Refinery based crude oil cooking w/out aquatuner or volcano

[ruhrohraggy]

Decided to take a crack at this. Seen a fair few questions asking if this is doable. I did a lazy search for something similar, and did not come up with anything that piqued my interest. I'm sure someone should now direct me to their build if they have one. (yay)

I'll opt for a "brief" explanation, and risk complaints / confusion for not being overly specific.

Core concept : Use only metal refinery to cook crude. No aquatuners, no volcanos, none of that.

Personal agendas : I barely want to look at this thing after it's built...I just want it to give me steel and cook the petrol. It's early / mid-game, I could care less about perfect efficiency.

Assumptions : This is used in a play-through during early game steel production. No petrol or steel is acquired before this point. No space age materials are used, and no ceramic.

The crude pumped in is at a typical temp of ~200F. To clarify, this is a one time deal. A general amount of crude is pumped into a reservoir, and then cooked. I'll work on finding the approx. amount of crude needed to process ~30t of steel and then cook. This would be enough steel to bunker off all of space and make a rocket or two.

Power : I was testing this, so I didn't really care about simulating realistic in-game power. For running this in-game, I'd use 2x Coal gen + sweeper / smart battery.

How it works from left to right :

Use gold amalgam (preferable) to initially pump. Swap to steel pump ASAP. Material for insulating tiles and insulated piping was igneous rock. Granite works fine too.

Radiant piping is aluminum. Tempshift plates can be whatever. Diamond, aluminum, granite. Just something to help even out temps.

Process enough petrol to feed refinery coolant.

Shut off oil refinery and pump once coolant loop is established.

*Sip tea*

Wait for reservoir to come up to 763F and cook.

Use valve to bypass refinery and continue looping once 763F max safe time established.

Watch crude cook. Diagonally tile in glass windows / tempshift to and extract surplus heat as power with steam turbine.

 

 

And, cooked. ~ 38,000 kg of crude cooked into petrol. 27.9t of steel was refined over 32 cycles...Which is surprisingly close to my target goal considering the complete not-caring about math for this one. I could assume that if I filled a 4x12 tile full of crude @ 870 kg/tile, this would give me the desired 30t of steel.

Interestingly, this amount of obtained petroleum is enough to run 1 petrol generator constantly for 31 cycles. 38,000 kg / (2kg/s * 600s/cycles) = 31.66 cycles.

@LucidFugue To answer the question regarding how long this would take to cool to a usable temperature using a steam turbine, not long.

As predicted, given waters high heat capacity, pumping in room temperature water next door drastically cooled the petrol pool.

I added 4,000kg of water at 72F, or 21.1C and ran the steam turbine constantly for cooling purpose. It would probably take longer if you wanted to actually use the power, and that would be fine with me.

2 Cycles later, we can put a steel pump back in.

 

[Gus Smedstad]

While it's an idea worth exploring, you took a very brute force approach to this. As we discussed in the last thread on the subject, if you want it to be sustainable in a real game, you really need to make a counter-flow heat exchanger to reduce the amount of heat you need, and incidentally to cool the output if you want to use the petrol for something other than generators.

I don't think you need the multi-step refinery process to get the end result you want, which is petrol reliably heated above 400 C to act as your final heating element. Nor do you need sacrificial pumps if you're using a counter-flow process where you don't pump anything until it's back to roughly the temperature of the input crude oil.

If I were to take a stab at this, I'd modify one of John Francis's designs, replacing the lava heat source with a radiant pipe loop with >450 C petroleum.

[ruhrohraggy]

47 minutes ago, Gus Smedstad said:

While it's an idea worth exploring, you took a very brute force approach to this. As we discussed in the last thread on the subject, if you want it to be sustainable in a real game, you really need to make a counter-flow heat exchanger to reduce the amount of heat you need, and incidentally to cool the output if you want to use the petrol for something other than generators.

I don't think you need the multi-step refinery process to get the end result you want, which is petrol reliably heated above 400 C to act as your final heating element. Nor do you need sacrificial pumps if you're using a counter-flow process where you don't pump anything until it's back to roughly the temperature of the input crude oil.

If I were to take a stab at this, I'd modify one of John Francis's designs, replacing the lava heat source with a radiant pipe loop with >450 C petroleum.

To your first point. Yes, of course. This is designed to be a one off process that takes advantage of the early to mid-game desire for steel. It's not meant to be sustainable. It's more like...I need to make 20-30t of steel without space-materials, what can I do with the heat? Well, you can cook some crude.

