All Master Class Blueprints are available on FactorioBin Overview and direct links to all Blueprints: nilaus.atlassian.net/l/cp/HBEUm524 (Pastebin links no longer work)
You kinda let me down here with nuclear blueprints Nilaus, I can understand steam storage versions are bad but some people want their powerplant to shut off when they're not using the power, I was expecting you to dig into the challenge of creating the circuit logic since there are so few tutorials on it.
I took one look at nuclear power and thought "Nope! Sticking with my 264 boilers, 528 steam engines, 14 pumps, aannnddd lots of drills." But after seeing this, nuclear is much less ominous to think about! I appreciate your video a lot :)
Same for me! Really the only thing that is annoying to set up are the drills, because (for me at least) my mining drills are very far away from my dedicated power generation area😂 but yeah its so much easier to set up than I originally thought
Just here to confirm that I've just built a 10GW nuclear power plant, with almost exact ratios, that works at peak capacity indefinitly as long as there is fuel (it really does work, its not those ones you find on the internet that works for a few minutes before running out on water). I did it only cause I wanted a very big one, not because its highly needed. Your points are tottally valid.
But the kovarex isn't for needing less miners, it's for allowing you to have that one patch last LONGER so you don't have to constantly run around and find new ore patches for your fuel.
Kovarex is for disposing of excess 238 automatically by turning it into 235. It's not about making patches last longer, it's about making damn near everything you dig up stop being useless dead weight.
@@williambarnes5023 You're both talking about the same thing. When almost 100% of the patch is U-235 because of kovarex, it lasts longer. You aren't mining lots and lots of ore just to use 0.3% of it, you can use almost all 100%. The U-238 that used to take up space, and that makes up 99.7% of what you dig up, can all be made useful. Everything you dig up stops being useless dead weight... and thus you get the maximum amount of usable U-235 from every patch. Having 100% of the result be useful absolutely means a patch lasts longer, since you're using all of it instead of the vast majority being dead weight.
I agree with your point but I'll set the numbers right: It's 99.3% (or 99.28% according to some sources), and slightly less if you module the centrifuge, so ~135:1, give or take 5. Kovarex needs a substantial working set (i.e. more than the stuff it actually converts) but the net conversion is 3 U238 to one U235. Not sure how a prod mod could change that, but anyway you should use them if you can afford the power and production speed hit. Anyway, first-hand U238 is converted to ~45 times as much U235 than the first-hand U235 you get, and Kovarex ain't random. Finally, second-hand U238 from spent fuel reprocessing can be turned into even more U235, but that doesn't increase the 45:1 figure for reactors - because a batch of fuel rods takes 10 U238 to make; you're merely reclaiming that from spent rods. (Not sure how a prod mod would change _THAT_ tho...) Anyway, I got here because the math looked off to me -- 1 _thousand_ ore per first-hand U235??? But that's right; the centrifuge takes a batch of 10 ore units per unit of U235. It's essentially more than just the centrifuge. It separates uranium from acid, tailings, and other impurities before separating the isotopes. So, one miner at 18 ore per minute makes 1.8 batches, and ~0.013 of those are U235. Which is enough to make 10 (or 11.6 with prod magic) times as many fuel rods, or ~0.15 of them. Each can run the reactor for 200 seconds, so you get ~30 seconds of reactor runtime out of a minute of mining operation; two would keep the reactor more or less supplied 24/7 while ore deposits last. Multiply that by 4 if you want to multiply the output of each reactor by 3, and you get the "8 mines" bottom line which Nilaus mentioned. (His treatment was a bit confusing because he didn't mention the ore shrinkage in the centrifuge, just that 0.7% (of the total output) is U235, and a bogus "18 per second" figure when he meant per minute. I hope my math is on point. I'm interpreting the "8 miners" figure as the requirement per plant of 4, rather than a figure per reactor block.)
Me: Taking so frckn long to design things that I could effort building 10GW+ in Solar, having a SIlo (not working yet) but building my first Nuclear Reactor ever on a little island that is supposed to be my Science-Hub to generate the Energy for it :D
Hey Nilaus, good stuff as usual. Good series this is. May I request a Master Class vid on train unloading techniques? I'm really struggling to balance train unloading using inserters onto belts.
Some of this work has already been done in the twitch streams like stations. Try the twitch stream! There are usually blueprints that go to pastebin you can get before the proper master class comes out. Join!
@@Nortonius_ Perhaps I'll give it a shot. Though I never use other people's complete blueprints copy/paste. I love getting inspiration though, especially for circuit networks
FWIW I agree with you on just stamping down others blueprints. What happened with me was that I was really into the loaders and unloaders (2x8 trains but unloading into 2x full blue belts out) he was using in his twitch series braver new world, and those ideas influenced the stations from the workshop. I then modified to fit my applications like adding LTN support 🚂🚃
I would have gone with 6 offshore pumps rather than 4, better to overkill than underpower especially since nuclear stuff is much more expensive than pumps. I actually made one of my nuclear setups have the reactors and exchangers on a landfill island offsite, and I have a few big trains deliver the steam to an array of tanks and turbines, it's kinda dumb but I like it.
So im a noob here who's only just exploring blue science, but can the steam turbines be used with regular boilers? (the ones that consume solid fuel) Kinda in a energy bind while trying to get to nuclear.
@Mizuna You absolutely can use steam turbines with boilers. They only have a higher max temperature, but they work just fine with 165°C Steam. However, a boiler-fed steam turbine maxes out at 1.8MW, the same as two steam engines but way more expensive. You basically double the cost to halve the space. EDIT: To clarify, the efficiency stays the same. Steam turbines just consume double the steam compared to steam engines.
@@dakon2154 Does the game keep track of the temperature of the steam? I've never noticed that. I wonder what happens if you try to mix steam from boilers and heat exchangers.
@@DFPercush Yes, in technical terms, with fluid.temperature you can read the temperature of the fluid in an entity's fluid box (in this case steam). You can mix fluids of different temperatures and their temperatures averages. In vanilla only steam has more then 1 temperature, but some mods add other liquids that have different temperatures. For more information take a look at the wiki page.
