Optimizing Oil industry

[TehDwArF]

I'm working on an expandable oil processing line and I wanted to optimize based off a few conditions. Sadly I can't seem to import from Excel to Google Sheets easily so I can't link my maths, instead I'll have to summarize.

I also have a couple of things I can't find the answers to.
What is the indirect power production of fuel in boilers (MJ to KW or how much power one solid fuel produces)?
How much liquid can pass though a pipe at full flow (how many Oil Refineries can one pipe supply before starvation starts)?

Comparison of Lubricant output
Oil Processing obviously takes the cake on this one.

Comparison of Petroleum output
Advanced Oil Processing is better in both output of Petroleum and lower power consumption.

Assuming 100% cracking of all resources. Numbers based off 5 "Time" (1 crafting cycle for all factories).
Oil Processing:

Code: Select all

4 Oil Refineries
3 Heavy Oil Cracking plants
7 Light Oil Cracking plants

Total of 30 units of Petroleum

Advanced Oil Processing:

Code: Select all

4 Oil Refineries
1 Heavy Oil Cracking plants
7 Light Oil Cracking plants

Total of 36 units of Petrolium

Comparison of Solid fuel output
Advanced Oil Processing is better, my gut says that cracking Heavy Oil is better then not, cracking Light Oil is a clear net loss.
DISCLAIMER: Not sure how to compare power gain of extra Solid Fuel vs power loss of higher consumption. Basically, is 1 Solid Fuel per 1 "Time" enough to power 8 chemical plants. If anyone knows the conversion or where I can find it please let me know.

Numbers based off 15 "Time" (3 crafting cycles for Refineries and Cracking, 5 cycles for Solid Fuel production).

Oil Processing, no cracking:

Code: Select all

20 Oil Refineries
18 Heavy Oil Solid Fuel
36 Light Oil Solid Fuel
24 Petroleum Solid Fuel

Total of 390 units of Solid Fuel

Advanced Oil Processing, no cracking:

Code: Select all

20 Oil Refineries
6 Heavy Oil Solid Fuel
54 Light Oil Solid Fuel
33 Petroleum Solid Fuel

Total of 465 units of Solid Fuel

Advanced Oil Processing, Heavy Oil Cracking only:

Code: Select all

20 Oil Refineries
5 Heavy Oil Cracking plants
63 Light Oil Solid Fuel
33 Petroleum Solid Fuel

Total of 480 units of Solid Fuel

Best ratios ignoring production of lubricant.
To get the highest throuput of either Solid Fuel or Petrolium (not both at once), the following ratios are the best.

Code: Select all

4 Oil Refineries
1 Heavy Oil Cracking plants
7 Light Oil Cracking plants
~13 Light Oil Solid Fuel
~6.5 Petroleum Solid Fuel

[jorgenRe]

I would say the smartest thing to do is cracking all the heavy oil to light oil with 2 productivity 3 . This will be much more efficient.

[DarkenDragon]

the only question I have is, do you really require that much output of oil products?

Im in the 4th tier of science production (alien science packs) and I only run 3 refineries and 1 of each. I dont even bother cracking heavy oil into light oil anymore either. so what does each make,

well all oil products makes solid fuel, though each at a different ratio.

Heavy oil will make either lubricant or cracked into light oil. lubricant is only used for electrical engines and express belts, both of which you dont really need excessive amounts of, (unlike copper cables or electronic circuits)

Light oil can only be cracked into petroleum and nothing else, so I use this to make my solid fuel with since this is usually the only oil product that gets stored the most out of the 3.

Petroleum is used to make either sulfur or plastics, and sulfur is used to make explosives or sulfuric acid. and then sulfuric acid is used to make processors or batteries, and each of the items mentioned here takes longer to create than the previous items to build, thus you can make a 1 to 2 of each and never run out or at least I never had that problem.

so all in all, you really dont need such a large scale amount of oil processing at all. there's no point in making more production of an item if there is no demand for it. otherwise it'll just be stuck in storage for no reason.

[tralala]

If you go mass module production you need lots of petroleum. Modules need lots of adv. circuits which need lots of plastics which needs lots of petroleum gas. So if you're going to complete the rocket defense or just want a ton of (tier 3) modules you'll need lots of refineries.

[TehDwArF]

the only question I have is, do you really require that much output of oil products?

The real question is how awesome would it look... I'm planning on not using any solar so I want to make sure I have all the production in place to supply it all. That said, even without that plan, knowing the optomized ratios could be interesting for some people regardless of scale, hence the rough count for only 4 refineries.

Also, I just realized this is all flawed. Refinery and Chemical Plants have different base speeds. I'll have to redo the math....

[therapist]

I've been wondering, is it more efficient (oil cost wise not time wise) to make solid fuel out of light oil, or crack that light oil into petrol and make the solid fuel out of petrol?

