Nuclear Power from a Nuclear Engineer

[BeCurieUs]

I saw that nuclear power is being considered for the next round of updates. As a huge fan of Factorio (new comer, but very obsessed) and a future nuclear engineer, I wanted to share my thoughts on some thoughts on how nuclear should be implemented in the game.

1) Mining

I have 2 thoughts here, one more in line with current mechanics and another a bit more unique.

a) Mining should be chemical rather than physical

So modern day uranium mining will, many times, will be a chemical process rather than a physical one. We will pump some type of acid down to dissolve away only the uranium and pump it back up. So to mine the uranium, maybe we use some kind of modified chemical plant with a sulfuric acid input. This gives you a uranium powder which we will use in a later process (enrichment).

b) Ore should be discovered with radiometer rather than surface ore body

This seems like a more tricky one to implement. But ore bodies for uranium are usually sub-surface. We find them,sometimes, using radiography. Literally some dude walking around with a gamma scout looking for hot zones. We then have to do a core sample to get an idea of the ore yield. This could be similar to how oil extracts in the game, sometimes you find a rich vein, sometimes not. Instead of it being duration of how long we can mine a resource, however, I think it should be how much chemical processing it takes to get a similar yield of ore resource. So a low yield = more wasted sulfuric acid. Ore yeilds in the real world can range from .25% to 12% uranium by weight, so there is HUGE variety in the real world already here, might be reasonable to have similar scale in game.

So you could basically have them like little oil spots in the game that you can't see without some kinda "night vision" (really gamma vision) and you would need to get pretty close to it. Since we magically know how good an oil site is gonna be, maybe we just let the same thing happen with uranium deposit.

2) Enrichment

Now that we have our powder, we need to make it into something usable. Now, there are reactors that can use regular uranium, but doing it the way I am going to suggest has game play benefits.

a) New unit, centrifuge

centrifuge is the modern system used for enrichment. It should likely operate in a similar way as the boilers do, except powered by electricity, where you smack the output of one onto the input of another. Having more = faster you enrich, but also the more energy you use. Later, we might talk about different levels of enrichment, but for now, it should likely just have 2 outputs, high enriched and depleted uranium

b) energy use

Enrichment is a pretty significant task in terms of energy use. In the real world, it is about 1000 to 1. So a 1000MWe reactor will need about 1MW of power enriching things. In game, we could scale this any direction we want to, make it worse or better for game play purposes

c) depleted uranium

The waste stream can be used to make bullets. Just do a research on the gun to be able to use it, and whamo, you got yourself some depleted uranium shells! Likely just produced out of a factory using steal and depleted uranium. Maybe they are so good that they need a drawback, like they also increase pollution on use?

2b) After you get your enriched fuel, we could likely just stick it in a reactor, but we could also do like we do in the real world and have to fabricate it. This could be done in just a regular fabricator with uranium. In the real world there is a wet and a dry process. The wet process uses a base and water, but we don't have bases in the game, so a duel input lab with water and enriched uranium or just uranium. Or just skip this part and just put the uranium from centrifuge directly in a reactor.

3) Power plant

In the real world, the lions share of all energy on the front side of a reactor is in Build, operate & decommission plant. Over 60% of the energy in the nuclear life cycle is from this part. So the nuclear power plant itself should be the more resource hoggy of all this. Lots of steal, AND finally a use for concrete! It will need lots of both, copper to, wires, not many chips as more reactors are analog controls (computers and digital circuits fail in unpredictable ways, we don't like that), lots of pipes...so many pipes.

Should also just be a heat source for a steam plant, so it can heat water up almost exactly like a boiler and power many different steam generators. I will leave the math up to yall, but the math in the real world is for all the energy inputs, nuclear will output about 100x what you put in. a little less if the ore quality is poor. Poor quality ore can raise input as a share of output from about 1.7 to 2.3% to upwards of 4%...basically low quality ore doubles the energy inputs, but still, payback is pretty huge.


Now, if you want to make an entire nuclear side to the game, you could start to add in other reactor technology, like reactors that don't need enrichment (just fabrication), or reactors that operate at higher heats (power more steam generators), perhaps more R&D into better power cycles than steam (but only work at higher temperatures), and other such thermodynamic goodies.