Thanks for the responses everyone. its definitely not going to be worth getting steam turbines quite yet if they are only give a fraction of their max efficiency. Probably going to try to expand to the next coal patch for more steam engines.
Yeah, I have an artillery outpost really really far away from my base, and I train off uranium ammo, artillery shells, and obviously, steam to it and it works really well!
Sacrificing the ratio slightly to make everything more organized is definitely something I hadn't considered, but boy does it result in a nice design. Love it.
hello, here's my 2 cents : - while i'm 100% OK with dumping "perfect ratios" when they're not relevant, (so, most of the time :D) - in the 4 reactor design, i'm very surprised that you choose to use the maximum potential of offshore pumps which are so cheap and trash the excess nuclear reactor potential which is so expensive. I usually go like this : 4 reactors + 6 pumps -> 6*8 heat exchangers > 6*14 steam turbines. Too many pumps & steam turbines, but every bit of the reactors power is tapped (i get 480MW out of it instead of 464 MW) - doing maths makes people leave ? what ? that encouraged me to STAY 8) - i agree with most of the opinions of the second half... although nuclear power without kovarex relies on having enough space to store excess u-238, and that's more than 1 steel chest per hour PER REACTOR, even with prod modules... so should it be advised to beginners ?.. - nice video overall, well explained, well organized.
kirkmcdonald.github.io/calc.html#data=0-17-60&rate=h&min=4&k=off&items=used-up-uranium-fuel-cell:f:1 According to this, fueling 1 reactor without kovarex yields an excess of 221 U-238 per hour, or 21.7 hours to fill a steel chest. That's plenty of time to tech up to kovarex and set it up.
I should have read your comment before I built the reactor exactly as shown in the video--except that the water was a bit further away--and then spent a couple of hours trying to figure out why I couldn't get the full 464 MW of power. I noticed I was seeing 'Low Input Fluid' in several of my heat exchangers. It was easy enough to add a 5th offshore pump to make the design start showing the full 464 ME of power.
This was my favourite Factorio Masterclass yet. I'm not a fan of tutorials that end up feeling prescriptive, showing designs and saying "use this blueprint" rather than talking through the underlying theory and tricks being used, encouraging the viewer to try out their own ideas. I really liked how this time you deep dived into the numbers and reasoning, and provided good insight for experienced players and those following the wider community trends. In my main factory, I put off building a nuclear power plant even with a high effective Kovorax processing system*, because the effort to design my own power plant with all the adjacency bonuses was too daunting compared to the ease of just letting my bots place another solar farm. Hopefully this tutorial will help future players feel less challenged by the switch to nuclear! * When you cover Kovorax processing, please show how no circuits are required. There's very simple circuitless setups, and you really don't need any of the complexity of circuits. The one I use is simple, tileable, beaconed and doesn't need any circuit logic, and ultimately far far too effective.
Although this master class series tackles some concepts that are mostly useful for new players, I can honestly say I learned a ton from this one. Thanks very much mate, keep em coming!
apropos the steam storage; I use steam storage and 20-50% extra turbines, to handle peak variable power load from laser turret arrays. My resting factory load is like 1.2gw but when the wall's getting pushed that can peak at nearly 2gw so that extra capacity just being there ready to go is v v nice, and the turbines/steam storage aren't terribly expensive especially compared to battery banks.
They are incredibly expensive in the Update calculation side of things though. Each pipe and steam engine needs to do their own calculations. Battery banks are the exact opposite in that no Matter how many you have, they will always need the same amount of updates. So if youre building really big or your PC sucks, use batteries and preferably solar over nuclear.
@@VoidplayLP oh yeah, that's a thing. I'm about to set up a Ryzen based system next month so I'm not super worried about it but yeah, my mega base lags a bit.
I know this is an old video, but I still kinda want to discuss some things if possible. First of all while I do agree with many of your arguments regarding steam storage and big reactors, but they are based on the question of whether you need to save fuel or not. So I feel a bit dissatisfied about the argument that you never need to build bigger than this one based solely on the fact that fuel is cheap. 2x1 would make for an even simpler build and probably even easier to tile - it's not more complex than a typical coal burner setup, so why not use it? I think saying that 2x2 is all you need needs a little elaboration, on why not smaller. Personally, I think the question regarding the size of the reactor is about what you feel comfortable building yourself. On my last save I personally went for 2x3 reactor supplied by bots with nice symmetrical design, meant to be stamped all over the lakes. 2x3 is not much more complicated than 2x3 - requiring 80 heat exchangers and 138 turbines, and it outputs a nice round number of 800 MW Next I do kinda disagree with you design process of limiting your reactor output by water pumps, like other comments have mentioned. Water pumps are really not that expensive, and sometimes building more of them just produces a simpler design with less piping involved, especially if we are doing landfill design that's meant to go over lakes - you can just set up a pump for each row of heat exchangers, and leave it at that. So I personally overbuild my water pumps and turbines a bit for the sake of simplicity and symmetry - like for example my 2x3 design actually has 144 turbines instead of 138, because it consists of 8 separate sections with 18 turbines each, and I want each being able to give maximum output. I can connect them together and remove some of the turbines, but honestly I don't see the need for that and the current version is cleaner and aesthetically pleasing. The setup has also been tested, and it does produce exactly 800 MW at maximum capacity like it should. There is of course a consideration that you usually want to have more power available than your base consumes, so having your design limited by water pumps is probably pretty much irrelevant in the long run, I guess our difference here is that I prefer to round up and you opted to round down in this case.
Agreed. If the argument is that uranium is so cheap/plentiful then why not just build 1x1 (lone reactor) setups? Anything larger is for the purpose of getting the neighbor bonus which *saves fuel* . So on one hand he says you need 2x2 for the efficiency, and then on the other hand limiting the water (and no steam/circuit) because uranium is cheap and plentiful so just ignore the inefficiencies. Can't have it both ways.