As to the second point. Yeah I made an Oopsie. Specific heat of petrol is 0.978 (DTU/g)/F vs. crude which is 0.939 (DTU/g)/F.

Got those mixed up.

It would be more efficient to simply convert a batch of petrol, and let the petrol loop run until crude is cooked.

However, if one were to favor the refining process over the crude cooking, then the way I originally had it setup would be more preferable.

Basically, remove everything to the left of the oil refinery and decon the pump once the process is going. I'll fix this oversight asap.

Aaaaand....Fixed. Should work without all the extra steps. Again, I had mixed up the specific heat for some reason. Had fun automating bypasses though.

Also learned that a 1 g/s valve loop with a liquid detector on it can be useful for limiting how much a prior process will input to your system, and keep a passive loop going. So, not all wasted energy.

[Zarquan]

What is your mechanism for getting the petroleum out of that room?  A liquid pump made of steel can't survive in there, and that is the best you can get before space.

Also, you probably do want that counterflow setup heating up the incoming crude oil with the outgoing petroleum.  If you make it a physical construct (where petroleum flows out of pipes and crude oil flows in pipes, you can use the crude oil to get the petroleum down to a temperature that you can pump it out.

Finally, what is this F unit you speak of?  It sounds like a silly unit.  /s

[LucidFugue]

Interesting, though I have two questions:

1. Last time I tried this the hot petroleum was destroying my liquid shutoffs, though now I think on it it was likely because the shutoff was placed between the refinery and the boiler in a VERY compact setup. What temp is your shutoff getting to once it is going? 

2. Possibly the same point as Gus made already, but since you've cooked your steel pump in the reservoir, is the plan to run a steam turbine until the petroleum is cool enough that you could immerse a new pump in there? Seems like it'll be a while before you can actually use that petroleum for anything without changing what's going through those radiant pipes. 

[Lilalaunekuh]

36 minutes ago, ruhrohraggy said:

To your first point. Yes, of course. This is more of a one off process that takes advantage of the early to mid-game desire for steel. It's not meant to be sustainable. It's more like...I need to make 20-30t of steel without space-materials, what can I do with the heat?

My early/mid game solution is and was always the same:

Placing the metal refinery right before my electrolyzer setup.

 

I did mess with a refinery based petroleum cooker aswell [but with access to space materials], but my petroleum consumption was always higher than the amount of metals I had to refine.

[Maybe I should reconsider this since metal ore is falling from the sky^^]

 

My conclusion: It´s a nice idea and fun to build it once, but to satisfy some reasonable petroleum needs you need a counter flow heat exchanger.

=> But if I go for a "perfect" (/ space material) heat exchanger, then the metal refinery isn´t really needed anymore ...

 

3 minutes ago, LucidFugue said:

1. Last time I tried this the hot petroleum was destroying my liquid shutoffs, though now I think on it it was likely because the shutoff was placed between the refinery and the boiler in a VERY compact setup. What temp is your shutoff getting to once it is going? 

You can build a 1x2 room without an atmosphere and place the shutoff inside.

[Build a 3x4 or 4x3 box without the corners and deconstruct the middle 2 tiles afterwards.]

=> A shutoff inside a vacuum will not transfer heat with the pipe content

[ruhrohraggy]

15 minutes ago, Zarquan said:

What is your mechanism for getting the petroleum out of that room?  A liquid pump made of steel can't survive in there, and that is the best you can get before space.

Also, you probably do want that counterflow setup heating up the incoming crude oil with the outgoing petroleum.  If you make it a physical construct (where petroleum flows out of pipes and crude oil flows in pipes, you can use the crude oil to get the petroleum down to a temperature that you can pump it out.

Finally, what is this F unit you speak of?  It sounds like a silly unit.  /s

Your points are similar to Lucids...I'll cover both below.

F is Fahrenheit. I can spell it correctly, so I feel entitled to use it.

I'm not solving engineering equations, therefore, I use Fahrenheit. If I was doing Statics / Dynamics or real Thermodynamics, then I'd use Celsius or Kelvin...because screw the imperial system IRL.

13 minutes ago, LucidFugue said:

Interesting, though I have two questions:

1. Last time I tried this the hot petroleum was destroying my liquid shutoffs, though now I think on it it was likely because the shutoff was placed between the refinery and the boiler in a VERY compact setup. What temp is your shutoff getting to once it is going? 

