Factorio Forums

After hours of experimenting what's the highest efficacy of transporting nuclear heat and converting it into steam while providing a maximum load without any losses, I've created the following setup:


I recommend a mod like Pump Anywhere for problem-free water support if you don't like to landfill. There're two Offshore Pumps required on each reactor line!

For Stress-testing I recommend BT's Waste Electricity.

Benefits:

• 100% reliable & stable load even if satisfaction is in red area
• minimum temperature never below 685°C on the last Heat Exchanger in the row
• adapted power distribution from Nuclear Reactor's power delivery to Heat Exchanger's consumption rate at 100%, no loss of heat, no overload of any reactor
• only one row of HE's per reactor, no unused Reactors somewhere in the middle
• this implicates that this setup is 100% tileable (see section "Disadvantages" below) (see change notes at the bottom)
• the Heat Pipes have the shortest line possible, only 2 tiles of distance to the next Heat Exchanger to prevent waste of heat energy
• the 944 Steam Turbines do not consume 100% of the Steam
• the remaining 124.8 MW of Steam can be used for Coal Liquefaction or other modded recipes
• controlled Steam output via monitored Steam Tanks: Energy production has the priority. Only full Steam Tanks provide additional Steam production. If the "System Ready" tick starts flickering then there's too much Steam consumption
• the more Reactor lines, the more additional Steam production is possible
• 24k Uranium Fuel Cell buffer for endless support with monitored income: if the income drops to zero, there's a global alarm (income is always zero unless the buffer has been filled)
• Steam Turbines are always satisfied with Steam, allowing consumption of maximum 60/s every single one without drops
• Roboport coverage supplies Construction Robots everywhere (in case something should get damaged)
• the "System Ready" signal from the green wire can be used for power switches while the system isn't up (make sure not consuming too much steam to prevent UPS drops if large networks are merged often in a short time)

Disadvantages:

• Logistic Robots cannot work in the middle of the power plant
• scalability suffers Roboport Coverage completely in the middle if the size of 36 Nuclear Reactors is significantly exceeded; however, it is still fully functional
• that being said, this setup cannot work with Logistic Robots; Belts must be used to supply the Reactors
• no control about consumption; it is at fixed vanilla rate (36 fuel cells every 200 seconds)
• Medium Poles must be used for this setup. It was impossible for me to perform a true scalable reactor setup with Substations. The limitation is the size of the Heat Exchangers together with the necessary input Heat Pipe and the output piping system compared with the reactor's size of being 5x5.

The Steam production at the end of the lines can be cut & additional Steam Turbines added resulting in a total of 980. This leads to the maximum of 5.6 GW electrical production and even a higher peak unless all steam buffers are depleted. The system will then show no more "system ready" if it is at full load due to empty steam tanks.

If you have comments, reviews, stimulations or criticism, feel free to let me know.

Blueprint (too large for this forum):
https://justpaste.it/9y46h
(if this link will be dead, I will search for alternatives)

Edith: corrected a typo

Another Edith: Changed Topic to "extensible", since the interpretation about what is "tileable" differs from person to person. This isn't a "just paste & have more power" blueprint.
Changed power description to plain energy production since a modular power plant with extra steam production doesn't seem to be welcome.

vangrunz wrote: ↑Thu Mar 30, 2023 5:12 pm For Stress-testing I recommend BT's Waste Electricity.

You could just use /editor and the EEI

FuryoftheStars wrote: ↑Thu Mar 30, 2023 6:30 pmYou could just use /editor and the EEI

Thanks, I'll keep that in mind and will try out some day. At the moment, it's interesting to me to do "live tests". Since I have an infinite resource mod installed, I don't worry about shortages any more, so the mind is free for experiments.

A few things to think about:

• why do steam buffers exist, while there is no fuel throttling? Buffers are needed for fluctuation in production to bridge higher consumption at a time of lower production, however since this setup always does produce the maximum amount of heat, there is no such thing as lower production.
• the setup isn't able to produce the maximum possible amount of energy. You claim your setup is tileable, so in theory the maximum amount of energy is 160 MW for each reactor. Only the 4 reactors at the ends of the long row produce 40 MW less due to the missing neighbor bonus, You have 36 reactors in your blueprint, so 36 * 160 = 5760 MW if tiled. This is what your setup should be able to produce (and what other blueprints will actually provide). For 5760 MW, you need 576 heat exchangers. These will produce and 576*103.1/s = 59385.6/s steam. To consume this steam, you need 59385.6/s / 60/s = 989.76 or rounded up 990 steam turbines (at least. Usually one will place a few more due to rounding). If you don't tile (this setup isn't actually tileable), you create 35*160=5600 MW and need 560 heat exchangers with (560*103.1/s / 60/s) = 962.3 = 963 steam turbines. You have only 944, which wastes (962.3 - 944) * 5.82 MW = 106.5 MW. So this setup only creates 5.49 GW, not 5.6 GW.
• the setup isn't actually tileable. It can of course be placed next to itself as many times as you want, however that's not what is meant with "tileable". Tileable means you are able to seamlessly put blueprints next to each other and just extend the existing functionality. With nuclear setups, this means it is able to directly connect to the existing reactor rows to benefit from the added neighbor bonus.
• leaving surplus steam for coal liquefaction is inefficient and wastes energy, since it's 500°C hot stream. Coal liquefaction works fine with 150°C steam from an ordinary boiler that can be located directly in the coal liquefaction plant and fueled simply from the byproducts of coal liquefaction.
• consider placing your offshore pumps in pools of water, so you can deconstruct and place them again in the vanilla game without mods. If you make them water well pumps by first placing them in water properly, then filling all water with landfill, you cannot replace them with the vanilla game. So if you made a small mistake and deconstruct a pump by accident, you will not be able to place it again. It's also tedious to first place landfill to create the spots where the offshore pumps are located, then place the pumps, then place the landfill to fill the remaining water. You seem to use a mod for this, however I consider this cheating out of a challenge given by the game developers for good purpose.
• the footprint of your setup is very large with much unused space in between, so you need much roboport coverage and much forest to clear/landfill to place. It's about double the size other designs with the same power generation will use.
• the setup fills the belts completely with nuclear fuel cells just waiting to be grabbed. These inactive fuel cells are not used productively and just laying around.
• the setup isn't resistant to brownouts and cannot start/restart without considerable external power. There isn't a secondary solar powered grid to supply the reactors with fuel, and since you replaced pipes with pumps, you need to supply a considerable amount of pumps with power before you're able to create even one Watt of power. For such a plant, it is expected to auto-start itself.