Are you sure about the math? If I do the math, 60*1.08*1.16*.0007=0.52 fuel per miner every 200 seconds. In that case 2 miners are enough for 1 reactor. 7.6 miners seems more like what you need for the entire setup of 4 reactors you build
You're right about the 2x2 going to a 2x3 reactor grid that it consumes just 50% more fuel, but your power output goes from 480MW to 800MW, almost doubling the power. So very much worth it to make such a design. I made a design which is almost a perfect mirror, it even includes roboports and has 1 belt going through the design which supplies the fuel and carries the spent fuel out as well without letting the input fuel passing through, it's stopped right after the reactors. And my design also lets you tile it perfectly in case you need more power. Both belts run parallel to the steam engines, one supplies the fresh fuel, the other gathers the spent fuel and they connect nicely. I may have to redesign it someday because my roboports aren't perfectly on-grid atm.
This reactor design produces 480MW(th) out of which 464MW(e) is used to produce electric power. A4W reactors on US Navy Nimitz-Class Aircraft carriers produce up to 550MW(th) each (1100MW total), of which only about 100MW is used for electricity production (wikipedia), the rest of the capacity is used for ship's propulsion, plus some extra capacity for safety margin and sustaining the aircraft carrier with one of the two reactors down for maintenance or recovery from a casualty. This design uses reactors that, unlike A4W reactors, operate at 100% capacity at all times. When operating A4W reactors, we typically maintain power as low as practical for the ship's operating conditions. Let's just say with both reactors running and normal day-to-day operations when in a mission area, they're typically operating at less than 50% each. The reason for using both reactors when one could carry the load demanded of both is to have operational flexibility if one reactor needs to go offline for maintenance or in response to a casualty. Factorio reactors also use fuel enriched to... exactly 5% (1/20 parts U-235, 19/20 parts U-238) whereas the enrichment of navy nuclear reactors is... I forget exactly but it's legit well over 90%. It may be 98-99% even. This allows navy nuclear reactors to load enough power-production capability to last a ship for literally decades. Someone with a lot more dedication than me could probably calculate the mass of U-235 consumed in 200 seconds to produce... well... actually, the fact that factorio reactors have this neighbor bonus thing kind of throws off the calculations. Why does a separate independent reactor adjacent to another reactor cause the fuel to produce more usable energy?
Maybe the neighbor bonus is modelling a larger combined reactor instead of many smaller adjacent ones. In a larger reactor core, the otherwise wasted neutrons could be used.
I remember my first time playing factorio and i couldn´t (and didn´t want to) figure out the nuclear reactors, so i just stamped down a random blueprint from the internet. I think with that tutorial many new players can learn it fast, great tutorial! Wish you did this when I was new to factorio :D
One touch I like to add to my nuclear reactors is to have the fuel creation, fuel reprocessing, and the inserters for the reactors all on an independent power grid run on a few solar panels and accumulators. If you make a mistake that results in a complete blackout, it can be a nightmare to get things started again without that.
i usually tap off some heat from the end of the reactor rows, closer than any other heat exchanger and with an own water supply, plus backed up by a Boiler fed from a chest with some surplus burnables, and an Accumulator, both with a Global Alert if it ever gets discharged/used. This powers everything directly related to the reactor (Inserters, Combinator stuff, Pumps), and also the Uranium refinement Process. The Output Turbines are then on their own Grid(s), completely isolated from the reactor-internal Grid. (a few Turbines on the Edge might feed both Grids to allow for higher consumption on the internal Grid, but i think thats not necesary)
I would say that applies to coal/solid fuel power much more so. If you screw something up and lose power, kickstarting nuclear reactors isn’t that big of a deal you should be able to just clear out a clog and then manually fuel the reactors and it should get going again long enough for you to find the issue and correct it. I just keep a buffer of nuclear fuel and uranium if I need to do so. If your large coal/solid fuel generation facility blacks out (say you accidentally redirect a main fuel belt) getting that going is a pain because you need to feed coal to dozens of reactors long enough to get fuel produced and to the boilers, which is why I like to have a fuel supply on its own circuit that needs relatively few boilers to maintain that can easily be fed manually or with a single chest.
Overall I like it. - Totally agree with your opinion on steam storage, waste of effort. - Not sure I agree with your conclusion on 2x2 layout vs infinitely tile-able, but you make a good case. - The only thing I really disagree with is capping your design on offshore pump capacity. I would increase to 2 pumps per row (for a total of 8), or do 3 pumps per 2 rows (for a total of 6), so the reactors are your cap. I agree that nuclear fuel is cheap, but so are pumps. - I definitely agree that you can run nuclear power for quite a while before requiring kovarex, just stashing 238 in a dozen steel chests for a while. - 28:20 maybe some people, but when you said math my ears perked up! Loving this master class series!
I keep getting the figure of 1,9 miners per reactor. Can I see your math? Here's mine. (Bare with me as I try to lay out the equation in youtube comments and make it readable. ) 1 Reactor = 1 Fc/200 sec 10 U235 = 116 Fc 1 U235 = 1323 ore 1 Drill = 18 ore/min ------------------------------- 1 reactor = ? Drills 1 Fc * 10 U235 * 1323 ore * 1 min * 1 Drill * 60 sec 1 Reactor = ----------------------------------------------------------------------------------------- = 1,9 Drills 200 sec * 116 Fc * 1 U235 * 18 ore * 1 min
I've also tried the equation in reverse and I get that 1 Drill can fuel 0,53 Reactors. which means that 1,9 Drills can fuel 1 reactor. One of us is making a mistake.
I think Nilaus meant to say 8 Miners can support 4 Reactors. 1 Fuel Cell / 200 sec is 0.3 Fuel cells / min and here are the calculations for that with +20% Mining Prod. and Prod 1 modules: kirkmcdonald.github.io/calc.html#zip=bY3NCgMhDITfJqeGuv2FhTyM1VikRiUq7OPXpbSnwjCH+YYZb7slg8sdbwa00hXcbhIzXeBFJQR4cOrEW1VuDbva3GrRjnsMUrV4OhmInaXRmDQOwTA4oeOUVl3Px8WAFD8S/ytU/unwpXPUzbOYnx/wBg==
@@dakon2154 8 miners per 4 reactors would make sense with the equation as well (1,9 miners per reactor * 4 = 7,6). Thanks for clearing that up. As an engineer it was really eating away at my soul.