2. Possibly the same point as Gus made already, but since you've cooked your steel pump in the reservoir, is the plan to run a steam turbine until the petroleum is cool enough that you could immerse a new pump in there? Seems like it'll be a while before you can actually use that petroleum for anything without changing what's going through those radiant pipes. 

1. Vacuum

2. Yeah, Steam turbine. I will figure out exactly how long it takes and get back to you on that. But yes, use a steam turbine to cool it back down.

I personally only really use petroleum for rockets. Power I usually handle with coal / Solar / Nat-Gas. With launch upgrade... scavenging random pools of ethanol.

@Lilalaunekuh

Yeah you pretty much nailed it. I typically sack a cold biome below my base for the extra water using the metal refinery. Plenty of decent options.

I understand counter-flow heat exchangers, but like you said, why bother using a metal refinery for something like that.

This was more of a fun idea, and I kept seeing posts asking "Can you use metal refinery to cook crude?" Yeah, you sure can.

 

[Lilalaunekuh]

12 minutes ago, ruhrohraggy said:

This was more of a fun idea.

I like your designs and your work with space scanners is amazing

I prefer more practical builds even if I over-engineer them to a point it isn´t worthwhile anymore.

Spoiler

Till the last change to meteor showers, I always solved my bunker door automation WITHOUT any space scanners^^

[You could use just clocks to determine "save" launch and re-entry windows for my rockets. Don´t get me wrong this is still possible, but quite a lot trickier if you don´t want your rockets to wait a couple cycles if there was already a launch window.]

 

 

 

 

PS: If your in the mood here is my current "engineering project":

Spoiler

Turn the gravitas into a (fully) automated tropical pacu farm aquarium.

Goals:

1. Keep the gravitas clean.
2. Allow for an optional population increase with a fed pacu.
3. Allow for an optional husbandry training.

 

 

[ruhrohraggy]

12 minutes ago, Lilalaunekuh said:

I like your designs and your work with space scanners is amazing

I prefer more practical builds even if I over-engineer them to a point it isn´t worthwhile anymore.

  Hide contents

Till the last change to meteor showers, I always solved my bunker door automation WITHOUT any space scanners^^

[You could use just clocks to determine "save" launch and re-entry windows for my rockets. Don´t get me wrong this is still possible, but quite a lot trickier if you don´t want your rockets to wait a couple cycles if there was already a launch window.]

 

 

I like your designs as well, I thought the concept of using critter cooling was interesting and hilarious. It also seemed pretty darn good, especially with shove vole eggs.

As for non-scanner automation. I assume you kept track of 10 Cycle Angry times vs. 4 Cycle peace times? I've seen it mentioned, and it seemed interesting, but I never got around to it. What threw me off was the fact that breaks and meteor shower duration were a randomized range of values....and that was annoying to someone trying to optimize their solar power.

[Lilalaunekuh]

17 minutes ago, ruhrohraggy said:

I assume you kept track of 10 Cycle Angry times vs. 4 Cycle peace times?

Yes.

Having 4 out of 14 cycles with open doors is enough to run a fully automated rocket, but if you want to go for solar power this isn´t really an option.

 

PS: About critter based cooling:

One breeding stable with 8 hatches produces enough eggs to keep the population up and cool down my 300g/s gold volcano.

[No water lock is required and not even ceramic^^]

Works like a cooling solution and a killing chamber.

=> If you provide a "natural" tile above 70°C for the hatches to burrow, it works really well^^

Spoiler

In the 50 cycles dormancy period will the volcano cool down below 70°C and there might be up ~15 hatches waiting to die soon after the volcano starts errupting again.

 

My next implementation will be a gulp fish aquarium inside my base.

[Gulp fish will absorb heat and die when the temperature reaches 25°C, so they are a good solution if you want to stabilize the temperature in a liveable range without "real" temperature sensors.]

 

 

[0xFADE]

Counter flow designs while more work are quicker and result in a usable temperature when the petrol comes out. Usable with steel pumps at least. 

[ruhrohraggy]

36 minutes ago, 0xFADE said:

Counter flow designs while more work are quicker and result in a usable temperature when the petrol comes out. Usable with steel pumps at least. 