You might want to take a look to a reactor setup a few threads below in this forum section (2x6 tileable Nuclear Reactor 1.76 GW) and see how it solves the above things. It solves everything I mentioned above, as well as all of what you mentioned as disadvantages. Look at that included screenshot with the 156 reactors for 24.8 GW. This is what is meant with "tileable". It's just placed the same blueprint 13 times directly on top of each other, no additional work.

There are many more reactor designs in this forum and on Reddit for multi-GW plants solving all these challenges with different approaches. I encourage you to search and look at these a bit for inspiration and see how other people solved the given challenges. I don't want to appear as destructive, however I still miss a challenge this setup solves in an interesting and effective new way. You excessively use pumps, however I miss the purpose, because plants work as effective with far less. The mentioned other reactor setup for example provides a solution to the challenge to integrate in one single setup almost all of the challenges solved individually by other setups. The only challenge it doesn't solve is to provide water externally, for which there is another thread with a big plant just a few threads below as well.

For developing such setups, I suggest you create a sandbox game in map editor mode:

• Main Menu->Single Player->New Game->Sandbox
• In the map generator: disable Enemy->Enemy bases, disable Advanced->Pollution, then ->Play
• Research all tech: yes
• Starting items: irrelevant
• Cheat mode: yes
• Always day: yes
• Enter map editor by opening console, then type /editor
  Map editor description: https://wiki.factorio.com/Map_editor Don't worry, in map editor, there is no character running around. Just place stuff.
• Activate Map Editor Settings->UI->Show character tab in controller GUI and Show infinity filters in controller GUI.
• Tweak other map editor settings as desired
• To remove terrain and make the whole world one clean lab: choose the "Surfaces" tool of the map editor menu
• Click "Remove all entities", click "Fill with lab tiles", activate "Generate new chunks with lab tiles"
• The map is paused as default. To make it run like an ordinary map, choose the "Time" Tool and click the "Play" button in the "Speed section"
• The best thing about developing things with the help of the map editor is the ability to pause and single step through every tick with the "Tick once" button, resp. with the corresponding hotkeys.
• Notice the new item category "unsorted" with the surprise box icon. Use the electric energy interface for infinite energy and energy consumption, heat interface for heat production without reactor, infinity chest for infinite items, infinite pipe for infinite fluids. To create fully compressed belts from chests, use the loaders now found in the Logistics item category instead of stack inserter combinations.
• For slightly more inventory space, generate yourself a Power Armor MK2 and wear it.

No mods required for all this. This is still the vanilla game.

Tertius wrote: ↑Fri Mar 31, 2023 1:42 pm [pumps... pumps everywhere... why?]

Because it's this person: viewtopic.php?f=18&t=105648

Edit: Also, in case anyone is unaware (or cares, it's not like it makes much of a difference), the need to open the console and type "/editor" can be removed by just simply using Map Editor -> New Scenario right from the main menu. Same game modes and everything, just immediately starts you with the /editor open right from the start.

Also, if you're just running the new game (via either method) for the purposes of the test building, all of this can be skipped (in other words, it's optional):

Tertius wrote: ↑Fri Mar 31, 2023 1:42 pm

• [...]
• Research all tech: yes
• Starting items: irrelevant
• Cheat mode: yes
• Always day: yes
  [...]
• Activate Map Editor Settings->UI->Show character tab in controller GUI and Show infinity filters in controller GUI.
• Tweak other map editor settings as desired
  [...]
• For slightly more inventory space, generate yourself a Power Armor MK2 and wear it.

Tertius wrote: ↑Fri Mar 31, 2023 1:42 pmwhy do steam buffers exist, while there is no fuel throttling? Buffers are needed for fluctuation in production to bridge higher consumption at a time of lower production, however since this setup always does produce the maximum amount of heat, there is no such thing as lower production.

I didn't know that I have to play Factorio with fuel throtteling. While I have an infinite resource mod (like I stated) there's no need for me to throttle anything.

the setup isn't able to produce the maximum possible amount of energy. You claim your setup is tileable, so in theory the maximum amount of energy is 160 MW for each reactor. Only the 4 reactors at the ends of the long row produce 40 MW less due to the missing neighbor bonus, You have 36 reactors in your blueprint, so 36 * 160 = 5760 MW if tiled.

There's a mistake in calculation:
32 * 160 + 4 * 120 = 5600

This is exactly the value of megawatts the reactors produce in heat.

This is what your setup should be able to produce (and what other blueprints will actually provide). For 5760 MW, you need 576 heat exchangers. These will produce and 576*103.1/s = 59385.6/s steam. To consume this steam, you need 59385.6/s / 60/s = 989.76 or rounded up 990 steam turbines (at least. Usually one will place a few more due to rounding).

Each Heat Exchanger converts 10 MW of heat to steam, so we need for the main lines 16 per reactor, for the two others 12 each.

These Heat Exchangers provide a steam flow of 103/s:

• 12 x 103 = 1236 / 60 = 20.6
• 16 x 103 = 1648 / 60 = 27.5

32 x 27 + 4 x 20 = 944

This is exactly how I set up. You can add another 7236 turbines to gain the maximum of energy production but with slightly reduced steam flow.

You have only 944, which wastes (962.3 - 944) * 5.82 MW = 106.5 MW. So this setup only creates 5.49 GW, not 5.6 GW.

Each turbine produces 5.8 MW. 944 x 5.8 = 5475.2 MGW. The "unused" rest of 124.8 MW (assuming the steam turbines run at full power) can be produced by the heat exchangers.

the setup isn't actually tileable. It can of course be placed next to itself as many times as you want, however that's not what is meant with "tileable". Tileable means you are able to seamlessly put blueprints next to each other and just extend the existing functionality. With nuclear setups, this means it is able to directly connect to the existing reactor rows to benefit from the added neighbor bonus.