Since the refined uranium is 0.7% chance of earning, it means, on average, 7 for each 1000, which means 1 every 142.86, rounding up to 143 ore. So, if every 143 ore, it would output 1 U235, I don't get where you got those 1323.
@@mobius1qwe One refined uranium requires 10 ore. So 1429 ore would be needed for one. You actually right. Where does the 1323 come form? He does say on average so maybe this is enough for a reasonable chance to get one. I missed the workshop session and OD doesn't appear to be available ATM, so I can't check.
Why do you say sulfur for sulfuric acid? If sulfuric acid is too hard to say wouldn't just "acid" be a more correct and less confusing word? I find it very distracting.
"if you are not an absolute master in designing these, don't do it, its not worth it" Its a designing game, you design something and work out to increase its efficiency.
Well, I think it depends on what your goals are. Some people like to make the larger setups just to do it and because they enjoy the design process; others are more practically oriented, so beyond a certain point if it doesn't actually further benefit the functionality of the factory it isn't worth doing.
@@strategicsage7694 If you do not understand how a setup/blueprint design works; you shouldn't use it. Thats almost all the fun of this game. I don't know a single person who enjoys factorio just by copy pasting some blueprints off the internet and launching a rocket.
@@kitsukeita Where do you draw these people you know who play factorio from? That's certain not what I observe. I see a lot of channels where the biggest request is for the blueprint downloads. A number of people specifically say it's because they can't be bothered to make their own designs and/or just want to get past whatever aspect is causing them trouble. People play Factorio for a lot of different reasons, but there's definitely a sizable group of people who are more interested in the solution (blueprint) than the understanding of it or the design process.
@@strategicsage7694 People might ask for blueprints for various reasons, maybe they want to assimilate it in their own design or they want to further improve it or they just like it more than their own design; but if you do not understand what the blueprint is supposed to be doing & how, i see no point in playing; might as well copy those small starter base prints from the internet(meant for advancing faster in modded games), hand craft the components/buildings, launch the rocket and call it a day. My data of people comes from my steam friends/factorio steam groups/Bobs,Py + some other mod discord channels.
I'm struggling to bring enough water from my distant offshore pumps because pipes really limit fluid flow. So I need some tanks and electric pumps to supply throughput. I couldn't find a good recent discussion on factorio fluid mechanics. I'm wondering why 1 electric pump with in-game listed 12000 units per second can't push more than 1200 units per second of fluid through a water pipeline supplied by 10 or more offshore pumps at 1200 units per second of water each.
Here is why I disagree with your perspective about steam storage: One: steam storage is eeeeeasy. Just slap some tanks down in the free space along the sides of your boilers. The design difficulty is zero, and compared to the cost of the rest of the parts, a dozen tanks is basically free. Two: peak capacity being higher than sustained capacity is useful for e.g. situations where you are transitioning from heavy solar or have lots of laser turrets or other situations where your power consumption needs might be "bursty". Three: reactors which are hot (>900C) EXPLODE if they are destroyed by damage, with the same effect as a nuke going off. Nothing like a few biters getting into your power plant, and then the entire thing literally blows up and you have to rebuild all of those extremely steel heavy structures on a factory with no power. Four: It adds resilience to brief outages. If I need to disconnect the water supply to re-route pipes, I can do that, and it will run for 10 minutes. Technically, storing heat directly in pipes is more space efficient (500MJ/tile compared to steam tanks at ~269MJ/tile), but just building more heat pipes in random places is ineffective because as you point out, heat transfer becomes a bottleneck. I disagree with your perspective about circuit-limited reactor designs too, because they are very simple to implement well. I homerolled a controller design which is 3x4 tiles, contains only 6 combinators, and it's both jam-resistant and idiot-proof. It's the difference between waiting until kovarex to spin up nuclear fuel for your trains, and starting it as soon as you can.
You make good points. I guess it depends on the situation: a big base with low UPS means not using tanks will make the game run faster, because it doesn’t deal with the fluid calculations involved, and a base on a limited-resources-run is way better off implementing efficiency measures. It ultimately comes down to preference and convenience.
Well, if bitters chew your nuclear reactor, you have something very, very wrong with your base defence anyway. I do agree that steam and circuit based nuclear power are simple to build, and they are scalable. You can easily build a 480 MW power plant (that's 4 reactors) even if you only need 60 MW right now (it's usually the power I need when I start looking into nuclear power). You won't use that much fuel, stockpiling for the future, and you won't need to deal power for a while.
@@giovannidueck9094 Megabase skip nuclear power entirely and only use solar. Solar have no UPS cost. Nuclear power can't beat that, with or without steam tanks.
Regarding point 3: With the addition of the spidertron, this has become a non-issue. Simply building the reactors on an island that is only accessible by spidertron makes them completely invulnerable.
Doesn't really matter once your Kovarex process is up and running, as you basically get unlimited nuclear fuel. Get a couple uranium patches being mined and processed and it's pretty much impossible to be short on it.
Hi Nilaus, Hope you are doing well! I'm aware this video is a little old. But I just started with Factorio and I'm at the stage where I'm setting up nuclear power. This guide and your blueprints are just GOLD. Like Jesus, this helped me out SO MUCH!! A very big thanks for your help and effort, you Mr are GOLD!!
Thank you for another excellent master class series. I really enjoyed your math and motivations on why anything larger than 2x2 plants are irrelevant. Your arguments that a decent sized uranium patch will essentially provide unlimited power is very convincing. The modular design of these nuclear power plants are beautiful and I will be sure to use them in my next Factorio playthrough!
