Jump to content

Oxygen Liquificator (tm) Agricultural Addition


Recommended Posts

Seeing as the fluid-in-pipe/duct heating bug had been solved, I wanted to try and build an oxygen liquification setup.

Couple issues that have to be addressed: If your hydrogen cools to below -252C, it liquefies, damaging your ducts. If your liquid oxygen heats above 183 while in a pipe, it boils, damaging your pipes. 

After spending a bunch of time playing around with constellations of valves, and thermo-switches, etc to try and maintain control, I came to a very simple realization: Heat extracted by the thermoregulators is equal to the heat extracted from the polluted oxygen. The heat transfer in the liquifier is determined by the temperature gradient (ie difference between the polluted oxygen and the cooled hydrogen), the conductivity of the conduit (all granite), and the area of the radiator (16x9, in this case). Ergo, for a given number of regulators, and a given polluted oxygen input temperature, if you make the chamber large enough, you can ensure the thermoregulators run constantly, and never hit hydrogen's dew point. 

In this example, the maximum cold temp reached is negative -233 celsius. By my calculations, the room could be *slightly* smaller (about 8 tiles high) and still just narrowly avoid liquifiying the hydrogen. 

Other design features:

The liquid oxygen is dripped out where the thermo-regulators live, so their heat can partially reheat the liquid oxygen. Because i want to avoid the liquid oxygen boiling in the pipes, that pump is set to a hydroswitch set to 100kg (it could probably go a little lower). This ensures the pump only pulls relatively large (>5kg) packets of liquid that have enough thermal mass not to boil in the pipes). Each time the pump kicks-on, it pumps out ~60kg of liquid O2. Because these bursts are relatively infrequent (~250kg O2 / cycle total), the thermoregulators will overheat in between cooling cycles, which is why i installed the polluted water cooling system off to the left. 

I route the polluted oxygen through the cold biome near where the liquid O2 is dispensed. This pre-cools the polluted oxygen, increasing system throughput, while reheating the liquid O2. This should probably be larger; it's nowhere near equilibrating in temperature. 

Curious for your feedback! Also, if someone wants to give me how many seconds per cycle, we could compute how this compares energy-intensity-wise to other methods of oxygen generation. 

Ultimately, scaling presents some interesting problems: if you run your thermo-regulators in series, like i do, your average cold-gas temperature has to be higher (Each thermoregulator drops the temp by 15C, so it means you need a 30 C rise across your radiator. This means your average cooling temp is going to be at least -235C. If you your thermo-regulators in parallel, on seperate loops, you run the risk of un-even cooling, where one of your two loops over-cools. I think such a design is possible but would require careful balance. You could also just make multiple, seperate rooms, each with their own thermo regulator, but then you're wasting space with walls. 

Overview: Clockwise from upper right: Polluted air intake, Liquid O2 pump, polluted oxygen pre-cooling loops, twin thermo regulators

Overview.png

Single loop, dual thermo regulator setup: 

gas.png

Liquid setup; a batch of liquid O2 is on it's way out. Polluted water cooling stack on the left. 

liquid.png

Link to comment
Share on other sites

A way to avoid having to use liquid cooling (and dealing with polluted ice or released polluted oxygen) is to use more regulators and have a standard pipe junction before them that will send packets in alternating manner to them. Then each regulator will only get a packet to process every once in a while (e.g. every third packet) and will have time to cool down in between them. It also won't increase power consumption as the regulator only draws power when a packet is coming through it. With such alternating, you can keep cooling them with wheezeworts.

Link to comment
Share on other sites

I think I'm avoiding any polluted liquid-> polluted oxygen, as it is all sitting under a CO2 blanket. As it's going through a valve at low flow rate, the power to the pump is pretty minimal. 

Going to multiple thermo-regulators, I'd worry about making sure the pipe lengths were coordinated such that they didn't all kick on at the same time, overloading the circuit...

LIQUID COOLING.png

Link to comment
Share on other sites

19 hours ago, Kasuha said:

A way to avoid having to use liquid cooling (and dealing with polluted ice or released polluted oxygen) is to use more regulators and have a standard pipe junction before them that will send packets in alternating manner to them. Then each regulator will only get a packet to process every once in a while (e.g. every third packet) and will have time to cool down in between them. It also won't increase power consumption as the regulator only draws power when a packet is coming through it. With such alternating, you can keep cooling them with wheezeworts.