It is tileable. You just can cut off the most upper or most lower line (CTRL+X or blueprint), copy a 2 x 160 MW line and add it how often you like. Then you add the "end line" and you're done. You can create at any time a blueprint of a line and tile it to your likings.

Having each line exactly the same heat exchangers & steam turbines would result in a number that cannot match (26.2 turbines each).

leaving surplus steam for coal liquefaction is inefficient and wastes energy, since it's 500°C hot stream. Coal liquefaction works fine with 150°C steam from an ordinary boiler that can be located directly in the coal liquefaction plant and fueled simply from the byproducts of coal liquefaction.

Why is that a waste? The reactors permanently produce heat, no matter how much is consumed. Using boilers will need extra fuel. A power plant that isn't completely satisfied can do this job good.

consider placing your offshore pumps in pools of water, so you can deconstruct and place them again in the vanilla game without mods. If you make them water well pumps by first placing them in water properly, then filling all water with landfill, you cannot replace them with the vanilla game. So if you made a small mistake and deconstruct a pump by accident, you will not be able to place it again. It's also tedious to first place landfill to create the spots where the offshore pumps are located, then place the pumps, then place the landfill to fill the remaining water. You seem to use a mod for this, however I consider this cheating out of a challenge given by the game developers for good purpose.

Using mods is cheating? Why did the developers then allow them?

As you said, you can landfill a pool of water to achieve the same result. Using more than 11 heat exchanger requires a second offshore pump. This is difficult to achieve with the short lines of heat exchangerspipes to get the most heat out of the reactors.

the footprint of your setup is very large with much unused space in between, so you need much roboport coverage and much forest to clear/landfill to place. It's about double the size other designs with the same power generation will use.

Comparisons?

the setup fills the belts completely with nuclear fuel cells just waiting to be grabbed. These inactive fuel cells are not used productively and just laying around.

See above.

the setup isn't resistant to brownouts and cannot start/restart without considerable external power. There isn't a secondary solar powered grid to supply the reactors with fuel, and since you replaced pipes with pumps, you need to supply a considerable amount of pumps with power before you're able to create even one Watt of power. For such a plant, it is expected to auto-start itself.

I could add my solar panel/accumulator setup with a power switch that monitors the steam tanks, like said in the description. And as I also stated, if satisfaction is in red area the system works very well.

You excessively use pumps, however I miss the purpose, because plants work as effective with far less.

Feel free to use other setups. Pumps make sure you have always the maximum flow, so I don't see a problem using things in the game that are given (and therefore, every other comment about pumps usage is superfluous).

Addendum: Without pumps, it is difficult to supply so many turbines, if that's ever possible without requiring much more space than we have in a line with 5 of reactor's height, breaking tilability. Hover over a pump around a storage tank at full load to see flowing speed.

It's not clear why you need so many pumps while a setup like in this post from your previous thread a week ago works just as well completely without a single pump. Steam flows on its own. It achieves exactly the same thing. It's truly tileable without first removing anything, it truly outputs maximum power all reactors are able to create. I don't ask you copy other people's setups, I suggest you use such examples as inspiration and create your own things that perform better, not worse.

You are of course free to use any mods you like to overcome all of the games' challenges you don't like, not care about resources but instead infinite resources. However I am free to remark this is not the vanilla game as the developers conceived it and meant it to be played while they developed it. My personal goal is to play a game as near to the developer's vision as possible, and my above remarks are targeted to this.

I tried to help understanding these original challenges. The game is about automatisation. It's the core and the spirit of the game. Automatisation also means optimization. Optimization means using resources carefully and as sparingly as possible. And because of this, I expect players will try to develop optimized setups, and I tried to show you ways to develop your own optimized setups. Optimized setups are beautiful in a mathematical and engineering sense.
I understand you see the game differently, so I will leave you with that and not bother you with my remarks again.

Tertius wrote: ↑Sat Apr 01, 2023 12:05 amIt's not clear why you need so many pumps while a setup like in this post from your previous thread a week ago works just as well completely without a single pump. Steam flows on its own. It achieves exactly the same thing. It's truly tileable without first removing anything, it truly outputs maximum power all reactors are able to create. I don't ask you copy other people's setups, I suggest you use such examples as inspiration and create your own things that perform better, not worse.

I've seen that. If a setup is better or worse that's in the eye of the beholder.

From my point of view, that's purely technical: what energy is produced and how is it distributed, for example. There's no lack of steam, no buffers will ever run dry (not even at heavy load), and a pump consumes a maximum of 30 kW which might scale with flow speed: so what? In my "Electric network info" they are the 8th in place, long behind lamps, for example. Unused energy in form of heat is converted into steam if necessary. If someone likes to have a buffer, then feel free to modify my setup and add much more storage tanks. I would love to see modifications.

To me, it was important that I only calliper exactly that amount of energy a reactor can generate. Because of their neighbour bonus, the ones at the first & last lines need to be loaded less than the others. The Wiki doesn't talk much about energy heat flowing through a running reactor, only in reference of unfueled ones, using them as a better heat pipe -- what exactly feels as a kind of cheating: why would you build an expensive reactor as a heat pipe replacement if you could use heat pipes instead? Either the mechanic of heat pipes has a lack, or an unfueled reactor should not transmit heat energy, even better as the equivalent heat pipes.

You are of course free to use any mods you like to overcome all of the games' challenges you don't like, not care about resources but instead infinite resources.

Even with limited resources there's nothing to say against a sushi belt where nuclear fuel cells just "lying around". They will be used, definite. What's about "tileable" reactor setups if you don't scale it with your power needed? At first, you need perhaps 4-6, and it will grow with your base. All that's possible with my setup, too.

However I am free to remark this is not the vanilla game as the developers conceived it and meant it to be played while they developed it.

You are a developer and can take an assessment what's the goal or spirit of this game?

One likes PvP, the other PvE. The third looks from a technical view, the fourth likes immersion like hell. Well, that's the diversity that makes a game interesting, Including a highly mod-able aspect all the more.