A few remarks: You are using assembling machine 3. Those are not unlocked at blue science, but at purple science. With only blue science, you can only use 2 modules per machine (8% prod instead of 16%), reducing the amount of fuel you gain from an ore patch. It doesn't change significantly your conclusion, nuclear power is dirt cheap in fuel and should absolutely be rushed in the mid game (as soon as you can afford the enormous oneshot setup cost and you need the enormous power generation). Productivity 3 is completely overkill in general, unless you already have a massive base. They cost a massive amount of resources for very little gain. By the time you can afford to productivity 3 your nuclear setup, you most likely have Kovarex. Using steam storage is mostly there to "future proof" your design (so you don't have to come back before a while) and to protect your grid from overloading because of laser turrets. Tanks and combinators cost next to nothing when compared to the enormous cost of the nuclear plant itself, so frankly there is little reason to not do it. If you wonder about UPS, DON'T DO NUCLEAR AT ALL, do solar. Solar cost nothing in UPS, that's why all the megabases are using solar. Once you have nuclear power up and running, you should upgrade your armor piercing bullets to uranium bullets. It only cost you U238 (the useless one) and uranium bullets does 3x the damage. They will shred through any bitters. You can do the same thing for your tank shells. Don't bother with nuclear missile however, those are next to impossible to build before Kovarex.
That Maths just blew my mind.... How can it be so low?!?! :D How much excess of the "Waste Uranium" does that "8" Miner process produce per minute after producing Fuel? If I am not converting it to Ammo, I would possibly need huge storage values until Kovarex Refinement comes online. I might try the maths on that myself.
Hey Nilaus, this was a definite master class. Factorio graduate studies vs your 4 year first degree. Btw it is so cool when things like your uranium non-scarcity math appears on YT and I can remember being in the twitch stream when the calculation first happened ((: Bravo! EDIT: I’m SO used to seeing radial symmetry nuke plans designs in your playthroughs. My eyes almost can’t believe this version. That’s the real controversial opinion hehe
What do you do with the thousands of U-238 ? I save them until i have rushed kovarex so i never need to mine uranium ever again :D the longer you run without kovarex,the more tens of thousands of worthless 238 you aquire... Love the videos by the way!
Thanks again for some more great content. Really appreciate your hard work and clarity on these guides. So helpful. Can I ask a question, though? I don't understand how you worked out that centrifuges require an average of 1.3k ore to be mined to produce only 1 piece of U235. My calculations work it out at 132.3, so that's makes processing of ore into U235 10 times more productive than you say it is (30:05). This being the case, it actually make your point that you don't need to bother with steam capture or kovarex 10x more potent! What am I not understanding? Am I missing something? Thanks! --- Here's my math, just in case. U235 mining yield @ 0.7% = 0.007; 8% production bonus on centrifuge = 1.08 centrifuge yield; the combined yield = mining yield x centrifuge yield; => combined yield = 0.007 x 1.08 = 0.00756 ∴ Combined avg. yield = 0.00756 Calculate the inverse to switch unit to U235 rather than ore; => 1 / 0.00756 ≅ 132.3 : 1 ∴ Avg. ore needed to produce one U235 is approximately 132
Another great design that is very efficient in space and power. My nuclear is blueprint of choice until now has been the KOS nuclear plant. It is modular, but not nearly this compact. Thanks for the fresh look and design!
I get it and I knew about the abundance of fision material since the first time I played the game (because I can read). I still don't like running my power plant at 1000°C.
Excellent design and it seems you came to lot of the same conclusions as I did when doing my nuclear setup (although I never did the miners per reactor calculation, that is kinda crazy). I also ended up with a 2x2 reactor setup (since you've got most of the neighbor bonus by then) and pulling water from one side without circuit controls or steam buffers. I sometimes add circuit control when I run these before unlocking kovarex but just adding a single storage tank is sufficient to monitor steam levels. Full steam storage is totally unnecessary. It is good to see such a useful design getting highlighted. Cheers!
yiu might not need koverax but if you don't have have it you have to store U238 somwhere. Also if you don't use koverax to feed your reactors it will take even longer to get the U235 you need to start koverax since now your also using up some of it to run your reactora
One thing I like to do is add steam storage tanks and wire the system to feed fuel cells only when the steam storage runs low. This way, I use the fuel cells efficiently.
@@littlebuilder8132then just build 1x1 reactor (lone reactor) setups if you think uranium is so plentiful that you don't need to be efficient. I love Nilaus, but his argument here makes no sense. He is saying to build 2x2 for efficiency, but then don't build steam/circuits because uranium is plentiful so you don't need to care about efficiency. Can't have it both ways.
you really don't need to be efficient with uranium once you have an enrichment process researched and running. you lose on a lot of free power if you build just singular reactors @@beepbop6697
argument for steam storage: unexpected power surges (or not looking at power consumtion) that enables you to put more steam turbines than the heat exchangers can usually handle thus allowing for a compact battery
Using those numbers, we should only need around 2 minuers per reactor. Either the ratios you presented aren't correct, or else I made a mistake in my calculations somewhere. Can anyone help with this? So we have: In the miners: Uranium ore per miner per minute = 18 Uranium ore per miner per second = 18 / 60 = 0.3 In the enrichment process: Amt. of Uranium Ore per unit of Uranium 235 = 1323 In the assembling machine: Amt. of Uranium 235 per unit of Uranium fuel cell = 10/116 ~= 0.08 In the reactor: Amt. of fuel cells per reactor per second = 1 / 200 = 0.005 Putting them together: # of miners to constantly power 1 reactor = (1/Uranium ore per miner per second) * Uranium ore per Uranium 235 * Uranium 235 per Fuel Cell * Fuel cells per reactor per second = (1/0.3) * 1323 * 0.08 * 0.005 = 1.764 This means we need less than 2 miners per reactor, as opposed to the ~5.6 miners mentioned in the video.
I think he said it wrong, it's 7.6 miners for the block of 4 reactors. Nilaus often misspeaks but then he is not a native English speaker. (Although it really irks me that he calls "Sulphuric acid" "Sulpher" ALL THE TIME! ) Doesn't detract from the amazing builds and tutorials he produces
Hi Nilaus, enjoyed the vid; one concern I have is the distance covered by the steam-pipe network--I'd wonder whether steam pressure is lost due to the long pipes?