4) Waste?

If you want to deal with waste accumulation, we could just do the "dry cask" method of it literally going into a barrel. So power plant eventually depletes uranium, take that uranium throw it in a barrel and take that barrel somewhere to store it. Or you can just have it disappears like other waste in the game disappears and it just turns into pollution. As a nuclear guy, the idea that nuclear is generating pollution upsets me greatly, so I like the barrels idea better, but might be to hard a mechanic to manage.

4b) Refueling

So in the nuclear world, we have to offline the reactor for refueling and to remove spent fuel rods, and refueling can take awhile. This could be an interesting mechanic to balance against. In the nuclear world, we can plan downtime to refuel when we can ramp up other sources to compensate. Similarly, if you have 2 nuclear plants that are only extracting 50% of their heat energy, you could use this buffer as a way to refuel. For gameplay reasons, I would say refueling should take longer than one could reasonably expect to store energy in batteries...like 2-3 days to refuel, but once refueled, online for a long time. SO refueling ends up being a slight mini-game you have to keep up with. Dunno, this might be overly annoying and make nuclear not worth the hassle, unless nuclear ends up being so damn good energy wise that all the hassle above is worth it. Or just tone down all I have suggested above and make it like "coal 2.0".

Anyway, just some food for thought. Dunno, maybe someone already has a thread like this, but I saw the announcement and wanted to throw in my 2 cents.

Cheers @BeCurieUs

[Rhamphoryncus]

I quite like how you've approached this. In particular inputting acid into something like a pump-jack sounds like an interesting mechanic.

I'm unsure about having to locate deposits via a radiometer. If you have to walk everywhere to find them (rather than look at your radar) you cover much less ground, so to balance that the deposits must occur at a much higher frequency. That then means your map has disproportionately more nuclear ore deposits than other types of ores, which on a long running map seems a balance problem. However, that all could be resolved by adding in aircraft to increase the ground you cover, or even autonomous drones so you don't have to.

Waste gives me an idea. A long running complaint I have about factorio is the large jump from green to blue science. Nuclear, being a new ore, provides a natural gate for another science pack, which would allow existing recipes to be reshuffled (and blue somehow easier.) Waste from a basic reactor, being full of tasty transuranics, is a great candidate for scientific experiments (meaning the science pack). However, that has limited consumption (especially once all science is done), so a second, more advanced reactor could be unlocked with the option to burn either regular fuel (cleanly) or transuranic waste.

[BeCurieUs]

I quite like how you've approached this. In particular inputting acid into something like a pump-jack sounds like an interesting mechanic.

I'm unsure about having to locate deposits via a radiometer. If you have to walk everywhere to find them (rather than look at your radar) you cover much less ground, so to balance that the deposits must occur at a much higher frequency. That then means your map has disproportionately more nuclear ore deposits than other types of ores, which on a long running map seems a balance problem. However, that all could be resolved by adding in aircraft to increase the ground you cover, or even autonomous drones so you don't have to.

Waste gives me an idea. A long running complaint I have about factorio is the large jump from green to blue science. Nuclear, being a new ore, provides a natural gate for another science pack, which would allow existing recipes to be reshuffled (and blue somehow easier.) Waste from a basic reactor, being full of tasty transuranics, is a great candidate for scientific experiments (meaning the science pack). However, that has limited consumption (especially once all science is done), so a second, more advanced reactor could be unlocked with the option to burn either regular fuel (cleanly) or transuranic waste.

Ya, the gamma scouting stuff seems pretty hard to work out. My brain was thinking about it kind of how Civ does it, where it was there the whole time, but you have to unlock it. Maybe just having gamma scouting upgrade for radars would be enough, though, like a tech upgrade?

And to the waste/research thing, I had the same idea but it was using a "test reactor" that uses highly enriched uranium instead of regular low enriched uranium. Basically, if you fed regular enriched uranium back into the centrifuges, it would become HEU (and even more depleted uranium) which could then be used in a test reactor to transmute things, or just make some blue science or something. Similarly, on the backend you could reprocess the waste to net some kind of material and depleted uranium. Just trying to avoid creating to many "nuclear unique" buildings...already have a special mining one and a centrifuge AND the nuclear power plant.