Well- your concern was prescient. I let the system run for a while longer, and the temperatures dropped enough to freeze my cooling water... which would cause the thermo-regulators to overheat (Now, the idea of an AC unit "overheating" when surrounded by 5kg of -30C oxygen is another story...) 

Looks like parallel processing is perhaps necessary... 

Link to comment
Share on other sites

Well, here's my particular unit running.  It took a number of cycles to setup, tweak, and prime but it's more or less working fine.  There's a few things I'd probably change, if I was to rebuild it.

20170611220500_1.jpg

20170612103349_1.jpg

20170612104433_1.jpg

 

So, my basic idea when building it was whatever heat was going into the unit, should mostly stay in the unit to reheat the oxygen coming back out.  And it seems like I've generally got it to work.  There's the two separate radiators.  The primary one with hydrogen, and a heat exchanger for the polluted oxygen to exchange heat with the clean oxygen coming out of the unit.  I had to add that second part on the right because it wasn't big enough.  Now, it seems to be about right.  The clean oxygen is coming out at a decent temperature, although it might go up higher over time.  

 

Using what I learned from the hydrogen bubbler, I used valves for a bypass of the regulators in the hydrogen loop for when they don't run.  (Which pretty much doesn't happen I've noticed.)  And the regulators process half of the hydrogen packets.  I had originally tried just one, but it could just barely not cool itself.  I did have to use liquid cooling to originally prime the unit with the one regulator.  I used the polluted water from that bio distiller which worked quite well, actually.   The right regulator does tend to run a little warm, but it does not seem to get any where dangerously hot.  As long as fresh cold oxygen is pouring into the room.

 

I had to add that dip in the center of the exchanger room because of CO2 buildup.  It was starting to muck up with the cooling of the regulators, so its basically a CO2 trap.  Occasionally, I have a Dupe or two hop into the room to mop up puddles of liquid CO2 and that seems to be cleaning it up okay.  Interestingly, as long as the freezing room is filled with PO2, I don't seem to have a problem with the hydrogen getting too cold.  It always seems to level off around -230 ish.  Even with the regulators running non-stop.

 

I think I need another section of mesh tile where the liquid oxygen is coming out as it sometimes completely evaporators from there and clean oxygen back-fills into the primary cooling room, but it generally works okay.  I would also consider adding another parallel regulator, just as a safety margin.

Link to comment
Share on other sites

On 6/12/2017 at 9:25 AM, The Flying Fox said:

I think I need another section of mesh tile where the liquid oxygen is coming out as it sometimes completely evaporators from there and clean oxygen back-fills into the primary cooling room, but it generally works okay.  I would also consider adding another parallel regulator, just as a safety margin.

When you were doing the initial cool down, how did you stop polluted oxygen from filling the entire chamber (my setup separates the heating and cooling, using a pump to transfer the L-O2)

Link to comment
Share on other sites

5 minutes ago, QuantumPion said:

Could you explain how you used valves to make a bypass for the regulators to continue flow when they weren't running?

If you put regulator input on the pipe, it will steal all its contents while the regulator is running.

KtNwjwU.png

While when not running, it will continue through the pipe.

LHV96kq.png

So all you need to do is to safely merge the output from the regulator and the bypassing flow back to one pipe, which is best done with a bridge or fully open valve.

Z6O7mEM.png

v3VELiJ.png

Link to comment
Share on other sites

18 hours ago, caffeinated21 said:

When you were doing the initial cool down, how did you stop polluted oxygen from filling the entire chamber (my setup separates the heating and cooling, using a pump to transfer the L-O2)

Well, where I built the machine was pretty much all P-O2, if I remember correctly.  So, the whole machine was already pre-filled with the stuff :D  I distinctly recall when I was priming the hydrogen cooling loop that I hadn't hooked up something to the PO-2 fill-line just yet, so when the primary cooling room got cold enough, L-O2 started raining and pouring out until there was nearly a vacuum in the room.. and some hydrogen liquidized and broke a pipe.  Ooops!  That did create a good cloud of oxygen in the pre-cooler that pushed the P-O2 around and, using a gas filter, I was able to eventually scrub most of the free floating P-O2 from the machine.  There's still some little pockets of the stuff in there though.

20170614015916_1.jpg

Here's the machine after about a 100 cycles.  It has seemingly stabilized although the regulators do run quite a bit hotter.  The left unit runs around 55-45C and the right unit runs around 95-85C, but they've stayed that way for a while now.  The output oxygen has leveled around 10C.