I went with a circuit and steam storage design because otherwise of dozens nuclear power plants would have been sitting at 1000 c degree all the time (maximum capacity) and still consume fuels. All other energy demands are fulfilled by solar arrays during the day. Would you rather have fuel cells be consumed all the time? Or occasionally be turned off? It is kind of like having regular boiler consume coal/tree/etc... all the time. Lot of wastage from not utilizing all of the boiler output (1.8 MW) and miniscule power to move items.
I don't have the funds to support, and thereby show my approval at this time. But I am truly grateful that I can benefit from your knowledge, effort, and passion like this. I hope that my gratitude counts as a small token. (But I also hope that you are getting worthy financial support as well). It's like the feeling when you see a street performer who is breathtaking, but you have no cash on hand. You feel almost like you've taken from them as you walk away, without having given back.
Thank you. Don't worry about it :) I provide this content for free and hope that a small percentage want to and is able to support. I do not want any paywalled content, so just enjoy it :)
Came here only for the blueprint for Nuclear Reactors. Ended up with abundant amount of knowledge. Every single video you've done, you never disappoint
Funny, I grabbed the biggest uranium patch and put 177 miners on it and then had 52 centrifuges going. Once I watched and built this, I realize that is way overkill and there is no chance of my power plant ever running out of fuel, even if I duplicate it a lot. Thanks for the awesome info as usual!
i cant believe i just watched a 40 minute video about factorio nuclear power plants... and i gotta say, not a single minute was not needed great vid thanks for the help
@@legafredo I don't. Perhaps you misunderstood my original comment? Or maybe I was unclear. Either way I'd like the previous 'anti fluid mixing' feature back. Where you can't place a water pipe down next to a crude oil one because they contain different fluids for example.
@@legafredo I don't know. Back when I posted the original comment. The fluid prevention system got changed to the stupid way, where you can delete contents in the pipes if fluids got mixed. But now it might be back to the better way. I'm not quite sure
I need to pull the blueprint, but I spent around 20 hours one time a year or so back trying to design the perfect nuclear power system. the end system used 6/8 reactors, double heat pipes, tanks for water and steam but not for stream storage because there was exactly 0 fluid dynamics checks because everything was moved by pumps. the water came in through the middle, between the turbines and the reactors, and across the bottom. if you flipped the blueprint 180*, you could stick another one on the bottom and get a slightly higher bonus, and if you put them side-by-side they could share heat to speed up the initial heating process, and would share a mutual water input stream. the pumping system could handle just less than 3x8 reactor setups horizontally before it would start to have problems getting enough water into the heat exchangers at the back. the last heat pipe held a constant 525 degrees when the system was fully hot and fully utilized, which is why I chose to stick to 6/8 reactor systems. can't get the heat out far enough to need any more than that. I think that's the earliest hard stopping block. when I say 6/8 reactor setups, I mean 8 to start untill all steam tanks are full, then take the bottom 2 away to leave 6 which maintain enough output to get 1.2gw steady, but built with enough turbines to do just about double that because that's what the system should be able to handle since it's duty cycle is only about 50%. the best and only argument for steam storage is the day/night cycle and solar panels. Free(ish) power, and you definitely shouldn't need all of your reactors at full steam during the day if you aren't relying solely on them for your whole base's power. why not bank the rest of the steam for later? it's only a few tanks. pump in pump out and you won't have to worry about fluid dynamics checks. sure it's a bit bigger but, like, is your map size limited or something? just tack on a tank field connected by turbines to burn off the overflow in times of very high usage. obviously I used bots to deliver the fuel. I mean, who would even consider trying to belt that stuff in and out. lol :P (I kid) getting water in was always the problem, but can be overcome with proper placement and delivery techniques (pumps mostly) All in all, that's a great guide though. well thought out and reasoned and the design is great for a 'bases first nuke field' setup. the math may work out to being able to do about 1 reactor to 8 miners, but I'd like to see that actually work in practice. what would the power consumption of 8 miners and (I think) 8 centrifuges with 8% productivity modules be compared to the power output of a reactor? what's the real power profit there? or is that just another really roundabout way of banking solar power when you think about it?
@Tommy Taffy has been a long time since I was a factorio master, but, when I designed this fluid dynamics were far more rudimentary than they are today. at the same time, it's likely that pumps bypass the standard fluid dynamics that seek to balance population and flow between connected pipes and fluid containers. pumps move fluid from one vessel directly to another with no mind to balancing or viscosity. if you chain pumps, or pump->vessel->pump, once filled, will flow at pump speed throughout the system (minus any fluid used or forked in previous steps). fullstop. the gain in flow rate and fluid availability in deep systems is not negligible. try it for yourself.. chain pipes, underground, tanks... then do pump/tank/pump and see how much faster the end tank in the pumped system fills over any other layout.... then to prove it to yourself, in the middle of the pump setup, switch 3 pumps for 2 tanks and put 3 in a row in the middle of your loop, or connect with pipes instead, and see how just 1 fluid dynamics calculation substantially slows the rate the end tank fills. I'm going to dig up that blueprint... I'm sure I still have a vanilla world from 3 years ago with it built somewhere.
Great guide! It's really surprising how few miners and centrifuges are needed even without kovarex. I always start doing on-site uranium processing as soon as I unlock the research. I don't use modules, so I go by a 10 miners : 3 centrifuges ratio. I usually start with a 2 reactor build because its cheaper to build early on and satisfies my power needs for a long time, and just the 10 miners and 3 centrifuges is enough to support it indefinitely. The nice thing about having a tiny uranium processing build is that you can put efficiency modules in everything and never have to worry about biters. Sometimes I will use steam buffering for the 2-reactor setup because you only need a couple fluid tanks (maybe 10 or 20) and the wiring isn't too bad. For the 2x2 build though its definitely not worth it, especially after you get Kovarex.
Err, is no one going to point out that's not a 464MW Plant? The water limitation means your exchangers are only going at 97-point-something efficiency. Knocking 464 down to 450. Unless I missed something? EDIT: Wait, no, I did miss something, that's 48 exchangers, the lost capacity from water shortage is less than the number of missing turbines, nvm. EDIT2: Steam Storage, seems like it could be nice for lasers. Turbines and tanks are cheaper than accumulators right? Well, if you're short on oil. Maybe not otherwise.