[starxplor]

Petro needs 2 to make one solid fuel, light oil needs 1. There is no reason to make solid fuel from petro unless you run out of light oil somehow, but then you have bigger problems and probably need more pumps.

[therapist]

Petro needs 2 to make one solid fuel, light oil needs 1. There is no reason to make solid fuel from petro unless you run out of light oil somehow, but then you have bigger problems and probably need more pumps.

Okay cool, I have the game open so I can add a little more "Oil Math"

4 Heavy + 3 Water = 3 Light Oil
3 Light + 3 Water = 2 Petroleum

1 Light Oil -> 1 Solid Fuel
2 Heavy Oil -> 1 Solid Fuel
2 Petroleum -> 1 Solid Fuel

This means that, indeed, Light Oil is the best resource to turn into Solid Fuel. I was under the impression that cracking Heavy to Light or Light to Petrol was efficient, but actually, some of the product is lost in the process. I guess I must have assumed from the little picture that 1 Heavy = 2 Light and 1 Light = 2 Petrol, but that is NOT TRUE. Thank you for pointing this out for me, now I know only to resort to cracking after my storage is full in order to prevent a stall to the refineries.

[starxplor]

Using piping priorities you can design it to automatically do this overflow processing. I currently have a setup where the heavy oil flows to both a lube processor and a cracking plant. The heavy oil prefers the lube path so only when i fill my tank of lube or over-produce heavy does it actually flow into the cracking plant.

Play around with the liquid distribution mechanics a bit and you can find easy ways to do this.

[therapist]

Interesting, can you screenshot how you set up these "pipline priorities".

I myself have been using a "T" in the pipe, one path of the T is the liquid input, one path of the "T" has a pump, (which makes it the preferred path), and the third and final branch of the "T" goes out to cracking or wherever i want the "overflow" or "excess" to be directed to.

Is this the setup you use or are you doing that trickness with storage tanks I've heard others speaking about? I still don't understand the storage tank way to set up "preferred pipelines".

I think the GIF-tutorial series needs to get cracking on all the techniques we are attempting to hash out here.

[starxplor]

I will post a screen shot later tonight when I get home. I am not using tanks in this particular example, nor am I using pumps. I believe it is a similar phenomenon to the belt location placement of inserters.

I am sure pumps would be more intuitive since they visually point to what liquids are doing.

[TehDwArF]

I have recalculated the optimized ratios for the oil industry taking into account the speed modifier of each building. The optimized paths are still the same as before but the exact ratios are (very) slightly different.

Comparison of Petroleum output
Advanced Oil Processing is better in both output of Petroleum and lower power consumption.

Assuming 100% cracking of all resources. Numbers based off 5 "Time" (1 crafting cycle for all factories).
Oil Processing:

Code: Select all

5 Oil Refineries
3 Heavy Oil Cracking plants
7 Light Oil Cracking plants

Total of 7.5 Petrol per "Second"

Advanced Oil Processing:

Code: Select all

5 Oil Refineries
1 Heavy Oil Cracking plants
7 Light Oil Cracking plants

Total of 9 Petrol per "Second"

Comparison of Solid fuel output
Advanced Oil Processing is better IF 75 solid fuel produces more power then 8 chemical plants consume in one minute (my gut says says it does).

Advanced Oil Processing, no cracking:

Code: Select all

25 Oil Refineries
6 Heavy Oil Solid Fuel
54 Light Oil Solid Fuel
33 Petroleum Solid Fuel

Total of 38.8 Solid Fuel per "Second"

Advanced Oil Processing, Heavy Oil Cracking only:

Code: Select all

25 Oil Refineries
5 Heavy Oil Cracking plants
63 Light Oil Solid Fuel
33 Petroleum Solid Fuel

Total of 40 Solid Fuel per "Second"

Best ratios ignoring production of lubricant.
To get the highest throuput of either Solid Fuel or Petrolium (not both at once), the following ratios are the best.

Code: Select all

5 Oil Refineries
1 Heavy Oil Cracking plants
7 Light Oil Cracking plants
~12.5 Light Oil Solid Fuel
~6.5 Petroleum Solid Fuel

[starxplor]

In this image, you can see where the heavy oil goes, to a cracking plant and a lube plant. As it is, the heavy oil goes into lube plant first, and only spills down into the cracking plant if the lube plant is full of heavy oil, either from too much heavy oil production, or from a full tank of lube backing up the export pipe.

[therapist]

Hmmm, most curious. I believe you when you say it works, although the logic of it I do not comprehend.

In my vision of fluid mechanics, after that "T", the one you see just before the lubricant chemical plant, the liquid would be spread equally (pressure wise) in both directions. So, in my mind, if 50 pressure units of liquid were to enter the "T", 25 would be forced towards the lubricant plant and an equal 25 would be forced to the Heavy Oil cracking facility. What you seem to be saying is that, in reality, for some reason, all 50 end up being processed thru the lubricant plant, unless that plant's output is blocked, in which the fluid is diverted to the cracker.