Also, I should have mentioned the NPP should have to have a water source and it should use more water per unit energy (maybe requiring multiple pumps). Basically, the nuclear cycle is basically the opposite of the solar cycle. Solar cycle you can make small investments overtime but end up being rather significant in terms of land area and time spent making it vs nuclear has a ton of up front time before power starts getting generated, and has some management (albeit minor) along each phase, but huge payoff in terms of power. Be a good balance without being OP, slightly annoying but easy to expand and maintain vs solar easy to start but maintain (continual additions get cumbersome).

This all might just be to much, in the end just having an ore you mine and throw in a reactor like coal would work as well, basically coal 2.0, but I like making it more "jump through hoops" and requiring the advanced infrastructure of mid-game.

[BeCurieUs]

Looks like this mod has a lot of these ideas, I like it!

viewtopic.php?f=93&t=21676

It uses sulfuric acid AFTER it is mined, I kinda like it being a combined mining/acid thing, but this is a good solution from a mod, easy!

Some other things after seeing it splayed out make it seem a bit complicated, so I will wonder what the game developers think. Good work on the mode @Simdezimon

[Solyx]

The latest FFF asked for suggestions on nuclear power, so I figured I'd throw out a mechanic I haven't seen suggested yet, and open it to criticism and refinement.

Largely I see a lot of complicated, sophisticated suggestions for making Nuclear Power, and I don't think that's a way to go for something that's going to be in the Vanilla game. The Mechanic should be straightforward, with just enough complexity to make the process interesting and rewarding, while making it enough of an effort that you feel you've earned the reward.


Game-play is the only really important thing - reality meshing is nice but incidental. With that in mind, the inspiration for this idea comes from the ideas of Thorium Nuclear power. The general Thorium Fuel cycle is triggered by the fission of a Uranium-233 atom, which fires off several neutrons. Thorium absorbs neutron and turns to Plutonium, and then decays down eventually into another Uranium-233. Thus Thorium is the fuel, and U-233 acts as a pseudo-catalyst that drives the fission process but on the net isn't consumed by it. Thorium IRL has some interesting properties and could be a very good idea, but it also has some practical drawbacks. My idea for Factorio emphasizes one of them: The scarcity of U233. U233 exists in very small amounts, and it takes a very long time to breed. We're talking potentially years to double the amount of U233 you start with. When it's in the reactor, it'll perpetuate indefinitely, or even breed a little, but you're limited by the number of reactors you can have going by your total supply of U233. If the United States were handed the perfect designs for a thorium reactor, it'd still take 50 years to get enough U233 bred to run the US power grid entirely.

So that's the mechanic for Nuclear in Factorio. Thorium for Fuel, which will be straightforward to get, and Uranium as a requisite catalyst for running the reactors you must breed in sufficient quantities if you want to run your whole power grid off the stuff.

Here is an example implementation. I'll have comments below about things that could be changed. And numbers of course can freely be tinkered with to improve balance/progression - this is just to get the general sense of the implementation.

::::: EXAMPLE :::::

Phase 1
Add an Ore Processing Plant that consumes 10MW and takes in, say 50 coal and outputs 1 Thorium after 180 seconds. For an approximate amount, one thorium ore could have 5GJ energy content. Including the 1.8GJ spent refining it, that yields a net energy of 3.2GJ compared with the 400MJ contained within the consumed stack of coal. This represents an 8x increase in total energy from refining the coal, making nuclear an attractive option, especially if resources are set to a more scarce level.

Additionally, the Ore Processing Plant will also have a production-module-esque progress bar. Every time you refine a stack of coal into 1 Thorium Ore (Th-ore) the bar is advanced by 1%. Being Fed continuously, with no power interruptions, it would take 5 hours to generate that first U233 ore. And while you could make multiple first-ores in parallel, you can't use 9 women to make 1 baby in 1 month. This minimum time requirement could not be side-stepped (though speed modules could perhaps be employed).