True. Though I did finalize my post. Unfortunately I had a bit of an oversight and had to flip my design plan to favor cycling petroleum immediately, instead of using crude up front...Had to put my post back in piecemeal.

If you corner tile in some diamond windows and add a standalone steam turbine, you can stick a steel pump back into the pool in ~ 2-3 cycles, and get some power back from the steam turbine.

[Aelfled]

1 hour ago, Lilalaunekuh said:

Yes.

Having 4 out of 14 cycles with open doors is enough to run a fully automated rocket, but if you want to go for solar power this isn´t really an option.

 

PS: About critter based cooling:

One breeding stable with 8 hatches produces enough eggs to keep the population up and cool down my 300g/s gold volcano.

[No water lock is required and not even ceramic^^]

Works like a cooling solution and a killing chamber.

=> If you provide a "natural" tile above 70°C for the hatches to burrow, it works really well^^

  Hide contents

In the 50 cycles dormancy period will the volcano cool down below 70°C and there might be up ~15 hatches waiting to die soon after the volcano starts errupting again.

 

My next implementation will be a gulp fish aquarium inside my base.

[Gulp fish will absorb heat and die when the temperature reaches 25°C, so they are a good solution if you want to stabilize the temperature in a liveable range without "real" temperature sensors.]

 

 

someone call the RSPCA!

[Lilalaunekuh]

1 hour ago, Aelfled said:

someone call the RSPCA!

Is it just me or wouldn´t you consider my way of killing critters more humane than most other alternatives (except dieing of old age).

• I only kill adult critter.
• My hatches die (partialy) on their own terms.
  • => It´s more like placing the (adult) critter in a room with an option for suicide

[Only if a hatch decides to burrow into a to hot tile will it die.]

[Gus Smedstad]

12 hours ago, ruhrohraggy said:

It's more like...I need to make 20-30t of steel without space-materials, what can I do with the heat? Well, you can cook some crude.

OK, I totally get that motivation, since I've recently gone through the early stages of a colony again.

However, this project, at least as presented, was pretty darned big. I want my one-off "what to do with that heat" projects to be cheap, small, and easy. Big, complicated, and expensive is for sustainable stuff. Though I can see you've revised the screenshots to something less massive.

Speaking of which, once I get crude oil, I'm not thinking one-off short term solutions to metal refinery heat. Crude oil means I can set up a turbine for long-term sustainable use of the metal refinery.

Now that we're talking about this, I'm imagining it would be possible to make a long-term sustainable solution that brute-force heated the crude oil up to 405 C, and then shunted the heat to a turbine for power until it was down to 125 C, where the turbine shuts off. Essentially the sort of heat-to-power setup I have now, only with an oil cooker as an intermediate step.

A sustainable solution would use an airlock door as a trapdoor, dropping the finished petroleum into a tank where it can be pumped out. That keeps it away from pumps until it's a safe temperature.  

The challenge would be to make it small. Even in the revised screenshots, that's a pretty big tank of petroleum. Each tile is what, 700 kg * 1.69 DTU/g/C * (405 - 100 C) = 360 million DTU. A steel run is 93 million DTU per 100 kg, so each tile in the tank requires that you make 380 kg of steel to make petroleum. I'd want my working tank to be maybe 3-4 tiles of oil so it cycled after about 1 ton of steel.

[ruhrohraggy]

14 hours ago, Gus Smedstad said:

However, this project, at least as presented, was pretty darned big. I want my one-off "what to do with that heat" projects to be cheap, small, and easy. Big, complicated, and expensive is for sustainable stuff. Though I can see you've revised the screenshots to something less massive.

Yeah, it probably looks bigger than it really is. (dad joke...)

I didn't think it was too big. Stripping everything else away leaves us with :

Spoiler

If I did some re-arranging of the valve location and used conductive wire, I could probably shrink this a little bit more.

I guess this was more of a proof of concept. I never planned on using this myself...but as I look at it, I'm becoming more attracted to what it's giving me for the minimal effort / brainpower spent to set it up.

I'm really not a big fan of implementing long-term systems too early. To me, a long-term system is one that gets put in, and stays in. I am truly sick of ripping out *complex* contraptions that become obsolete once space-age materials are acquired.

This one is simple, and would take the dupes less than a cycle to tear out.

If i were truly to go long-term with crude-oil, it would be with a sour gas condenser.