 

I realized that without a major change to the machine, my idea of another mesh tile wouldn't work very well.  The left regulator is just too close to the output of the L-O2 and its heat boils off the L-O2 too fast for a seal of L-O2 to work in that location.  This is what was causing oxygen to back-fill into the primary cooling room and then, when enough cold oxygen built up within, it would all suddenly rain out of the room in a big gush into the pre-cooler room.  This was causing the regulator's temperature to bounce around and wasn't exactly ideal.

 

So, I modified the cooling room slightly, as you can see.  I realized that the 30 grams of L-O2 on that bottom step was always there and so, could be used as a liquid-lock.  It worked perfectly!  Now, L-O2 tickles onto that bottom step and then into the trap where the thermal switch is, warms up, and then flows into the pre-cooler room to cool the regulators more.. regularly.  It no longer back-flows into the primary cooler.  This has allowed a more uniform stream of P-O2 to flow into the primary cooler and likewise with clean O2 out of the unit.  Every 10-15 seconds it pumps out around 2.5 to 3Kgs of O2.  Not bad!  Enough for 2 to 3 Dupes.

 

And it's cheap to run too.  Since the regulators split the hydrogen packets between them, I would assume that, on average, they only consume around 240watts, but could be a little more.  Then there's the gas pump that feeds the machine P-O2 and the pump that pulls O2 out of the machine, neither of which run all the time.  Reheating the cold O2 is solely being done by the regulators and the incoming P-O2 that sits in the pipes in the pre-cooler.

 

Another thing of note I find interesting.  Now that I've stabilized the output of the primary cooling room, the output temperature of the hydrogen from the regulators is around -216.  I've not seen it go below -220 at this point which seems quite ideal as the regulators are running constantly.  Did I just happen to stumble upon the ideal ratio of exposed granite pipes in the primary cooling room to keep the hydrogen from getting too cold?  I would assume that with too few, the hydrogen gas wouldn't be able to get enough heat from the P-O2 and eventually get too cold and break the pipes.  However, with too many exposed pipes the hydrogen gas would warm up too much towards the end of the loop, creating warm spots in the room where you may not want them.

 

The other thing I find interesting about the machine is that its power requirement is.. roughly in-line with with an electrolzyer setup.  Ball-park at least.  I think it might be feasible to design a P-O2 generating room to feed it.  Something with a lot of shallow pools of P-water.  Hmmmm...

Link to comment
Share on other sites

5 hours ago, Kasuha said:

If you put regulator input on the pipe, it will steal all its contents while the regulator is running.

KtNwjwU.png

While when not running, it will continue through the pipe.

LHV96kq.png

So all you need to do is to safely merge the output from the regulator and the bypassing flow back to one pipe, which is best done with a bridge or fully open valve.

Z6O7mEM.png

v3VELiJ.png

Oh duh that is pretty simple. Why do you need the bridge or valve at all though, and just route the bypass pipe to intersect the main pipe? Is that just to prevent back flow/ confused packets?

Link to comment
Share on other sites

40 minutes ago, QuantumPion said:

Is that just to prevent back flow/ confused packets?

Exactly. With valve/bridge you get one end of pipe with two inputs, and one end of pipe with two outputs. You can join them together to get a closed loop and it will circulate.

Link to comment
Share on other sites

49 minutes ago, eloy2030 said:

with this setup I gain ~70-80 C

 

http://steamcommunity.com/sharedfiles/filedetails/?id=944176487

 

... and that's after dropping CO2 in the chamber.

..Oh, huh!  You're.. using a granite wall separating the two chambers to transfer the heat from the P-O2 to the hydrogen in that little room.  That's kinda neat, you can more easily regulate the temperature of the hydrogen that way.

 

OHHHH!  Idea!  Why not make the two rooms join together, like fingers inter-lacing!  1-tile wide sections from the hydrogen room that poke into the P-O2 room.  This would hugely increase the surface area -and- you could put wire bridges that cross the gap to further increase the transfer of heat from the PO-2 to the hydrogen!  Vents could be installed at the ends of these 'finger' rooms to pump fresh cold hydrogen and P-O2 into the two entwined but non-connecting chambers.

 

The only issue I could see is having the hydrogen gas stalling within those 1-tile wide finger sections and getting too warm.  The gas pump have have to keep pumping all the time to keep the hydrogen mixing and at a constant temperature. 

Link to comment
Share on other sites

4 hours ago, The Flying Fox said:

The only issue I could see is having the hydrogen gas stalling within those 1-tile wide finger sections and getting too warm.  The gas pump have have to keep pumping all the time to keep the hydrogen mixing and at a constant temperature. 

yeah, I don't think that would be efficient, I'll try it though... What could be is mixing this with running the hydro pipe through the O2...

