Weird mass deletion with mixed liquids in steam chamber

[Malcultimas]

!!LONG POST WARNING!!

I have a salt water geyser, volcano, gold volcano, and water geyser all in close proximity to each other.

I built a volcano tamer with the purpose to boil the salt water and use the water for electrolyzers and the salt combined with the gold volcano to make bleach stone in a bleach stone hopper to feed geo tuners I use to tune the water geyser as "free power" using steam turbines.

The volcano requires 2 steam turbines to manage its average output so I pump in 1000 g/s of magma which I use to heat up the steam chamber, which averages around 186 degrees Celsius.

The average output of the salt water geyser is 3383 g/s which isn't enough to feed both steam turbines with steam continuously so I pump in additional water from the nearby water geyser to hit a total of 4000 g/s of liquid being pumped into the chamber.

Salt water is supposed to lose 7% of its mass when boiled so I pump 3383 g/s of salt water using a liquid valve and an additional 720 g/s of regular water using a liquid valve. Accounting for salt water conversion rate, this should mean I'm pumping in just a bit above 4000 g/s of water, enough to feed both turbines. The room also has a "buffer" of 10 kg of steam per tile already in it.

The problem is I noticed the steam turbines were bringing the steam chamber down to vacuum levels in about 10-15 cycles with both turbines being half starved for steam. Both liquid inputs and both turbines ran without interruption each time.

At first I thought maybe the info I had on salt water conversion was out of date so I pumped in just 10000 kg of salt water and collected it in liquid containers. When the room was vacuumed out the containers had 9307 kg of water, obviously the math was right. I did the same with just 10000 kg of regular water with same results.

Than maybe I thought since I had the 2 liquid drains aligned vertically next to each other that there was some weird tile deletion happening so I moved each drain to opposite ends of the room. Chamber would still be brought down to near vacuum.

Getting desperate I decided to see if my save was bugged so I rebuilt the engine thing in a new game using debug mode on same map. Still same results, I even used the debug menu to highlight each tile in the chamber so I could see exactly how much steam was in there. I could watch in real time as the amount of steam decreased by about 2-5 kg every few seconds.

It wasnt until I started pumping in 1350 g/s of water that I saw the pressure in room go up, but if I turned on debug 10x speed the pressure would start dropping again until I started putting in 2400 g/s of water at which point it would start going up in 10x speed and REALLY go up on 3x speed and slower.

At this point I'm pretty much at a loss as why it's happening, it was supposed to just be a self contained leave it and forget it igneous rock, water, oxygen, and gold factory that was self powered, self cooled and ran forever without downtime.

My solution was to just use an atmo sensor to detect when steam levels drop too low and just pump in water as needed. It works, but it irks me that it didn't work the way I originally wanted it too when it should of.

 

 

 

 

 

[blakemw]

You could try valving the salt water down to 1 kg/s, heating it to above the boiling point in radiant pipes, then Venting it such that it instantly vaporizes.

[wachunga]

Could be this bug. Or some variant. Don't know if the bug ever got fixed. Or maybe it got partially fixed.

The TLDR is that under certain conditions (a combination of liquid flow rate, liquid temp, and steam temp) liquid entering a vent boils before it is emitted by the vent. This results in every other packet of liquid being deleted. Speculation is that there is some race condition shenanigans going on.

In the linked thread, 2 kg/s of 95C water flow needed ~375C steam temp for this bug to occur. Your lower rate of 720 g/s would bug out with a lower steam temp, perhaps as low as your 186C steam. Look at the "splashing" animation at the liquid vent. When the bug is triggered, the animation happens every other liquid packet instead of every packet.

To work around this bug, you want to disrupt the specific temp and flow rate conditions. Try super heating the water like @blakemw suggested. You could also try doing bigger water packets less frequently. Valve your water to 1.4 kg/s but loop half the packets back to your supply or something like that. Or keep the 720 g/s but send the packets to multiple vents so each vent only sees a packet every 2 or 3 or 4 seconds.

 

[Malcultimas]

So I did some extensive testing and I definitely noticed the splashing on every other packet of the water output, the salt water spashed every packet, so I believe it was indeed this bug and it was being triggered by having my water output too low. 

My fix of just feeding additional water as needed using an atmo sensor has been working fine and I'm swimming in water on this map so the loss of some doesn't bother me.

This was good info tho, I wasn't aware of this bug and I will definitely keep an eye out for it if there's future weirdness happening.

[Malcultimas]

Did another quick test, I made a fake steam room and filled it with 186 degree hydrogen and used 8 gas pumps to simulate steam turbines removing gas from the room. Using the gas material overlay I could see that a pocket of steam would appear 1 tile beneath the water vent move like one tile up and than vanish, it wasn't exactly every other packet, but it was pretty consistent. The vent outputing salt water produced a tile of steam every time.

Using this info I was able to get a packet of steam from the vent at exactly 1271 g/s of water.