My personal goal is to play a game as near to the developer's vision as possible, and my above remarks are targeted to this.

Feel free to do so. Live and let live, is my motto.

I tried to help understanding these original challenges. The game is about automatisation. It's the core and the spirit of the game. Automatisation also means optimization. Optimization means using resources carefully and as sparingly as possible.

There's war in the game. War isn't meant to manage sparingly. If you don't invest enough in defence (power production is a necessary part of it) your base will be overwhelmed.

And because of this, I expect players will try to develop optimized setups, and I tried to show you ways to develop your own optimized setups. Optimized setups are beautiful in a mathematical and engineering sense.

My setup is beautiful in math & engineering. It matches all vanilla settings, aside the necessity of some water spots which are -admittedly- difficult to achieve without mods, but it's not impossible.

100% of the energy is used, nothing wasted, nothing overloaded or too undemanding.

I understand you see the game differently, so I will leave you with that and not bother you with my remarks again.

I asked for other opinions, and I thank you for that.

Tertius wrote: ↑Sat Apr 01, 2023 12:05 amIt's not clear why you need so many pumps while a setup like in this post from your previous thread a week ago works just as well completely without a single pump. Steam flows on its own. It achieves exactly the same thing. It's truly tileable without first removing anything, it truly outputs maximum power all reactors are able to create.

Addendum, for clarification:
Using more than one pipe (or a storage tank in substitution) always generates "levels" the game will balance. This balance is contrary to flow, might generate backwater or limiting maximum flow speed (so, in your named setup there're two pipe lines for the steam turbines to circumvent that bottleneck, what's a bad solution from technical view while we have pumps available).

I would go that far to say that a system with pumps on the right places, such as entity outputs & inputs, could be even more UPS friendly. Of course, that's only a theory that might not match reality.

For me, the only or the first line of monitoring in-game playing by simply hovering my mouse over an entity to determine whether my system is good or there's a lack somewhere, is the level of liquid: is the level high/at maximum, I don't have to worry. Is there a lack, I need a deeper view if flow speed is good or there's something wrong. So, the easiest way (for me) is to take a fast look at the levels. This is achieved best with pumps. They do not cost much, neither resources nor electric power, and they make sure that you can pull out the most of your system (literally). So, in opposite, I don't see any reason why I should not use them. Having a power plant you always need electricity, for distribution as well as consumption. The only (legal) scenario is a "cold start" for what you need solar panels/accumulators or another electric grid with boilers, for example. And, like I stated before, they do not consume much electricity. Having only a single beacon for your uranium processing factory is a mass of more energy that's consumed: currently, in my (yet small) base that should grow much more, with liquid setup nearly complete, beacons consume ~500 MW while pumps only consume 2.7 MW. In overall satsifaction of 720 MW, this is just 0.37% we are talking about. Is it really that much? And it is not night, so the lamps with a consumption of about 320 MW are not lit.

I try to explain some reasoning behind creating efficient designs.

You need to concentrate on the task at hand. What is your goal? Again: What is your goal? Formulate this goal on the highest abstract level as possible.

If you want to build a nuclear power plant, what is the goal of a nuclear power plant?

The goal is to produce electricity.

Nothing else. The goal of a power plant is to produce electricity.

How can I reach this goal? By producing electricity. This is the challenge to solve. Produce electricity.
The goal is not to distribute steam. The goal is to produce electricity.

You reach the goal to produce electricity by providing machines that produce electricity. In case of a nuclear power plant: steam turbines.
The secondary goal is to provide them with the least resources spent. You don't want to pay more than necessary. You want to buy as cheap as possible. It's a waste to pay more than necessary.

One of these machines (steam turbine) is able to produce 5.82 MW energy. It uses steam, an intermediate created by the limited resource heat. The other steam resource is water, however water is infinite in the vanilla game. Heat is produced in portions of 40/80/120/160 MW by nuclear reactors. The most efficient heat production is with a 160 MW reactor, so we try to get as many 160 MW reactors as possible. This results in a double row of reactors, as long as possible. If we have an infinite line ("tiled"), every reactor will output 160 MW. This is one end of our electricity production. The other end is the steam turbines.

So the begin produces 160 MW portions of heat. The end consumes heat in portions of 5.82 MW.
So to completely consume a portion of 160 MW, we need a portion of 160 MW / 5.82 MW = 27.49 steam turbines. There are no half steam turbines, so we actually need 28 to fully consume 160 MW. One of these will run with only half utilization.

If we continue to design and add reactors, we will find we have to divide the total steam turbines into groups, so we are able to cut down some of the additional turbines from rounding up. Each group will still have one turbine not completely utilized due to rounding up. But that's fine, since our goal is to fully convert the reactor heat into electricity, which is achieved, and not to fully utilize each turbine. Fully utilizing steam turbines isn't possible, because the ratio of energy production (160 MW) and energy consumption (5.82 MW) isn't an integer.

The goal is achieved, if the total electric power created by all of our steam turbines is equal to the total power generated by the reactors, so the heat->energy conversion is 100% efficient.

I gave an example how to do this, and for some reason that worked without actively managing steam levels. So the person who designed that (it was not me) didn't add any active steam managing with pumps, because it wasn't necessary. Pumps use more resources than pipes to build, need power supply (even more resources to build), need managing (defend against brownouts), so it's efficient and desirable to get away without pumps. The fewer active components you have, the more resistant it is against malfunction.

You try to reach a different goal: You fill the steam turbine steam buffers to the maximum. But that's not required. It serves no purpose, reaches no useful goal. Steam turbines will output maximum power without their buffers filled to maximum. The power plant goal is already reached, if the power output is at maximum. The actual steam level inside is irrelevant, as long as the output is according to our goal. The inside is a black box. Don't micromanage every entity, if the whole is working properly.

By pursuing that different goal, you forfeit efficiency and the maximum power output. You output less electric power than was generated as heat. You also waste material (you happily pay more than required, really?) by adding pumps and storage tanks not necessary. Maximum power output can be achieved without steam level at maximum within every turbine, so maximum steam level is irrelevant.

Tertius wrote: ↑Sat Apr 01, 2023 3:08 pmBy pursuing that different goal, you forfeit efficiency and the maximum power output.