[Wild Marker]

I made one myself six months ago, It made me learn about "blob" cooling, which is like drip cooling but with blobs and the fact that they break and don't cool if they hit a wall. The coolant gets so hot that I had to place it in vacuum so it wouldn't exchange heat with the outside, which lead to this crazy contraption. Made a reddit thread about it so I'm just posting it because I can't be arsed to reformat the thing. It's probably outdated in some capacity and frankly making it was a learning process for a lot of mechanics so there's likely room for improvement, but you might like some of the ideas.

 

[ruhrohraggy]

8 hours ago, Wild Marker said:

I made one myself six months ago, It made me learn about "blob" cooling, which is like drip cooling but with blobs and the fact that they break and don't cool if they hit a wall. The coolant gets so hot that I had to place it in vacuum so it wouldn't exchange heat with the outside, which lead to this crazy contraption. Made a reddit thread about it so I'm just posting it because I can't be arsed to reformat the thing. It's probably outdated in some capacity and frankly making it was a learning process for a lot of mechanics so there's likely room for improvement, but you might like some of the ideas.

 

Interesting idea. At least one major thing that's outdated is wheezes though. Gotta feed 'em phosphorite now. Might be able to use some corner tile trickery to get a dupe or sweeper to feed 'em while still trapped in the hydrogen pocket.

The core concept seems interesting to me, and I like learning as many mechanics as possible. The design itself probably holds no real use to me in a play-through though. Unfortunately I'd have to see how blob cooling works and watch the interaction of it myself to get the point.

If there was a more abstract demo on the method, I'd better be able to understand it's usefulness.

[mathmanican]

8 minutes ago, ruhrohraggy said:

I'd have to see how blob cooling works and watch the interaction of it myself to get the point.

It's the same principle as the bead pump, but for the application in @Wild Marker's reddit post, I think you'll find the EZ Bead simpler.  Great to see others posting about it, and seeing another name by which to call it (blob cooling). Thanks for the hidden gem @Wild Marker. I'll have to hunt out more posts under this name.  Have you seen any others?

FYI, @Wild Marker, you can cool the refinery by placing a tiny amount of liquid in the left corner on a solid tile. The contents inside the refinery interact with the middle tile under the refinery.  Then you can cool the refinery without blobs or beads. Both work great. See @Sevio's awesome explanation of this principle here

 

[ruhrohraggy]

33 minutes ago, mathmanican said:

It's the same principle as the bead pump, but for the application in @Wild Marker's reddit post, I think you'll find the EZ Bead simpler.

 

So, basically using a mesh tile to force liquid into a state which can interact with it's surroundings. Seems simple enough.

Haven't really bothered to delve into more advanced liquid / gas mechanics until recently...hence my ignorance.

[Wild Marker]

14 hours ago, ruhrohraggy said:

Interesting idea. At least one major thing that's outdated is wheezes though. Gotta feed 'em phosphorite now. Might be able to use some corner tile trickery to get a dupe or sweeper to feed 'em while still trapped in the hydrogen pocket.

The core concept seems interesting to me, and I like learning as many mechanics as possible. The design itself probably holds no real use to me in a play-through though. Unfortunately I'd have to see how blob cooling works and watch the interaction of it myself to get the point.

If there was a more abstract demo on the method, I'd better be able to understand it's usefulness.

 

Yeah it's certainly outdated, it was a thing I tried a while ago and I don't think I'd make it again. I spent like a week refining the damn room to make it work. 

 

13 hours ago, mathmanican said:

It's the same principle as the bead pump, but for the application in @Wild Marker's reddit post, I think you'll find the EZ Bead simpler.  Great to see others posting about it, and seeing another name by which to call it (blob cooling). Thanks for the hidden gem @Wild Marker. I'll have to hunt out more posts under this name.  Have you seen any others?

FYI, @Wild Marker, you can cool the refinery by placing a tiny amount of liquid in the left corner on a solid tile. The contents inside the refinery interact with the middle tile under the refinery.  Then you can cool the refinery without blobs or beads. Both work great. See @Sevio's awesome explanation of this principle here

 

 

I didn't know about the EZ bead, it's so much simpler! Also that refinery trick is slick, isolating the tile with the contents and using liquid on the other tiles is a great idea. I might revisit the oil-cooled vacuum refinery with this as inspiration.

 

No other instances of "blob cooling" I'm afraid, I kinda came up with the name (based on the old "drip cooling") while making that thread as I had never seen it before  The EZ bead is so much easier though.