Steam storage still needs to be converted to electricity through the turbines, which have a fixed maximum rate. Accumulators don't have such a limitation, and act like super-capacitors that can supply as much transient power as necessary, e.g. for laser turrets.
It is only tileable if we disregard water supply. If base needs several GW of power it gets real messy real quick. I prefer more spread thin design, it also allows for +3 adjacency even though it is not a big deal. Also, it would be excellent addition to show the design in action, to demonstrate that it is indeed producing at maximum capacity. Especially after what you have said about difficulty of designs with fluids.
I'm not sure if that's what the nuclear fuel recipe has always been but I swear I remember one of the main reasons you needed Kovarex was because of the sheer amount of U-238 that you'd have to store which was why so many people wanted an incinerator of some sort for the game.
Ah, I always wondered why he invented that.. It would make sense exactly in the situation that you describe. And it would also explain why we always have way too much uranium patches by default.
It's very cool to see that you actually don't need to do anything complex to support nuclear reactors, but part of the fun I get out of Factorio is optimizing every last thing down to the stuff that doesn't even need it
My entire factory consumed less power than a single reactor outputs (b/c solar) but I wanted better efficiency so built a 2x2 reactor anyway. No way in hell am I wasting all that fuel running full time. Definitely worth the circuit effort... though I needed to make a T-flop train circuit to keep accumulators charging on only solar. That part was a mess lol
when you're at nuclear reactors, you're at bots or am i a weirdo here? ^^ by having bots it would be easy to feed and cleanup your reactors with those, why design it in a way where you will have hundreds upon hundreds of fuel cells sitting on belts and design the entire thing with the space for additional belts when bots can do it?
1 fuelcell / minute is enough to just 3.333 reactors. Not the 200 reactors you said? If this is true you need 456 miners / reactor and not "7.6" as you said.
@@Nilaus I checked it and you are right. I missunderstood you when you said "1 fuelcell can support 200 reactors" and I know that isnt true so I calculated from that to prove it wrong. I listened again and know you didnt mean it that way. 31:06
To make full use of the neighbor bonus, the most optimal nuclear build is obviously to place an infinitely large amount of reactors, that way every single reactor has every face connected to another. I'm both disappointed and concerned, as I saw you as a mind for the efficient, and yet your design completely trashed this obvious setup.
I like this but I feel the need to point out that you kind of missed to explain how essential it is to remove the used fuel cells from the reactors (considering you stated this is aimed at a new player) during the first part of the video. I bet its a common mistake amongs first-timers to forget/miss this since it is a very unique function in the game. In the build of course you show it but its really not talked about alot.
An interesting point is that even if you're committed to steam storage for your reactors, you probably only need to build your first reactor with steam storage. You'll be building a second reactor because you're about to exceed the capacity of the first one, so every other reactor you build can be running at full load without "wasting" power anyhow.
This is what i do, I rush nuclear to build a 2x2 reactor with removable steam storage, I use the steam storage function while my base is under 480MW and remove the tank section when i hit full load and paste down another 2x2 reactor when I need more power. I know wasting uranium isn't really a concern but i like the puzzle of it and like not wasting things.
good design but inefficient. You only put in enough 235 to heat the exchanges. then take the 235 back out. Storing the Heated Steam is the most Efficient Method Period. All you have to do is create a simple circuit to register the 500/750/900c or what ever your mod is. Once the Exchangers hit target temp filter back out the 235, when target reaches below like 30% of max temp filter back in.
I just put some tanks for the steam down and oversize the turbines with like 25% overhead if i run single nuclear systems that stand alone with solar. If i have peak demands it wil draw from the tanks but they will always be topped off when demand lowers. Its a safeguard for me that when i do a expansion that is energy intensive i don't make the mistake of using more energy than a single nuclear system can deliver. It gives me some time before i can upgrade.
I use steam storage with additional turbines as a means to store energy for turret spikes. I also setup an alert system so that once it is below a certain point(power use exceeds actual power production) I can go add another module to the power plant.
you can also use the steam storage to prevent over heating of your reactors : nuclear plants heat up to 1000°C no matter the consumption, any excess heat is wasted. If you setup things so that you only empty your reactors when steam storage goes bellow some level, and you only fill back when emptying, you will minimize fuel waste.
If you show us that you need an 8:1 ratio of miners to nuclear plant just after unlocking this tech (and having prod mod #1 available), then I don't follow how, for four (4) plants, you only need 24 miners? My math says 4 * 8 = 32 miners. Later on, when you have prod mod #3 available and the ratio goes down to 6:1, then I can see 4 * 6 = 24 kicking in. But I just don't see how 24 miners "in the beginning is really enough for 4 nuclear power plants". Please enlighten us?
When he did those calculations on twitch he was calculating it in terms of the 2x2 power plant. According to plantkirkmcdonald.github.io/calc.html#zip=bY3NCgMhDITfJqeGuv2FhTyM1VikRiUq7OPXpbSnwjCH+YYZb7slg8sdbwa00hXcbhIzXeBFJQR4cOrEW1VuDbva3GrRjnsMUrV4OhmInaXRmDQOwTA4oeOUVl3Px8WAFD8S/ytU/unwpXPUzbOYnx/wBg== you only need 1.91 Miners to support 1 reactor
That was a little bit of a mistake on his part, not in his math, just in his speaking. 8 drills with productivity modules can actually fully support 4 reactors. He undersold it.