So it is not legitimate to post creations that don't meet a certain definition of efficiency?

Please let me know if I misunderstood the invitation labelled on this forum:

Show your Creations

Tertius wrote: ↑Sat Apr 01, 2023 3:08 pm Fully utilizing steam turbines isn't possible, because the ratio of energy production (160 MW) and energy consumption (5.82 MW) isn't an integer.

oh it is possible to solve the equation, it just gives very annoying result

According to my calculations, if you have lane of 146x2 reactor, that would create 2*(144*160)+(4*120) MW or, 46 560 MW which is exactly the same as 8000 turbines would need. It represent 4656 heat exchangers which can be grouped as 291*16, so it can match 500*16 turbines to make it easier to build you can make a blueprint with 291 heat exchangers for 500 turbines and copy it 16 times next to the reactors

I agree with your advices to round off the number of turbines eventually

@vangrunz
Beware, the 5.82 is a correct value to use for precision coming from here => https://wiki.factorio.com/Steam_turbine
But the 103 is not, it is a rounded value already from 103.0927835 => https://wiki.factorio.com/Heat_exchanger
I found it easier to use 10MW as the value for heat exchangers, as reactor produce in MW and turbines consume in MW, this allow to avoid counting steam at all.

mmmPI wrote: ↑Sat Apr 01, 2023 4:33 pmBeware, the 5.82 is a correct value to use for precision coming from here => https://wiki.factorio.com/Steam_turbine
But the 103 is not, it is a rounded value already from 103.0927835 => https://wiki.factorio.com/Heat_exchanger
I found it easier to use 10MW as the value for heat exchangers, as reactor produce in MW and turbines consume in MW, this allow to avoid counting steam at all.

Ah, thank you.

I've re-calculated each line and it stays at 20 turbines for the "small" and 27 for the "normal" lines. Indeed, it's easier not to calulate steam flow rates but megawatts with the same results.

vangrunz wrote: ↑Sat Apr 01, 2023 4:28 pm

Tertius wrote: ↑Sat Apr 01, 2023 3:08 pmBy pursuing that different goal, you forfeit efficiency and the maximum power output.

So it is not legitimate to post creations that don't meet a certain definition of efficiency?

Please let me know if I misunderstood the invitation labelled on this forum:

Show your Creations

Just don't do any false advertising.

Since you claim tileable please do fix your top and bottom row. While you can have fewer boilers and turbines there you can take half the heat exhcangers for that row and move them out to where they would be for a full row. The few extra heat pipes are not adding anything significant to the price of the reactor. Same with pumps, being so cheap you can use the pumps for a full row, don't replace them with underground belts.

Next the roboports. Replace the steam tanks with an underground pipe. If you don't use any fuel control then they are utterly pointless. Replacing the tanks with an underground pipe give you space to place a roboport in there. You can also move the whole row of heat exchangers away from the reactor making space to a) leave water under the offshore pumps and b) place a roboport. It will raise the reactor temperatur slightly for controlled fueling but you are overheating the reactor anyway.

Do both of those things and you will have a truely tileable blueprint. You can then just place a second copy of the blueprint overlapping the first so that the roboports align (make sure they are placed a multiple of an reactor apart). The overlap will fill the in the short row without any conflicts or having to remove anything before expanding. You won't have to interrupt power production, the reactor will just grow and produce more power.

PS: you also need to put roboports between the steam turbines at the right places by having an undergorund pipe in there at the right spot.

PPS: If you were to add fuel control then you would want to add a buffer chest to every reactor limited to 1 fuel cell. That way you don't get refueling fails when there are gaps in the belt. You can even monitor that every reactor has a new fuel cell waiting and raise an alarm if fuel is low.

To throw in some other comments & recommendations based on my observations:

• As others have mentioned, the blueprint is not tileable as is. I recommend removing the extraneous stuff from the blueprint that is not needed for it to be tileable, such as the uranium input feed, writing in heat pipes, the status lights, etc, etc. If you want, put these in a separate blueprint. Some other minor tweaks (like the end of the belt feed) may be needed to make it truly tileable (if possible, I even suggest removing the idea of a belt loop and have the feed-in belt fork at the beginning of the reactor rows, and then have the return fuel be on separate belts; if need be, you can braid underground belts in the same row by using different tier/color belts).
• You mentioned that there wasn't space to place substations: while this appears to be true with the heat exchangers (which wouldn't be necessary if you weren't so insistent on placing pumps everywhere), there appears to be plenty of space around the turbines? Not a big deal, but felt it worth mentioning.
• You have a speaker set to alarm on lack of fuel on the input belt to the whole setup, but you have a lot on the feed belts themselves plus a huge amount of buffer in chests. Seems like the wrong place to place the alarm condition?
• Your setup has 944 turbines. At 5.82 MW each, this means your plant is actually only capable of producing just under 5.5 GW (5.49408 GW), not the advertised 5.6 GW. Yeah, I get it, the reactors are capable of 5.6 GW, but someone looking at this will be more interested in the actual output, not theoretical that can't actually be reached without editing your setup.
• Pumps between the turbines are unnecessary and a waste, not just in resources and power (and space), but in UPS as well. Setup could be 104 tiles narrower, and sans ~912 pumps (that's 912 fluid entities using UPS that don't need to be there!).
• As this setup is meant to run with the reactors at full throttle unchecked and at no point can the setup consume more steam or heat than what can be produced, all of the tanks and status indicator lights are unnecessary. This all just uses extra UPS that doesn't need to be.

mrvn wrote: ↑Sat Apr 01, 2023 9:21 pmJust don't do any false advertising.

Pardon?

It was never my intention to name something wrong. I have changed the topic & added some change notes.

Since you claim tileable please do fix your top and bottom row.

It's in the description that those two rows are intentional, matching their reactor power.

The few extra heat pipes are not adding anything significant to the price of the reactor.

Reactor's "price"? My goal was a heat pipe line as short as possible.

Same with pumps, being so cheap you can use the pumps for a full row, don't replace them with underground belts.

I don't get this message: underground belts replacing or not replacing pumps?