This is phase 1 of the mechanic. The plant itself requires a large amount of power be dedicated to it, which forces the player to be somewhat well-established before embarking on the path to a nuclear power grid. Furthermore the exceedingly long amount of time it takes to get a U233 ore encourages them to both work towards this as a goal to start early, and also leaves nuclear unattainable in the mid-game. These factors ensure the progression will go steam->solar->nuclear without skipping straight to nuclear. Also I think it'll feel kind of cool to have to set up your own little Manhattan Project. I notice a lot of other suggestions to try and do this, but they use a ton of complicated refineries and multiple buildings and fluid pipes and etc etc. Which is IMO, putting in too much realism and complexity when you can get a similar feel with a more streamlined mechanic. One building - refine both fuel you'll eventually use, and slowly cobble together enough Uranium to start a reactor. Streamlined, representative, simple, and still with a feeling of significance because of the massive time and resource investment.

Phase 2
Add a Nuclear Reactor. The Nuclear Reactor consumes 1 Thorium and 1 U233 over 100 seconds, producing 50MW. It also outputs a U233 (replacement). Additionally, a progress bar to generate an additional U233 advanced by 10% per cycle. So that single plant will generate an additional U233 every 1000 seconds - or every ~16 minutes. After struggling to make your first or second U233 ore, now you can slowly build up your reactor count to breed it in earnest. After 16 minutes you'll have an extra U233 ore you can use to start up a second reactor. After 32 minutes you could have 4 reactors. For a 500MW power grid, you could build up your reactors in a bit over an hour.

At this point, you'd need multiple Th-ore processors per reactor, but you'll also have a huge stockpile of Th-ore to burn through as well. Rewards stockpiling, and also potentially makes min/maxers think beyond basic i/o balance.

And that's essentially it. Your power grid is set up. Took some deliberate investments, and lot of time. But now you really have a post-energy-scarcity base set up. And founding new bases with separate power grids just requires a reactor, a U233, and a box of thorium, and it'll run forever.

Phase 2+
Something else to consider would be having a nuclear base with a U-233 infrastructure. Nuclear powered trains as an example. Each would require a U233 to be able to run. They wouldn't breed extra U233, but if you fuel them with a stack of 50 Th-ore, it's 625 as energy-dense as coal. Even limited to a single stack, you would never have to worry about refueling your trains, which is nice from an infrastructure standpoint and provides another incentive to go beyond solar to nuclear.


:::: END EXAMPLE:::



The reason I like this mechanic and some similar sort of implementation is:

-no new resources added to map
-no significant coding changes required (works within current mechanics of game)
-requires deliberate investment to get up and going
-has a slow, but exponential payoff.
-rewards forward thinking and stockpiling, also changes things up for the min/maxers given the huge upfront surplus.

Overall it seems to me like a mechanic you wouldn't put in a mod, but rather something streamlined and effective enough to be worth putting in vanilla.



So that's the general mechanic. Feel free to play with the numbers, to elaborate on the implementation, to make other recommendations etc. Other people have suggested other things for nuclear, and that could be mixed in. The main mechanic I'm interested in talking about for possible use is just the breeder-mechanic. A process that takes a lot of time and resources to start, and slowly becomes faster as you develop your nuclear power more and more. Focus on that aspect for discussion.




Some practical thoughts/changes to the implementation:

-Could use stone or Coal. Stone made more sense from a science perspective, but stone isn't as abundant. Energy-density of thorium makes this not a huge problem. Main issue is coal would become largely useless after this is implemented, so effectively transmuting coal to thorium seemed like a more balanced approach.

-Could also require something like sulfuric acid in the refinery. Honestly you don't use acid to dissolve out Thorium, but it'd represent some sort of refining process going on, and require additional resources. It would make sense from the perspective of refining the U233. It'd also force players to have a good oil-infrastructure in place before they can move to nuclear, which would further help enforce a progression.

-You could make the Uranium 233 more abundant and continuous, so a plant needs something like 30 U233 to start running, and reaches max power output at 50. The general time commitment to breed would be similar(make it 30x faster), but people might find it more interactive to manage ten or a hundred times as much Uranium, making it more of a continuous process than a few short moments interacting with it. Only problem is the minimum-time to running nuclear reactors is set by the time to make a single ore in a single plant, since you can massively parallelism and make the first 30 in the time to make 1. Removes a big progression-enforcement mechanism, but if you make the energy/resource input excessive enough, that could still act to limit how fast they can push through phase 1. Frankly I like the idea of a few super-valuable items you have to arduously craft over time, but that's just me.