I wonder why this is. Perhaps after the "T", if pressure remains equal on both of the output sides of the "T", the amount of "pressure space" one one output of the "T" is 10 (or whatever the input buffer of the lubricant plant is), and the other side of the "T" has a "pressure space" for 10 (1 pipe segment) + 10 (the other pipe segment) + 10 (input of the cracking plant) for a total of 30. This means that, if my assumptions are correct, and the pressure remains equal after the "T", the lubricant plant is simply filling first.

Would it be possible for you to run a little experiment for me? I am curios what wold happen if we take your exact setup, but me move the Pipe that feeds both plants down 1 square. See where the Heavy Oil comes out of the refinery(ies) and then goes underground before entering an "L" section? I am curios what would happen in your setup if you moved the "L" down one space, and when the pipe emerged on the other side, it would be in the exact middle of both plants. If my assumptions about the game's pressure mechanics is correct, this setup would cause both chem plants to fill at an equal rate.

Anyway, maybe you could upload the save file or I could just replicate the save myself in sandbox mode, but I am curious if the only requirement to give one chem plant "preference" over another is to add extra length to the "non-preferred" side. This would be the key to teaching others how to create and understand how to give one machine preference over another in the liquid transport system.

In real life, fluids do not work in this way, and pressure should be equal in the input buffers of both the lubricant plant AND the cracker and even in each individual pipe, causing them to fill at exactly equal rates, no matter how much distance the pipes cover. Basically, if 50 fluid enters a system, the 50 should be equally divided between every pipe and every buffer in the entire system if the fluids obey the laws of physics. Every buffer and every pipe should have the same exact amount of liquid, and thusly pressure, at all times because this is how pressure works. Pressure should not be able to build up on the side of the liquid system where the lubricant plant is, without causing this pressure to be forced to become equal with the side of the loop with the cracking plant on it.

Sorry for the long post, but this behavior is most illogical. Simply confounding.

[starxplor]

Hmmm, most curious. I believe you when you say it works, although the logic of it I do not comprehend.

In my vision of fluid mechanics, after that "T", the one you see just before the lubricant chemical plant, the liquid would be spread equally (pressure wise) in both directions. So, in my mind, if 50 pressure units of liquid were to enter the "T", 25 would be forced towards the lubricant plant and an equal 25 would be forced to the Heavy Oil cracking facility. What you seem to be saying is that, in reality, for some reason, all 50 end up being processed thru the lubricant plant, unless that plant's output is blocked, in which the fluid is diverted to the cracker.

I wonder why this is. Perhaps after the "T", if pressure remains equal on both of the output sides of the "T", the amount of "pressure space" one one output of the "T" is 10 (or whatever the input buffer of the lubricant plant is), and the other side of the "T" has a "pressure space" for 10 (1 pipe segment) + 10 (the other pipe segment) + 10 (input of the cracking plant) for a total of 30. This means that, if my assumptions are correct, and the pressure remains equal after the "T", the lubricant plant is simply filling first.

Would it be possible for you to run a little experiment for me? I am curios what wold happen if we take your exact setup, but me move the Pipe that feeds both plants down 1 square. See where the Heavy Oil comes out of the refinery(ies) and then goes underground before entering an "L" section? I am curios what would happen in your setup if you moved the "L" down one space, and when the pipe emerged on the other side, it would be in the exact middle of both plants. If my assumptions about the game's pressure mechanics is correct, this setup would cause both chem plants to fill at an equal rate.

Anyway, maybe you could upload the save file or I could just replicate the save myself in sandbox mode, but I am curious if the only requirement to give one chem plant "preference" over another is to add extra length to the "non-preferred" side. This would be the key to teaching others how to create and understand how to give one machine preference over another in the liquid transport system.

In real life, fluids do not work in this way, and pressure should be equal in the input buffers of both the lubricant plant AND the cracker and even in each individual pipe, causing them to fill at exactly equal rates, no matter how much distance the pipes cover. Basically, if 50 fluid enters a system, the 50 should be equally divided between every pipe and every buffer in the entire system if the fluids obey the laws of physics. Every buffer and every pipe should have the same exact amount of liquid, and thusly pressure, at all times because this is how pressure works. Pressure should not be able to build up on the side of the liquid system where the lubricant plant is, without causing this pressure to be forced to become equal with the side of the loop with the cracking plant on it.

Sorry for the long post, but this behavior is most illogical. Simply confounding.

I suspect the code for the junctions does a round-robin check of available space in each direction, probably with input being last(in case of changes, like adding a new branch of pipe earlier in line). The other option I have thought of would be the shortest path is prefered.

I will try moving the input down one spot to see what happens.

[ssilk]

As far as I remember, this is known behavior.

I'm nearly sure Kovarex explained that in the forum anywhere... somewhere between v0.9.1 and v0.9.6