Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

Beowulfe

Thermal Update - Hydrogen Bubbler (Liquid Oxygen)

Recommended Posts

Beowulfe    86

As I'm sure many of you have noticed by now, with the new update, state changing gasses inside of gas piping causes the pipe to break. That means that liquid oxygen is much harder to make via the traditional cooling methods.

Just wanted to share a new method that seems to work well. Since we need to get oxygen to -183 degrees *outside* of a pipe, we need a clean way to transfer heat to it on a low power, low maintenance system.

The easiest way to do that is to cool something else, and allow it to pull heat out of the oxygen. Hydrogen has a condensing point of -250 degrees, which means we can safely cool it down much further than oxygen without breaking pipes.

 

So, what I've done is cool a room full of hydrogen down to about -210 degrees (any colder and it will actually freeze the oxygen), and then pump polluted oxygen into the room with it, "bubbling" it through the hydrogen (hence the "Hydrogen Bubbler"). It will cool the polluted oxygen naturally into a liquid form, which then turns into clean, purified oxygen, which then gets pumped to a secondary location. Fun! :D

 

The best part of this is that, using the new thermal switches, you can set this system to be completely hands-off! You can keep the pressure in the room from getting too oxygen-heavy, you can turn regulators on and off to cool the hydrogen if it starts warming up, and you can pump oxygen in only when the hydrogen is primed and at the right temperature. It takes a while (and some power) to get the hydrogen up to the proper temperature, but once you do, the ambient temperature of the hydrogen stays high enough that the power to maintain the condensing point in the room is *very* minimal (as in no more than running a single generator, pump and filter for about 5 seconds every second cycle).

Again, all this is automated. Once everything is connected, there is no manual management required. Very, very rarely, a pipe will get damaged due to the volatile nature of the gasses and temperatures involved, but maintenance is minimal.

The device itself, however, is very complicated.

Here's some screenshots, in-depth details, and a proof of concept, with the best hydrogen bubbler design I've been able to create so far:

Hydrogen Bubbler.jpg

This is the overall design - The hydrogen is sitting at a comfortable -203 degrees right now, which means that the oxygen coming into contact with it condenses into liquid form. The liquid pump will pump out the oxygen as it condenses, and the "staircase" ensures the liquid flows to the bottom of the bubbler.

Let's take a step back and look at the process.

After all of this is built, *before* turning anything on, you need to fill the chamber with pure hydrogen and nothing else. The bubbler should have a room pressure of about 1000 - 1150 grams of hydrogen per tile. This is how you start "priming" the system.

The regulators at the top are intended to cool the hydrogen, but *only* cycle on if the hydrogen warms up too much, and only some turn on at a time. When initially "priming" the system (when fresh, uncooled hydrogen fills the bubbler, with no oxygen in the system), all of them will activate in order to rapidly cool the hydrogen. As it gets closer to the appropriate temperature, the thermal switches will shut off the regulators one at a time. The valve/bridge system at the top of the regulators ensures that the gas naturally bypasses any regulators that are currently powered off.

While priming, the bubbler will use a lot of power - almost a full circuit's worth. Be aware of this. Once primed, the energy cost is *ALMOST NOTHING*, as hydrogen will barely lose temperature over time. If you are concerned about the immense power consumption of the system, or if you need to temporarily cut back on power on the system, turn off some of the valves at the top (starting from the right), and it will stop gas from flowing into the regulators, removing their power draw from the circuit.

There are several complex systems at work here, between the thermal switches and piping systems, so I'll try to cover each in detail. To start with, the overall construction, and a reference for specific part settings:

 