-Most any base could get by with a dozen 50MW reactors. Which makes this seem like a lot of buildup with no payoff. Though frankly that can describe Factorio itself atm, so at least it's in character. The draw for nuclear is that you get very compact, simple, limitless power once it's set-up. Nuclear being over-powered could make it boring, but if it's not, then why will anyone bother using it over Solar Panels anyway? Maybe add an extra mechanic that manages the power output of the reactors? Or add extra end-game structures that will consume ungodly amounts of power, making nuclear necessary? I feel like this is a larger problem with nuclear, not specific to my implementation. Still worth discussing. The Phase 2+ part tries to address this - make powergrid energy the initial goal, but also just a stepping stone to a further goal, breeding more U233 to get super-energy-dense fuel to run trains and other such things.


Let me know what you think, if you think this would be an interesting mechanic, if you'd have fun setting up your own ore refinery, and how you'd make it better while still keeping it focused enough for a vanilla implementation.

[Koub]

If we focus just on power genration, I like your idea a log. However, I think it's too focussed on the sole power gen.
The things that I feel *should* come with nuclear :
- Nuclear waste management
- Need of water source (cooling, nuclear turns water into steam, steam goes into steam engine to produce electricity), and subsequent meltdown risks in the lack of (I know you know, but it's for the audience ^^).
- Other uses for nuclear, namedly for weapons (no nuclear bomb, that'd be too powerful, but depleted uranium bullets and shells would ne awesome against behemoths and heavily armored targets)

That would add a continuous drain on the uranium produced, so that your exponential production finds an output.

NB : I'm nowhere a nuclear specialist, some things I just wrote are deformed from IRL, I know different isotopes have different uses, but, hey, it's a game .

[TheDagmaar]

[moonzteel]

I like this thread. Good stuff.
I'm not a fan of "minigame" for refueling or similar antics in a game about building automation. (though if it where a simulation ofc)
Other then that I think this thread got something.

Your thread makes me think along these lines for a somewhat working first phase of nuclear power.
Mine uranium rich material,
Add acid and energy in a combined refinery/enrichment plant (i have no idea what that would be called). There is no point in splitting them unless you intend to add a use for non enriched uranium. (pollution seems fitting at this stage)
Insert enriched uranium to a reactor core. Run water to it, let it consume water and output energy in liquid form, where amount of water consumed is directly proportional to the amount energy the reactor core can get rid off any energy still left starts to build up in the core. (I'm not fan of pollution at this stage would prefer waste that might be run in a next gen factorio nuclear reactor that being said some would probably be fitting if nothing else due to waste heat.)
The liquid energy is piped to combined turbines/cooling towers. Could have them be the end or out put water that you can run back to the reactor.
This would mean lack of water input or no where to get ride of the heat build up in-case of to few turbine/cooling towers could cause the reactor to overheat.

But please for the love of good don't have them explode. If anything have reactor core overheat initiate emergency shutdown and each emergency shutdown has a small chance to fail and cause a meltdown. If meltdown occurs make the make the reactor core building unusable.

[Koub]

[Koub] In fact these topics are really about very very similar subjects. Topics merged.

[bobucles]

Lots of stuff to read through! I wanted to talk about adding some inflexibility to nuclear power. Everything else provides X energy exactly as needed. What if nuclear had a fixed output? Failing to use the energy could cause problems, and sudden needs can cause shortages.

One possibility is that a nuke reactor takes several days to "warm up" to full output. As the fuel ages the output steadily decreases until it's time to shut down and refuel.

[afk2minute]

you can scout from satillite you launch with rocket.
Not very realistic but it can work in this 'model'

[vipm23]

The things that I feel *should* come with nuclear :
- Nuclear waste management

This. I think thorium fission still produces radioactive material-if nothing else the containers inside will need replacement

- Other uses for nuclear, namedly for weapons (no nuclear bomb, that'd be too powerful, but depleted uranium bullets and shells would ne awesome against behemoths and heavily armored targets)

I think, actually, that if we use the thorium cycle as elaborated earlier, we could have a really powerful nuclear bomb WITHOUT it being OP, because it would cost lots of scarce U-233 to make.