Thanks Nilaus, you completely convinced me to not use steam storage. My little 600k uranium patch with a fully beaconed kovarex should last forever. :D
Here’s my noob pov : he good thing about Kovarex is that it consumes excess U-238. I’d rather see my Kovarex setup blocked by the excess of U-235 than my whole Uranium refinery blocked by the excess of U-238. But maybe I’m just missing a way of consuming that excess
Ладно, сейчас я расскажу тебе почему НУЖНО строить баки с паром. Смотри внимательно. Ты построил электростанцию без баков -> у тебя закончился уран по ряду причин(закончился, кусаки съели, упало производство из-за перегрузки сети и т.п.) -> твоя энергосеть упала моментально как только уран кончился. Когда же у тебя есть хранилища пара, то у тебя в запасе будет еще пара минут, чтобы понять, где ты налажал и все исправить, прежде, чем перезапускать всё производство. Ну и да, хранилища пара>>>аккумулятор, который вообще говно. Поэтому их нужно строить даже у каменных бойлеров
There is a good way to measure the actual power produced with an "electric-energy-interface" (you have to create it from console). When you set it to consume power, it has the lowest priority, same as an accumulator. You'll be able to see on the charts how much power is going into it. This is nice because it not only reveals flow issues but also includes all the pumps or whatever you use to keep the reactors running.
Hmm, this is interesting. I didn't know *just* how abundant uranium was, though I was aware it was ridiculously abundant. I enjoy designing and have made some ridiculously large power plants. I've done a 2x16 model and I think it's the best way to learn about water and flow mechanics. The amount of pumps just to increase flow for the water I had to use was staggering and the steam required a lot of attention too. I've ended up with hundreds of thousands of Uranium, both shiny and dark, and I guess I never really thought about it beyond knowing that between steam storage as batteries and efficiency from the amount of reactors I was using that I was cutting down on my uranium consumption significantly. I'll probably keep doing that. Definitely once I get to the point where I need more than 4 reactors, the design problems involved with nuclear is very entertaining :D.
The math here seems pretty off. 1 fuel last 200s, and if 10 u235 produces 116 fuel. That means 1 reactor need 1/11.6/200≈0.000431 u235/s 1 miner produces 18 ores for 1 min, which is 0.3/s. If every 1323 ores produces 1 u235. That means 1 miner will produces 0.3/1323≈0.000227 u235/s. By dividing these two, you get 0.000431/0.000227≈1.9. You need 2 miners to feed 1 reactor!
I'm not sure I'm doing the math right but I got a different ratio of mines to reactors. Taking the numbers in the video I get the equation (18/60)*(1/1323)*(116/10)*200 =0.526 reactors per mine or 1.90 mines per reactor, which is off by a factor of 4. Did you mean it takes 8 mines per your design or 8 per reactor because I don't see where else the 4 could've come from
Something one may find interesting is in real life the contetration of u235 is about.7% and the amount of u235 in reactor grade fuel is 5% the amount of u235 in a fuel rod in Factorio is 1 U235 and 19 u238 which is 1in20 exactly 5% so the fuel is actually reactor grade also sulfuric acid is actually used in uranium refining
Wow...this video demonstrates what killed this game. There is a LARGE portion of players who LOVE to build...who HATE and are utterly dragged down by the REQUIREMENT of math calculation. I fall in between actually. I LOVE math ...but I can see (as this video clearly demonstrates) that complicated computation is the bane of in-game """FUN"""! Very sad to see that at the very least, the designers did not even try to make the mathematical proportion be easy ratios.
Excellent video, I love how well thought out and justified your design decisions are. One bit of push back though, your comment on storing steam, you say it's just not worth it. Not worth the cost of fluid tanks? It would seem that while fuel in this case is inexpensive, so is the storage tank. Even just a few storage tanks can help smooth out the power production and increase potential output of the system if used correctly. In my mind adding 4 storage tanks to you design and adding a few extra turbines (8 or so) could account for the variable power usage. It wouldn't increase steady state output but it would prevent brown outs for a small marginal cost.
Once I got kovarex I just stamp these down all over the place. I currently have 10 of this exact design in my base. Any time I get to about 75% power usage I throw another two in. While the extra materials would be no waste at all I feel that any time spent optimizing would be a waste, since I could be doing something else.
Steam tanks would only increase UPS consumption, if you design properly your nuclear power plant, you can guarantee a steady output of power without issues, not needing switches, tanks, and all those things. Also, Nilaus uses too many regular pipes on his setup that if you switch to underground pipes, you would improve the flow rate of the steam and preserve UPS as well.
Mobius one steam tanks effectively count as a pipe. You only need to add a couple of tanks for monitoring purposes, and that’s basically zero impact on UPS. If UPS was that much of a concern then optimizing the design with fewer pipes would be favorable.
@@malventano so, adding tanks you add 1 problem with it that is Pressure. Since you also have to think about keeping pressure on the pipes to a decent flow, you are adding more entities than using only pipes. 2 Underground pipes counts as 3 regular pipes, a tank might count as 1 pipe but it requires pumps that counts as 1 pipe + 1 engine, so it would be around 4 pipes worth since you gotta handle pressure.
Sounds like he's more worried about the CPU impact at that point because going into full megabase mode with the level of power surplus that nuclear can provide. Not so much about the plant operation in-game, but the simulation cost that can slow the whole game down.
My newest map has absurdly scarce nuclear power, so scarce that I dont think I want to run my base on nuclear power as it'll mean I wont be able to build my 70k nukes. (Patch got 85% devpurded by water in map gen, only 2.1 mil left with high richness (dont think I can fit over 12 miners) so gotta do kovarex)
I'm still going to stick with my 2N reactor over a lake. It's not just about "saving uranium". Honestly. mostly it's about pushing the limits and getting more bang for buck as standalone principles, but practically it's about space and time. Sure I can copy/paste a quad reactor, or I can copy paste my 3x2 segment of a 2N reactor and make fewer of them overall; my concrete factory is slow and already occupied in tiling the entire world. If you're going to argue to just make it bigger to make up for efficiency losses, why not just use bog-standard boilers and belt in some nuclear fuel?
To be fair, ive been watching for ages and actually took notice to check it i was subscribed and i wasnt. I swear i subscribed a while ago, sorry Sir! Love the content it really is absolute mastery
8 reactors in a 3x3 grid, with one of the sides removed is actually the most efficient in terms of the neighbor bonus. Use long handed inserters to put the fuel in, and you're good to go.