Hydrogen Bubbler Overlay.jpg

  • V1: This section controls the flow of gas through the regulators. It must be built EXACTLY as pictured here, bridges and all (see the gas piping picture below for connection details). The intent is to ensure gas never flows into more regulators than intended (to prevent liquefying the hydrogen), and also to provide a really neat bypass system. If any of the regulators are powered off (which will happen automatically with the thermal switches, aside from initial priming, most of them will be off at all times), the gas will automatically bypass them, in a way that ensures there is NEVER any gas backflow. If there is any, the hydrogen will liquefy and break pipes, and we do not want that (as it removes the automated nature of the system). Valves in this section should always be set to maximum flow, unless you want to conserve power during priming, in which case you can set some of them to 0 g/s, starting from the right-most regulator. Beyond this, there is no need to touch the valves at all.
  • V2: These valves both control the amount of incoming gas and provides the gas with a one-way backflow valve. It ensures that vents becoming over-pressurized will not cause the polluted oxygen to flow backwards towards the other output, reducing efficiency. They should be set at max flow at all times, unless you want to reduce the amount of oxygen you are processing (for some reason).
  • V3 and V4: These two valves are intended to provide an "even flow" output of hydrogen to the vent. Without them, the hydrogen can come in "spurts", which causes temperature spikes and slower overall temperature change (reducing efficiency). V3 should be set to 500g/s and V4 should be set to maximum flow. This will split 1000g blocks of hydrogen into two 500g blocks (with the second block being potentially slightly larger or smaller, depending on pipe pressure).
  • T1 through T5: These five control the regulators (one each) and the gas pump in the upper left. These are intended to gauge approximately what temperature the hydrogen is as it flows into the pump, so it can determine how many regulators to pass it through. If the temperature of the hydrogen is very high (AKA, the system is being primed), the maximum number that can be activated without liquefying the hydrogen (which occurs at -252 degrees) will be turned on. As it gets closer to -250 degrees in the bubbler, it changes it's target temperature to -203 degrees, and only enables enough regulators to match that. -203 degrees is our ideal temperature, and once reached, the pump and regulators will shut off entirely. Oxygen will freeze at -219 degrees (very bad), and liquefy at -183 (very good), so we need a temperature in the middle that ensures the oxygen will liquefy quickly, but not freeze or turn back into a gas once it's entered the liquid pump (which breaks the liquid pipes). For the settings for these specific ones, see the following:
  • T1: Warmer than -203.2 degrees
  • T2: Warmer than -198.2 degrees
  • T3: Warmer than -193.2 degrees
  • T4: Warmer than -173.2 degrees
  • T5: Warmer than -183.2 degrees
  • TP: This is, essentially, a "pressure sensor". This should be set to colder than -188.2 degrees. If the oxygen layer begins rising too high (pushing into the hydrogen pumping area), it will cross over this switch. Since oxygen liquefies at -183.2 degrees, we know that if it's colder than -188.2 degrees, it cannot be oxygen here. If it is warmer than -188.2, then oxygen has passed over the switch, which will shut off the external pumps bringing oxygen into the bubbler until more of it can be liquefied (since we don't want the oxygen overpressuring the room too much, preventing the hydrogen from doing it's job). This essentially acts as a balance between the atmospheric pressure of the oxygen versus the hydrogen.
  • TF: This is the "oxygen freeze" sensor. It should be set to warmer than -208.2. This sensor detects if the temperature near the oxygen output is dropping too much, and shuts off all regulators if it is. If the temperature drops lower than -208.2, we're at risk of freezing the oxygen coming in, which is very bad for our bubbler. This prevents oxygen freeze.
  • T6: This should be set to warmer than -183.2. This switch controls the second gas pump, and is intended to turn it off once we reach a temperature that means we're going to start pumping oxygen in shortly. Since we do not need as much hydrogen passing through the system at that point, and this pump is in a position that raises the chance of it needlessly pumping oxygen through the pipes instead of hydrogen, we regulate the system down to one pump once we reach "primed" status. This increases overall efficiency and removes the risk of liquefying oxygen inside the pipes, breaking them.
  • F1: This filter should be set to filter Hydrogen! Simple enough. :) The pipe leading out of the white outout should be built out of Abyssalite, in order to prevent gas from losing temperature and changing state in the pipe!

 

Whoo boy, okay, that's the "quick" rundown of the major systems. Now for the actual gas piping and power sections:

Hydrogen Bubbler Gas.jpg

As you can see, the gas piping, especially near the regulators, is not simple. It may look needlessly complex, especially with some of the bridges in the system (both at the regulators and elsewhere), but the bridges provide a "one-way backflow valve", that essentially draw gas in and ensures it doesn't backflow into the system. Without these, the system will break in various ways.