If we use a rocket from the silo to deliver it, the rocket might actually be the cheaper part.
Speaking of: for an after release expansion:
http://www.projectrho.com/public_html/r ... ibertyship

Or just an Orion.

What if nuclear had a fixed output? Failing to use the energy could cause problems, and sudden needs can cause shortages.

So something like, it heats up water in it constantly, and if the water gets above 90 degrees it starts overheating?

I like. Though i think you'd need a way to get rid of extra hot water. Evaporation cooling tower?

One possibility is that a nuke reactor takes several days to "warm up" to full output. As the fuel ages the output steadily decreases until it's time to shut down and refuel.

I like this too.

[bobucles]

So something like, it heats up water in it constantly, and if the water gets above 90 degrees it starts overheating?

Something like that. Say a generator is making 50MW but you can only use 40MW. The excess builds up as "stress" in the nuclear system. The excess stress can be reduced slowly over time or maybe with a "cooling tower" sort of deal.

If too much stress builds up in the system it will malfunction. The kindest solution is to simply emergency shutdown. The player is then stuck with the slow process of ramping it up again.

I think in any case the system should punish the player for mismanaging nuclear energy. The simplest punishment is to have waste energy still burn the normal amount of fuel. This would be in contrast to "smart" coal that only burns exactly what is needed. A more advanced punishment might be that an emergency shutdown "dumps the core" and you lose some portion of the remaining fuel. It could be half, it could be all of it, or you might have to reprocess the fuel to use it again. In any case a mismanaged nuke system would punish the player with some kind of resource penalty. Thus the player is encouraged to build all sorts of wacky things to keep their high tier energy stable and happy.

The "keep it stable" mechanic becomes FAR more difficult when you include the next system

Nuclear as a slow ramping, decaying power source

Unlike the instant satisfaction of steam power or accumulators, nuke power would slowly ramp up to full capacity over a day or so. The basic idea is that you don't WANT your nuke power to shut down, because then you have to wait to build it up again. It's made to run steady, day in and day out, in order to give the best results. Kinda like IRL.

Once a nuke plant is at full tilt, its power output might decay over time. Say that a plant gives 50MW, but after a few hours it gives 45MW, then 40, and after a few game weeks it's all the way down to 20MW. Now you can't just "set and forget" your energy supply. At some point you have to shut down, replace the core, and reset it to gain full power again.

A hundred generators might give 5GW of power or they might give 2GW of power. Over consuming means you don't have enough. Under consuming can cause the power to emergency stop. Without enough spare capacity you won't be able to change the power cores without shutting your whole base down. How is a player supposed to keep everything straight? It's probably a job for automation.

[vipm23]

I dunno if generic stress would be a good idea. It's simple, but not intuitive-"Why did my reactor just scram? 'Stress'? what the heck is this 'stress', and what causes it?"

Whereas the water getting too hot and the reactor overheating? "Why did my reactor just scram? 'Overheated?' Huh, the water temperature's at 95 degrees-guess that was too hot. But I've got enough steam engines to take all the hot water up-wait, steam engines only take enough to serve the current needs of the grid. Was I just not using enough electricity? Or maybe I need a way to bleed off excess hot water..."

It ties into the existing water mechanics, and suggests some additions to it. For example, a cooling tower could, well, cool water within a fluidbox. A circuit connected pump, with access to fluid temperature, could be used to pump excess hot water out when a circuit condition indicates it's too hot. etc etc.

[Rhamphoryncus]

We need to be wary of systems that frontload all their complexity on new players. Any sort of sensitive overload mechanic would be inscrutable to a new player while an experienced one would say "yeah, wire a circuit to an inserter and a pump and it self-regulates" and invalidate the entire mechanic.

I do like the concept of slow ramp-up/ramp-down or of a slow refueling cycle. Those are more of just a nuisance to new players, something they can ignore at first, then later devise a solution to.