Link to comment
Share on other sites

Well, one of the other things I was playing with in debug mode the other day, was another design of a P-O2 liquifier.  This one testing using a granite wall and some wire bridges for a pre-cooler.

20170619160208_1.jpg

 

It's roughly pushing 500G/s of P-O2 to O2.  I have the liquid valve set to 507G/s and the liquid pump almost kicks in whenever the liquid O2 is out of the pipe.  The problem that's staring to form is, the gas pump that's feeding P-O2 into the primary cooling room just can't keep up!  :D  The pre-cooler is so good at exchanging heat from the cold O2 to the P-O2 that the input temp of the P-O2 is now past -117c and is slowly getting cooler.  The other problem is that of 'bandwidth' through the bottom room of the pre-cooler.  Gas just takes soo long to flow from one end of a room, to the other.  I've had to steady increase the size of the bottom room and move the gas pump away from the the wall so that it can get more air.  The right side of the room is at max pressure, but the pump is sitting in 300 to 200 gram air.

 

The bottom room of the pre-cooler has an interesting temperature delta.  The left top corner is -135c, the bottom left corner is -123c, the top right corner -54c, and the bottom right corner is -81c.  So, even with a room that tall, the heat is sucked straight out of it, although I'd hazarded a guess that the performance has partly to do with the top part of the pre-cooler having XXKgs of oxygen pre tile.  Heh.  For comparison, the top part of the pre-cooler, the P-O2 is coming from the right, flows towards where the L-O2 is dripping, then goes back up along the center back to the right.  At the very mouth of the machine the P-O2 is 31c and by the time it reaches the vents is around -20c, but I'm sure a good 10c of that change is because there's an absurd amount of cold O2 in the top part.

 

Fun, absurd, but not very practical in the game. :D

 

The one practical thing I learned is that the primary radiator's length of 75-100 granite pipes seems to be a good amount to use, in general.  The hydrogen packets are coming back to the regulator at around the 14 degree difference it can cool.  The temperature of the hydrogen in the pipe is being regulated by the P-O2 raining out of the machine and being removed with fresh hotter P-O2 moving into it's place.  Thus, we don't really need a thermo switch to gauge the temperature of the hydrogen in the pipes!  Instead, an atmo switch works just as fine in its place.  If the air pressure gets too low inside the primary cooling loop, shut off the thermo regulator until more P-O2 is added to the machine, simple!

Link to comment
Share on other sites

I was trying my own hand at a compact oxygen purifier, taking the heat recycling idea from my thermoregulator boiler project and applying it to a design that can release its output straight into the base and shouldn't need wheezeworts to cool itself once it reaches a steady state. This is done by dropping any liquid oxygen produced straight onto the thermoregulators to cool them.

Oxygen Purifier.jpg

Overlays in the spoiler:

Spoiler

Oxygen Purifier-gas.jpg

Oxygen Purifier-liquid.jpg

Oxygen Purifier-power.jpg

The room made of insulated tiles and the thermoregulator setup are the important parts of the design, things surrounded in abyssalite are just external support systems to get the thing running in a steady state. (power, hydrogen to fill the radiator, a supply of polluted oxygen, void for produced oxygen gas)

I used two thermoregulators made of wolframite in parallel as suggested earlier to reduce temperature spikes and help transfer heat more quickly to the oxygen room. The cooling room is square because I didn't really like seeing wasted space with the staircase designs I saw before. It might be a bit on the large side but since this is a closed hydrogen loop I wanted to give myself some margin to avoid freezing any hydrogen in pipes.

Some damage was taken by the regulators to reach the steady state because of the tiny room of oxygen they're in, when I forgot to paint cooler oxygen in there in time. Hopefully that won't be a problem when it is built connected to a larger base. In the steady state with 22C polluted oxygen as input, the hydrogen reaches about -213C when it enters the cooling room and heats up to -199 by the time it exits the room.

An atmo switch inside the cooling room can stop the regulators when polluted oxygen pressure drops too low (prevents hydrogen breaking the pipes), and a thermo switch above the regulators can detect overheating and shut them off to give the air some time to cool down.

Setup should be fairly easy since it doesn't require vacuuming any rooms, having the entrance to the cooling room at the bottom while building it should be enough to prevent any carbon dioxide from building up there.

Link to comment
Share on other sites

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.

×
  • Create New...