Next the roboports. Replace the steam tanks with an underground pipe. If you don't use any fuel control then they are utterly pointless.

Like shown in the description, the tanks are used for steam control.

Replacing the tanks with an underground pipe give you space to place a roboport in there. You can also move the whole row of heat exchangers away from the reactor making space to a) leave water under the offshore pumps and b) place a roboport. It will raise the reactor temperatur slightly for controlled fueling but you are overheating the reactor anyway.

I see what you mean. Longer Heat Pipes: no, but I could in fact move the tanks/turbines, along with underground pipes, to place roboports in between without losses. Unless I expand the heat pipes there cannot be a roboport coverage around the reactors.

You won't have to interrupt power production, the reactor will just grow and produce more power.

Power production isn't just interrupted but satisfaction lowers, at least with my setup where a part stays and will be expanded. If someone monitors its power consumption, there should be enough time to expand.

PS: you also need to put roboports between the steam turbines at the right places by having an undergorund pipe in there at the right spot.

Yes, see above, that's a good idea.

PPS: If you were to add fuel control then you would want to add a buffer chest to every reactor limited to 1 fuel cell. That way you don't get refueling fails when there are gaps in the belt. You can even monitor that every reactor has a new fuel cell waiting and raise an alarm if fuel is low.

I didn't occupy with fuel control, since there's actually no true circuit network signal possible to read the electrical energy load.

FuryoftheStars wrote: ↑Sun Apr 02, 2023 5:10 amAs others have mentioned, the blueprint is not tileable as is. I recommend removing the extraneous stuff from the blueprint that is not needed for it to be tileable, such as the uranium input feed, writing in heat pipes, the status lights, etc, etc. If you want, put these in a separate blueprint. Some other minor tweaks (like the end of the belt feed) may be needed to make it truly tileable (if possible, I even suggest removing the idea of a belt loop and have the feed-in belt fork at the beginning of the reactor rows, and then have the return fuel be on separate belts; if need be, you can braid underground belts in the same row by using different tier/color belts).

I got the point: "tileable" is meant to "just place the same blueprint on its end to expand"; changed the topic to "extensible". I have no problem to cut a part of the reactor off to tile another part in for expansion. But, well, we are all humans with different thinkings & likings. Misunderstandings happen.

You mentioned that there wasn't space to place substations: while this appears to be true with the heat exchangers (which wouldn't be necessary if you weren't so insistent on placing pumps everywhere), there appears to be plenty of space around the turbines? Not a big deal, but felt it worth mentioning.

The pumps can stay, see above, expanding just the lines with underground pipes. A good idea, yes.

You have a speaker set to alarm on lack of fuel on the input belt to the whole setup, but you have a lot on the feed belts themselves plus a huge amount of buffer in chests. Seems like the wrong place to place the alarm condition?

Where would it be better?

Your setup has 944 turbines. At 5.82 MW each, this means your plant is actually only capable of producing just under 5.5 GW (5.49408 GW), not the advertised 5.6 GW. Yeah, I get it, the reactors are capable of 5.6 GW, but someone looking at this will be more interested in the actual output, not theoretical that can't actually be reached without editing your setup.

Well, if I'd sell anything for a high price that's not worth it, I would agree that it's "wrong advertisement". See above: changed the description.

Pumps between the turbines are unnecessary and a waste, not just in resources and power (and space), but in UPS as well. Setup could be 104 tiles narrower, and sans ~912 pumps (that's 912 fluid entities using UPS that don't need to be there!)

Without pumps, the steam level is falling.

I know, I know: it's irrelevant... (why is it shown then? Just remove it from hover description)

As this setup is meant to run with the reactors at full throttle unchecked and at no point can the setup consume more steam or heat than what can be produced, all of the tanks and status indicator lights are unnecessary. This all just uses extra UPS that doesn't need to be.

See description:
You can consume more extra steam than it is produced.

While reviewing all inputs, thoughts & concepts, I don't see a "truly tileable" reactor setup that matches exact heat power production/consumption. While the exact amount of heat exchangers can match with a reactor's power output (this is what I've done), the turbines cannot, unless we're using that 8,000 setup.

Another opinion for a next update:
Change the description for Steam Turbines in-game regarding available power from "5.8 MW" to "5.82 MW". It is said that Einstein would have calculated with (only) two digits after the point, too.

Thanks for your stimuli, they helped me a lot.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am I didn't occupy with fuel control, since there's actually no true circuit network signal possible to read the electrical energy load.

This can be done indirectly with steam buffers. Add storage tanks that will be filled along with the common steam level within the plant. If there is less energy consumption than production, the steam level in the plant rises, and so does the steam level in the tanks. You read the content of a steam tank, and if it's above a threshold, the circuit that does refueling will stop refueling. The refueling will always only insert one fuel cell, so if the refueling stops, the current fuel cell is burnt up but after that the reactors have no fuel and their temperature will lower.
Due to lowering the temperature, less steam will be produced, so the steam tanks will empty and continue to supply the steam turbines. If the content of the steam tanks got below some threshold, the circuit will continue refueling, so the reactors will heat up again.
In such a setup, the steam tanks have to be big enough to buffer excess steam, so the reactors will not heat up to 1000°C. And the threshold must be set not too low, so there is always enough steam buffered to bridge the reheating of the reactors.

An example for a generic circuit solution can be found in the wiki: https://wiki.factorio.com/Tutorial:Circ ... lear_power
There exist other approaches, for example not triggering refuel with the output of the used up fuel cell but use 200s (12000 tick) counters to get an exact timing.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

You mentioned that there wasn't space to place substations: while this appears to be true with the heat exchangers (which wouldn't be necessary if you weren't so insistent on placing pumps everywhere), there appears to be plenty of space around the turbines? Not a big deal, but felt it worth mentioning.

The pumps can stay, see above, expanding just the lines with underground pipes. A good idea, yes.

Pumps were not the primary point of this comment....

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

You have a speaker set to alarm on lack of fuel on the input belt to the whole setup, but you have a lot on the feed belts themselves plus a huge amount of buffer in chests. Seems like the wrong place to place the alarm condition?

Where would it be better?