IMPORTANT NOTES:

-Make sure you build the three central regulators (or all, if you have the materials) out of gold amalgam, or you will have overheating issues during priming!

-The liquid piping leading out of the liquid pump should be built out of Abyssalite, as well as the pump leading out of the white output on the filter at F1, in order to prevent liquids/gasses from losing/gaining temperature and changing state in the pipes!

Note the two external pumps bringing in polluted oxygen. As long as they are connected to the system as shown (a single pipe branching to both valves in V2), you can connect as many pumps as you would like, the system will automatically regulate and process as much polluted oxygen as it can handle, which is quite a lot (more than 2 pumps worth).

See above for descriptions of the labelled sections.

 

Hydrogen Bubbler Power.jpg

Fun stuff! This is where we get into the thermal switches, and the huge net of specific wiring needed to work with them. See the above section for details on thermal switch settings and what purpose each section provides.

Note that any external pumps into the system should be connected through thermal switch TP, as shown with my two pumps here, ideally branching off the same wire as the liquid pump. Failure to do this can result in over-pressurizing the oxygen in the bubbler, causing pipes to break and oxygen liquefication to slow down significantly.

Power requirements during priming are approximately a full circuit (1800w), which is very heavy on power. However, you can reduce this by disabling specific valves on the regulators (see comments on V1 above). This is not required to be "continuous" power, if it's interrupted during priming, it will not cause any issues. The ambient temperature will remain the same, and will pick up where it left off next time you provide it power, so this can be a longer term goal.

Once primed (hydrogen temperature is at about -200 degrees), power costs are almost nothing. Over three cycles, a single regulator, filter and gas pump ran for a total of about 5 seconds, making the system incredibly efficient. Temperature loss, and by extension power cost, is almost nil.

 

THE ACTUAL STATS:

With the exact setup shown above, with two gas pumps pushing polluted oxygen into the system with a steady, strong flow, it produces approximately 350kg of clean, liquid oxygen per cycle. This is approximately enough for 5.5 duplicants, assuming you can continue providing enough polluted oxygen (Morbs/slime anyone?). This amount can be increased by adding additional external gas pumps, or by increasing the vertical height of the chamber. Note that increasing the vertical height will require you to shift some things around, and increase the overall priming time - switch TF, along with the gas vent there, and valves V3 and V4, should be shifted downwards to match the liquid pump, everything else should stay in it's current position. Be sure to keep the "stepping" form of the lower right wall, or you will begin to lose oxygen on other levels due to surface tension of the liquid.

I think that's enough of a writeup for now. I had a lot of fun setting this up! There's some pretty complex/neat mechanics in play in this system, so if you have any questions, feel free to ask. I can post a video of the system in action if requested, as well. :D

Share this post


Link to post
Share on other sites
Fatmice    64

Yep.  Been experimenting with this.  Works very well.  Hydrogen is an excellent working fluid.

Share this post


Link to post
Share on other sites
QuQuasar    4

(Edit) don't bother building the device I describe below. Gas loss bug will destroy all the oxygen and hydrogen you pump in. I am currently attempting to re-work it to prevent this.

 

 

 

Based on your description, I've been trying to design a compact oxygen liquidiser. I haven't tested it yet, but I intend to give it a try the moment I get back to the game. 
 
Here it is, rendered in delicious ASCII (hope this works...):
 
 
Buildings

            ███████═╗ 
        PPP █ TTT═█ ║
        PPP █═TTT █ F
      █████████████═F<H2
    F═█ PPP V PPP █═F
════F █ PPP   PPP █
 O2^F═█ S       S █
      █████████████ 

P = Pump
T = Thermoregulator
F = Gas filter
V = Vent
S = Thermo switch

 
Piping 

            ╔═══════╗
          P═╗ T═══╗ ║
            ╠═══T ║ F
      ██████║█████╚═F<H2
    F═════P V   P═══F
════F ╔═════╝     █ 
 O2^F═╝           █ 
      █████████████ 



  
This 9x8 unit is my preferred design: it might be possible to remove the lowest level of the cooling room, but then you'd lose a thermoswitch and your gas pumps would end up in liquid.
 
