear in.

For this design specifically, I think it's possible for you to get some liquid teleportation if a droplet forms right as a door is closing.  I don't know how likely that is to happen in practice.  It'll depend on other things like temperatures and flow rates.

 

Edit: Actually you might be safe, I don't think you get teleportation if there's a non-solid adjacent cell for the liquid to appear in, rather than teleporting upwards.

[Gurgel]

On 4/16/2025 at 2:18 AM, ghkbrew said:

Edit: Actually you might be safe, I don't think you get teleportation if there's a non-solid adjacent cell for the liquid to appear in, rather than teleporting upwards.

I seem to remember there was a patch that changed this some time ago. Not sure though. 

[myxal]

11 hours ago, Gurgel said:

I seem to remember there was a patch that changed this some time ago. Not sure though. 

Changed in what way? 

To clarify, I have seen the teleportation occur (on current version) with a setup like this:

I noticed it 3 times before I changed the design to what's in the OP. The horizontal door design didn't leak, but it wasn't running nearly as long. I'm hoping to do a proper test with both builds running simultaneously soon.

[myxal]

I added a door-close event counter. Save attached.

Well here's what I observed:

• In the vertical arrangement, oxygen gets teleported on the very first close action, but then doesn't for at least another 40 door closures.
• Chlorine gets teleported periodically in the vertical door setup
• In the horizontal door arrangement, neither get gas gets teleported, ever (44 events on O2 chamber, 16 events in the CL2 chamber)
• The CL2 chamber with vertical door is racking up the door-close counter much faster than the horizontal one (17 vs 28). The counters on the O2 chamber are even higher, but the difference between the setups is much lower (44 vs 46)

Spoiler

 

Taking a closer look, re: the O2 not teleporting after the first incident in the vertical setup - I see that in both door cells, it's condensing into atmosphere, rather than droplets. This goes back to my other question - when/how do gasses condense into droplets vs atmosphere. As I'm looking at the footage, it seems the droplets only appear below the super-cooled liquid pipe, where there's free space.

But that seems to be the issue - with the door oriented vertically, there's always a 2-high gap where droplets may appear.

I'm still having the setups churn through 13-20t of each gas, but so far the horizontal setup is looking like a winner.

EDIT: Yep, no leaks on the horizontal version. Vertical version leaked:

• 1223.3 g of O2 (out of 12t)
• 743.8g of Cl2 (out of 24t)

So, probably not a dealbreaker unless I want to maintain vacuum outside the build.

 

cryofuel v10 leak test 2.sav