Well, your alarm is set to say "[the reactors] are about to shut down!", but they're not really, as you still have (potentially) up to 4800 pieces of fuel in the chests alone, not counting what is circulating on the belts (which, with some very rough calculations, is between 700 and 950 additional pieces of fuel) or in the reactors themselves (considering you run everything unchecked, they'll load the reactors up to anywhere from 5 (with no stack size research) to 16 (at full stack size research) + running fuel). This means you have ~7.4 hours of fuel in the chests, ~1-1.5 hrs of fuel on the belts, and ~16-56 mins with what's left in the reactors (depending on stack size research and the current running fuel % remaining).

It's not a major issue/point, I can just see someone using your design, getting the alarm that the reactors are about to shut down, but then anywhere from ~8.5-9.5 hours later, they finally actually shutdown if they don't ever provide more fuel. How you want to deal with it is up to you, if at all. I just wanted to point it out.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

Pumps between the turbines are unnecessary and a waste, not just in resources and power (and space), but in UPS as well. Setup could be 104 tiles narrower, and sans ~912 pumps (that's 912 fluid entities using UPS that don't need to be there!)

Without pumps, the steam level is falling.

Why would it? You have (on the long rows) 16 heat exchangers producing over 1648 steam (using the rounded down figure of 103) and 27 turbines consuming 1620 steam. You're producing more than consuming, so why would the levels drop? And really, even if you weren't, the presence of pumps wouldn't solve it....

Have you tried it, though? I did. I took out the pumps between the turbines and moved all the turbines so they were direct connected end to end. Mind you, those were the only pumps I removed - all pumps in the heat exchangers, the ones feeding to the start of the lines of turbines, and the ones taking out from the end of the lines of turbines I left in place. From what I saw, not only did the steam levels not fall (they continued to rise, in fact), but they all powered up just as quickly as the pump-between-each-turbine design did.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

As this setup is meant to run with the reactors at full throttle unchecked and at no point can the setup consume more steam or heat than what can be produced, all of the tanks and status indicator lights are unnecessary. This all just uses extra UPS that doesn't need to be.

See description:
You can consume more extra steam than it is produced.

This is not the primary point of the design though, is it? And considering that you don't care about wasted fuel or heat with this design, then getting every extra drop of steam out of the turbines shouldn't be important, either, correct? As such, you don't need a pump at the end of the lines of turbines. A simple pipe would do it, then you can put a pump to it after several pipe segments. This should prevent the pump from actually robbing from the turbines and not cause an issue with flow rate of the extra steam.

And to be fair, there's a very high probability that anyone using your design is not going to use 500°C steam for those extra things that work just as well from 165°C steam. I'd almost just leave that entire piping segment you have at the end of the turbines out and leave it up to the end user what they want to do. After all, by including it, you're also dictating to them in which direction all of that extra steam goes.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

mrvn wrote: ↑Sat Apr 01, 2023 9:21 pmJust don't do any false advertising.

Pardon?

It was never my intention to name something wrong. I have changed the topic & added some change notes.

Since you claim tileable please do fix your top and bottom row.

It's in the description that those two rows are intentional, matching their reactor power.

The few extra heat pipes are not adding anything significant to the price of the reactor.

Reactor's "price"? My goal was a heat pipe line as short as possible.

Same with pumps, being so cheap you can use the pumps for a full row, don't replace them with underground belts.

I don't get this message: underground belts replacing or not replacing pumps?

I didn't say to add more heat exchangers. The top and bottom row should keep the same number of heat exchangers as they currently have. But
The position of the offshore pump in the long and short rows differs. I said to move the offshore pump in the short row to where it is in the long row. That way, when you expand the reactor there will still be water where the new pump needs to go. In fact the old pump will already be there so even if you landfilled under the pump things still work.

And when the pump is moved you have to connect the heat exchangers to it somehow. You could use an underground pipe, but then you would have to remove that on expansion. Instead I suggested to move the heat exchangers (the half of the short row that gets fed by the outer offshore pump). Again that places the heat exchangers where there would be heat exhcnagers in the expanded reactor so nothing needs to be removed on expansion.

But now the heat exchangers are split in two and the steam isn't connected anymore. That's where the pumps come in. Connect the two groups of heat exchangers not with an underground pipe but with the pump-pipe-pump-pipe-pump-... setup you have in the long rows (most of the pumps are unnecessary but who cares).

The goal is to have no entity in the original reactor that isn't present in the expanded reactor. So you can change a short row to a long row simply by pasting a long row on top of it. In this design it would just add the missing heat exchangers in the middle of the short row. Everything else of that row already exists. Same with the turbines. The long row should just have more turbines on the outside, nothing to remove before expanding the reactor.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am

Next the roboports. Replace the steam tanks with an underground pipe. If you don't use any fuel control then they are utterly pointless.

Like shown in the description, the tanks are used for steam control.

Replacing the tanks with an underground pipe give you space to place a roboport in there. You can also move the whole row of heat exchangers away from the reactor making space to a) leave water under the offshore pumps and b) place a roboport. It will raise the reactor temperatur slightly for controlled fueling but you are overheating the reactor anyway.

I see what you mean. Longer Heat Pipes: no, but I could in fact move the tanks/turbines, along with underground pipes, to place roboports in between without losses. Unless I expand the heat pipes there cannot be a roboport coverage around the reactors.

You won't have to interrupt power production, the reactor will just grow and produce more power.

Power production isn't just interrupted but satisfaction lowers, at least with my setup where a part stays and will be expanded. If someone monitors its power consumption, there should be enough time to expand.

When you have to remove the top and bottom rows that's a drop of 4 * 20 MW == 80 MW of power. And that probably comes at a time when you have brownouts already. Because who notices power running out before it actually does?

You can make your reactor setup totally tileable by the simple solution of moving a few heat exchangers around and adding a few pumps. I think that's totally worth it.

As for longer heat pipes: There is enough wiggle room in the heat flow for longer heat pipes. You are not loosing any energy by adding a few more, the heat still spreads all the way to the outside and all heat exchangers will run as expected.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am I didn't occupy with fuel control, since there's actually no true circuit network signal possible to read the electrical energy load.