As you describe, it has two circuits: the hydrogen cooler (everything right of center) which activates when it is warmer than -220 degrees inside the room, and the oxygen injector/extractor (everything left of center), which activates when it's colder than that.
 
The pump in the top right provides Contaminated Oxygen from the surrounding area. A moderate amount of hydrogen would need to be pumped in at the start to prime it.
 

Share this post


Link to post
Share on other sites
QuQuasar    4

(Edit) Okay, I just noticed a design flaw. If enough non-oxygen is pumped in for whatever reason, the whole thing will hit max pressure and promptly seize up. 
         
Simplest to just ensure that never happens. I've changed the design to add a filter to the input line, to sanitize the inputs and ensure no chlorine or carbon dioxide gets in.

Buildings

      ╔═══F ███████═╗ 
      PPP F═█ TTT═█ ║
      PPP═F █═TTT █ F
    V═█████████████═F<H
    F═█ PPP V PPP █═F
═O2═F █ PPP   PPP █
    F═█ S       S █
      █████████████ 

Pipes

      ╔═══F ╔═══════╗
      ║ P F═╣ T═══╗ ║
      ║ ╚═F ╠═══T ║ F
    V ╝█████║█████╚═F<H
    F═════P V   P═══F
═O2═F ╔═════╝     █ 
    F═╝           █ 
      █████████████ 

 

Share this post


Link to post
Share on other sites

Since tiles change temperature now, could you stack two rooms on top of each other and use a single tile high floor as a cooling plate?

So the bottom room you would have the pump/thermo cycle with your working fluid of hydrogen and an temp switch to prevent freezing it solid, then above the cooling plate floor, you pump in as much contaminated oxygen as the room can hold, then pump out the liquid O2 with a fluid pump?

That way you don't have to worry about pressure locking the system.

EDIT:

It just occurred to me that this system would likely work better if the rooms were side by side vertically instead of stacked. That way the O2 side can be exactly two wide with the fluid pump at the bottom so no liquid O2 get's missed by the pump, and the temp exchange plate can be as tall as you want it to be.

Share this post


Link to post
Share on other sites
Fatmice    64

Not really.  Temperature exchange is poor so the system isn't very scalable.  I'm going to build one and see.

Share this post


Link to post
Share on other sites
Fatmice    64

Normal tiles take forever to heat up and cool down.  Very difficult to control temperature with such slow response.

Share this post


Link to post
Share on other sites
SDAFsdaf    3

you can use more than 3 themor for faster convert . maximum is 6 ( because your hidrogen will become liquid too )

i ll add themor switch and more pump to make better automatic convert

Share this post


Link to post
Share on other sites
Beowulfe    86

Just added screenshots and a write up of my system in the main post, let me know if you have any questions! It's rather complex, but automates everything, including preventing freezes, unintentional liquefication, handling atmospheric pressure, and priming the system (initial cooling). Aside from the incredibly rare case, there's no maintenance or manual control required at all. :)

Thoughts, questions, comments, concerns?!

Share this post


Link to post
Share on other sites
xbeo    8
2 hours ago, Beowulfe said:

Thoughts, questions, comments, concerns?!

Geez, great design! Thank you so much for sharing!

Have to test this right now...

Share this post


Link to post
Share on other sites
Mijae    9

Really nice design.  I haven't had a chance to play around with thermal switches yet.

Has anyone actually identified a significant difference in liquid flow down steps?  If the system is continual, I think you should get the same amount of liquid pumped if it was flat at the bottom (perhaps just the pump itself down one tile).  Worst case I would imagine is slightly more residual liquid if you were to stop the system (which seems like you wouldn't want to).  If so, it would just equate to slightly longer priming time but might allow for an easier build and/or gas balance.

Share this post


Link to post
Share on other sites
Fatmice    64

Step down saves half the space and "concentrate" the liquid so that you leave less residues on the tiles.

Share this post


Link to post
Share on other sites
Saturnus    332
On 20/03/2017 at 5:43 AM, Fatmice said:

Step down saves half the space and "concentrate" the liquid so that you leave less residues on the tiles.

A flat bottom and mesh tiles on top will also distribute liquids almost as well and is less complicated to build. Just leave the mesh tiles out at the 2 tiles where the pump is.