[BeCurieUs]

Perhaps, instead of making the load invariant, that load dramatically changes the water conditions of your system. And if you don't keep it supplied with enough water, you get a reactor scram and have to go through some type of cool down period. This is actually not super dissimilar to reactor power in the real world, the dreaded iodine pit.

Or to take it back to the beginning, reactor startup is generally a phased operation, you can take upwards of a week throttling up to full power. So perhaps just some kind of indicator like "reactor heating up" and it continues to do this over a period of time, then it reaches full power...eventually. But if the water flux to much, bellow an acceptible threashold, SCRAM...have to cool down for 'X' hours and being long repower up phase. This is kind of a harder punishment of a big coal setup. We have all been in the situation before were a coal mine gave out unexpectedly and your entire electrical grid shuts down. This would be a more finicky, long term, "NEVER LET THIS HAPPEN" sort of version of that. Basically, a slight modification of bobucles's commentary

Also, perhaps the only way to feed it with enough water is to use an electrical pump, and thus, having a complete blackup is basically unacceptable, you would have to make some backup coal generators just to power the pumps...which sounds almost to on the nose!

I don't think meltdowns make since from a gameplay and physics standpoint (it is hella hard to meltdown small reactors, even if you really try...which is why we let college kids operate them on campuses), and there all the sudden just being an area of your map you can't use any more is SO devastating it isn't worth the risk, just use solar, always.

There has to be a way, though, to let the player know that water levels for reactors are very sensitive, and hooking into a steam network must be properly maintained. A reactor just turning off for seemingly no reason isn't very intuitive. Steam network is hard enough.

I dunno, I shall ponder on some of the thoughts here more, but I like the suggestions so far. I think breeders and stuff are either to complex or to easy (resource wise). The entire idea is about managing scarcity, I think this way you create scarcity in interesting ways, more stress on your chemical plant, stress on your water management (which is basically not even really a thing now), stress on your power backups for critical systems (also not really an issue, spare lasers). It hits on a number of newish areas that currently are mostly barren. I would love to see more water management beyond just petrochemicals. I don't know what that would be, but something! And something critical, water is so important in real life, feels almost like a sin that is feels almost tacked on in many respects. I dunno, maybe I am just falling in love with my own idea to much, ya that sounds very likely

[Koub]

I'd add something : whatever mechanism is added for nuclear, please make it more or less automatable, like : If I need moar, let me just build moar, and provide with fuel, and voilà. I don't want to be constantly monitoring and doing manual things to my nuclear power source whatever it may be.

[BeCurieUs]

Agreed, the nuclear waste thing seems like the hard part (or maybe it is just late). Maybe the longer you go without putting it in some kind of storage unit the more pollution that generates? And, like solar, you have to physically (or with robots) place the storage casks on a tile. This means it could still be mostly automated similar to solar, except on a smaller scale (waste from the nuclear stuff generated much slower than having to always be running around setting up solar farms). That adds maybe to much complexity, though, but if implemented right, I think a user might understand the flow well enough. I dunno, maybe I should take a moment to mess with the modding tools and see if I can whip up a proof of concept.

[Rhamphoryncus]

Waste disposal isn't a concept factorio currently has. If you get too much of something you can just put it in a chest and shoot the chest. Having to do this is a sign you're doing something wrong or the mod you're using is doing something wrong. Put another way: everything you produce should be useful and balanceable.

Another thing is fluid flow in factorio is very crude. It's not a realistic simulation of pressure, volume, temperature, etc. As such steam engines already do stress it and it's not at all obvious why something doesn't work. It's not good gameplay; there's a reason why 1:14:10 gets repeated as a mantra rather than everybody experimenting and doing their own thing.

[bobucles]

The systems may sound difficult at first glance. But keep in mind there are accumulators, solar, and steam power all working together. If you nuke power is only 50% of the equation then maintaining it is EASY. All the spikes and dips get absorbed by the other energy systems so you get full use with no problems.

The challenge begins when you want 80-100% nuke power. Bases consume a varying amount of juice, which nukes would hate. At that point you need all sorts of things to keep your energy stable and happy.

Anyone can set up a little bit of nuclear power using ramp up and stress/overheat mechanics. They probably won't even be visible its that simple.

Factorio doesn't have established mechanics to deal with toxic waste. It just will not work that well.