Fuel control doesn't try to read the electrical energy load, it's trying to read the reactor temperature. By monitoring steam levels you detect when the reactors temperature falls so low that some heat exchangers drop below 500°C. And the steam tanks you use for monitoring also create enough buffer to bring the reactor temperatur back to fully working levels.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am While reviewing all inputs, thoughts & concepts, I don't see a "truly tileable" reactor setup that matches exact heat power production/consumption. While the exact amount of heat exchangers can match with a reactor's power output (this is what I've done), the turbines cannot, unless we're using that 8,000 setup.

There are a number of posts in the forum for truly tileable reactors. And I just told you how to make yours one too, twice now.

Lets try a third way:

Start by making the short rows as long rows. I know, I know. Now you have too many heat exchangers and steam turbines. So remove some of them without breaking the reactor or having to add any new pipes. You can remove any entity from the short row without breaking the tileability. Just don't add any.

FuryoftheStars wrote: ↑Sun Apr 02, 2023 2:43 pmWell, your alarm is set to say "[the reactors] are about to shut down!", but they're not really, as you still have (potentially) up to 4800 pieces of fuel in the chests alone, not counting what is circulating on the belts (which, with some very rough calculations, is between 700 and 950 additional pieces of fuel) or in the reactors themselves (considering you run everything unchecked, they'll load the reactors up to anywhere from 5 (with no stack size research) to 16 (at full stack size research) + running fuel). This means you have ~7.4 hours of fuel in the chests, ~1-1.5 hrs of fuel on the belts, and ~16-56 mins with what's left in the reactors (depending on stack size research and the current running fuel % remaining).

It's not a major issue/point, I can just see someone using your design, getting the alarm that the reactors are about to shut down, but then anywhere from ~8.5-9.5 hours later, they finally actually shutdown if they don't ever provide more fuel. How you want to deal with it is up to you, if at all. I just wanted to point it out.

It's some kind of "failsafe circuit". Primarily, of course, there's an alarm set up for uranium ore. If that runs out, the fuel cells will run out, too. Having limited resources and perhaps need to explore more ore that's far away, this should give enough time without running out of power. Here's my current uranium ore facility:


(hum...I cannot spoiler an attachment, interesting)

You may call it "overcautious", but, well, there's opportunity to manage nearly everything...

Why would it? You have (on the long rows) 16 heat exchangers producing over 1648 steam (using the rounded down figure of 103) and 27 turbines consuming 1620 steam. You're producing more than consuming, so why would the levels drop? And really, even if you weren't, the presence of pumps wouldn't solve it....

Just adding the turbines will let the steam level stay < 200. That's a fact. Another fact, and there, you're right, it will work without, as long as at least there's equal steam production that's consumed.

Have you tried it, though?

I have tried another thing yesterday:
The turbines were split apart from the heat exchangers. Then, 3 tanks were used as a buffer. Each 1/3 of a reactor side, 6 rows, are piped to a tank each. This is the maximum a normal pump can provide. Then, after the tanks, there're another 3 branches with turbines to match nuclear heat production (at least some pumps are necessary for this as a valve). The tanks can be merged.

However, it was not possible to me, no matter which setup, to have a higher pumping speed than ~1800. Even at full load, some heat exchangers didn't manage to shake off their steam. So, and that's another interesting thing, some branches of the turbines got loaded, then completely unloaded, over & over again. My intention is still to convert the maximum heat of a reactor to electric power.

With such a setup I could use an amount of turbines as near as possible to the heat the reactors produce with minimal loss. But it did not get it work, and it isn't also tileable, but expandable at at least two different places.

vangrunz wrote: ↑Sun Apr 02, 2023 10:02 am See description:
You can consume more extra steam than it is produced.

This is not the primary point of the design though, is it?

Indeed, it is. I often asked myself: where to get the steam for coal liquefaction or modded recipes, without setting up extra steam engines which consume additional fuel? Even if I produce solid fuel from coal liquefaction for steam generators, there's less oil from coal available.
Then I found the solution that the remaining heat (that isn't fully consumable by a tileable or expandable reactor setup (therefore, a turbine must produce 5.00 MW instead of 5.82 MW)), which just comes in the matter of steam, can be distributed. This is how this design was born.

As such, you don't need a pump at the end of the lines of turbines. A simple pipe would do it, then you can put a pump to it after several pipe segments. This should prevent the pump from actually robbing from the turbines and not cause an issue with flow rate of the extra steam.

They cannot rob: if they do, the tanks will start to fall under maximum and the pump will immediately stop until the tank is full again. This guarantees that the turbines never run out of steam (see the red copper lines which are individual for each half of the row).

And to be fair, there's a very high probability that anyone using your design is not going to use 500°C steam for those extra things that work just as well from 165°C steam.

I think this can be disregarded because if the system's electric usage isn't near maximum (and there's no or everything below 100% fuel control) the "over"produced heat has, in fact, a benefit, not to say: it's produced, but not used/wasted. Since nuclear fuel cells are cheap (someone mentioned that here in a forum), I don't see why we should not use this to our advantage.

Another thing:
Fuel control must be 100% accurate not to produce additional steam. I do not know of any setup currently that can prodive this, an adaptive steam level monitoring & adjusting is obligatory. Currently, the inserter can have only one signal set. It would be interesting if anyone had found out how to manage this with eg. accumulators, steam tanks and/or constant/arithmetic/decider combinators? Ideas welcome: it's no problem to add a second tank in each line for fuel throtteling (one dedicated is always necessary to guarantee 100% turbine load while steam is consumed otherwise).

On the other hand, a constant consumption by a nuclear reactor is a vanilla setup, that can be altered, though -- in contrast to boilers in combination with steam generators which will consume more fuel the higher the load is. I think this is due how nuclear power is managed in real life; a fuel rod isn't aware of any load, but steam engines do, like every other combustion engine.

I'd almost just leave that entire piping segment you have at the end of the turbines out and leave it up to the end user what they want to do. After all, by including it, you're also dictating to them in which direction all of that extra steam goes.

Take it easy: I'm not "dictating" anyone anything. All I do is to make an offer. This can be accepted or dismissed.