However, the stairs also concentrate gas flow so that the cold hydrogen comes in contact with the most amount of oxygen.

If I'd have one change to the system it would be to reverse the stairs from right to left side as hydrogen (as the lightest gas in the system) has a tendency to go to the top left corner where the circulation pump, is therefore having the stairs reversed would mean a slightly higher efficiency but it will be a miniscule difference. 

Share this post


Link to post
Share on other sites
phelaen    1

has anyone tried using normal pipes instead of insulated ones made out of abysilite (not sure how to spell that) ? maybe those don't break?

 

 

Share this post


Link to post
Share on other sites
xbeo    8
4 minutes ago, phelaen said:

has anyone tried using normal pipes instead of insulated ones made out of abysilite (not sure how to spell that) ? maybe those don't break?

Yes, i tried. They break.

Share this post


Link to post
Share on other sites
Dogbarian    8
15 hours ago, Beowulfe said:

Just added screenshots and a write up of my system in the main post, let me know if you have any questions! It's rather complex, but automates everything, including preventing freezes, unintentional liquefication, handling atmospheric pressure, and priming the system (initial cooling). Aside from the incredibly rare case, there's no maintenance or manual control required at all. :)

Thoughts, questions, comments, concerns?!

IT'S BEAUTIFUL! (said like the French guy in Raiders, right before the Ark melts his face).   Amazing work, dude!

Share this post


Link to post
Share on other sites
xbeo    8
On 3/16/2017 at 6:14 PM, Beowulfe said:

T6: This should be set to colder than -183.2. This switch controls the second gas pump, and is intended to turn it off once we reach a temperature that means we're going to start pumping oxygen in shortly. Since we do not need as much hydrogen passing through the system at that point, and this pump is in a position that raises the chance of it needlessly pumping oxygen through the pipes instead of hydrogen, we regulate the system down to one pump once we reach "primed" status. This increases overall efficiency and removes the risk of liquefying oxygen inside the pipes, breaking them.

Do you mean Activate if warmer than -183.2? If not, i do not understand the logic behind this. The second pump should turn off when it is cold enough, right?

Other than that, great manual!

Share this post


Link to post
Share on other sites
xbeo    8
8 minutes ago, Saturnus said:

No.

 

On 3/16/2017 at 6:14 PM, Beowulfe said:

we regulate the system down to one pump once we reach "primed" status.

With it being set to colder, it would start with one pump and then regulate up to two pumps once primed status is reached?

Share this post


Link to post
Share on other sites
Beowulfe    86
30 minutes ago, xbeo said:

Do you mean Activate if warmer than -183.2? If not, i do not understand the logic behind this. The second pump should turn off when it is cold enough, right?

Other than that, great manual!

You are correct!! It should be active only if warmer than -183.2. Good catch, typo on my part. Fixing now, thanks! :D

Share this post


Link to post
Share on other sites
xbeo    8

I tried it out, works like a charm.

But I had to pour water over the whole damn thing when starting it to prevent the air coolers from overheating. Did you use isolated pipes behind the coolers?

Then I had to find the right type of material for the pipes because granite broke sometimes :D

And then I tried to boil the liquid oxygen and pump it back into my base... which cooled my whole base it down to -50C on average, much colder on certain spots.

Then I tried some heating methods and my game started to crash, which almost never happens.

10/10, would build again!

 

Share this post


Link to post
Share on other sites
darkflames9    13

Finally, an oxygen liquifier that works!  It's a shame we have to resort to such convoluted and precise contraptions just to clean contaminated air.  Compared to the ease of an electrolyzer, it's just not worth it.  However, the ocd in me dictates that I must use up every single resource!  Anyway, your machine works, but I had to replace the liquid pump at the bottom with a gas pump; for some reason, every time liquid o2 is sucked up, it immediately boils, even if the pipe is insulated.  Therefore, I expanded the bottom bucket by 1 row of tile to collect the liquid o2 and placed a gas pump where the liquid pump is.  Now, liquid o2 collects at the expanded bottom bucket, quickly boils, and the gas is then sucked up by the gas pump.  Now the next issue to solve is how to warm up the o2...

Share this post


Link to post
Share on other sites