Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
Nuclear has the same issue, with different numbers. Each heat exchanger, which acts like the boiler making steam, produces approx 103 steam/sec and the steam turbines use 60 steam/sec. (The exact value for the exchangers is the nice non-round number 10,000/97 or ~103.0927835.) Each reactor (or reactor equivalent) makes 40 MW of heat and can urn exactly 4 heat exchangers. Those 4 can then run almost, but not exactly, 7 turbines. 6 turbines with a bit over 52 steam/sec left. With nuclear systems the exact ratios are BIG numbers (292 nuclear reactors, 4656 heat exchangers, 8000 steam turbine, and 400 offshore pumps worth of water to keep it running.) If you set the pipes such that the steam input from the exchangers is less than the steam used by the turbines connected to it, only the 10MW max will be produced by the turbines.
My design is going to need a few adjustments, lol
say something when it matters instead of this whatevs the ♥♥♥♥ you are doing
Maybe you got the topics confused...
if the fbi see's you talkin about this
The first part is somewhat true.
For heat pipes, the design and ratio do matter. A bit too long, the heat exchangers won't get the necessary heat, etc.
Steam Turbines, Steam Pipes, and Steam Storage matter a little less, but it can be useful to have a buffer between Nuclear and Steam Turbines. It's not mandatory, but it's nice to have.
The last part is slightly more complicated. People have different opinions. I belong to the school that thinks Nuclear Fuel cell input should be regulated.
Better not to immediately consume an extra set of Nuclear Fuel cells as soon as the reactor is empty.
Having no buffer between consumption and production can be good in some cases, such as ore smelting production lines and plate consumers, but not when doing so would consume unnecessary products to overproduce unnecessary extra energy to meet the current energy demand.
I have had conversations with people who have strong opinions in the past, and that is okay.
There is more than one valid Nuclear Reactor design out there that can meet different needs/goals.
One offshore pump will put 1200 water/sec into the pipes, or whatever is attached to it. This is a limitation of the offshore pump's specs, not the pipes themselves. Without any pipes in that line you can give water to 11 exchangers. One or three pipes will not be enough to affect that so you can cover a bit of a gap between the offshore pump and the line of exchangers, if you need to.
Pipes will carry, in theory 12000 water/sec., 10x what an offshore pump can make. That flow is nearly impossible, or impractical, to get in a useful layout, however, so don't build expecting that to be the rate you get. The higher the number of pipes the more the flow rate drops. Keep it in a small range, say under 50 total, and you can probably count of the flow from an offshore pump feeding all 10 exchangers.
Heat pipes don't always heat up all the way to the end. There's a limit to how many heat pipes you can have after the reactor and still be hot enough to make an exchanger work. That range can be extended by using parallel heat pipes, but that also uses another tile in width.
To make a print you can tile, it only makes sense to me to aim for the ability to create a growing reactor setup with 2 wide reactors that stretch both ways as far as you want. So you can start with say a 2x3 core and them make it 2x4, then 2x6, then 2x7, etc. If you make a print that tiles but has a gap between the reactors it's wasting possible power. In a 2x3 or higher grid every "inside" reactor is making 160 MW of power and each corner reactor is making 120 MW of power. Every time you add a "tile" of 2 reactors to that you are adding 320 MW, or for the next comparison, a tile of 2x2 adds 640 MW, to the total. If you make a print that tiles with a 2x2 reactor setup but that doesn't let the new reactors connect to the existing setup you are only adding 480 MW of new power. A print can be made in the non-connected style, but it's missing out on 160 MW every time you use it rather than a connecting version.
With the connecting version of a print in mind, you should probably make the print as if each reactor was making 160 MW worth of heat, using 16 heat exchangers for example. The wider you make the print, 2 reactors, 4 reactors or 6, etc., the more freedom you will have to be creative and to optimize the turbine to exchanger ratio. For a single-wide you're limited to 5 tiles and to make it always work using 28 turbines, wasting more than half a turbine every time. A double-wide, 4 reactor, print would give a width of 10 tiles to work with and only use 55 turbines and waste less than 2% of the turbine capacity.
Converting a built reactor setup from free-running to fully regulated is not exactly easy. It's possible, just not a simple change. Experiment with both, and there are several ways to 'regulate' the reactors, to find how you want to do it, if at all, and build your print that way. (Save prints, even if incomplete, of your experiments - you might change you mind later and half the work's already done.)
Reactor prints can get quite large. Collecting a bunch of them in your library will eventually make the saves load slower. The one you decide to keep can stay in the library, but consider using the export string function to save other prints, or a book of them, as regular text files on your computer. You can always import them again later. A single print per file with a screenshot of the blueprint from in-game saved with it as an image, with the same name, makes browsing prints on the computer almost as easy as in the game.
Unless you use a waterfill mod, reactor prints with the pumps in the middle, next to the reactor core, are hard to find places for on the map, when you want to grow them to a large size and requires a huge investment in landfill. Prints with the water piped in from outside the area are harder to design, but can be build almost anywhere you can find room to place a collection of offshore pumps. (The change in cliff explosives for Space Age will make placing them a bit harder, but they probably won't need to be all that big until we get the new explosives unlocked anyway.)
Though it is specified in the Wiki, somehow I didn't realize the reality, so I'll offer it here as well. For the neighbor bonus to apply the reactors have to be touching for the entire length of the side, all 5 tiles in common. There's no staggering the "connection" like there is with beacons. In addition, both to the reactors have to be running with fuel in them. A reactor without fuel just acts like a 5x5 heat pipe.
If you are trying to "calculate" the flow of steam in the system just remember that a steam turbine will take its 60 steam/sec out first and then act like a pipe for any left over steam, So in a line of 10 turbines, to the last turbine it will look like a line of 9 pipes and there will be 540 steam/sec less available.
Lastly, when building and testing your ideas, when the flow of water or steam that is happening doesn't seem right based on any math you've done, try replacing pipes, and pumps if you use them. Just delete and rebuild the pipe. Sometimes the order things are built will make a difference in the way the game calculates the flow of steam, or any fluid, and with experiments and redesigns happening as you go, the order is going to be all scrambled up. Just start at the beginning of a line and one-by-one replace each pipe/pump/turbine until you get to the end.
-----------------------------------------------
As far as the fuel conservation and added tanks issue goes I'll just add these caveats.
When you are going for nuclear power you are usually expanding your base to the maximum possible size you can. The size limit you will run into is not resource abundance, its calculations per second ie: frame rate and updates per second (UPS).
1) In a mega base most of your resource consumption will be producing science packs the rest will be expansion, and a tiny bit fighting bugs.
One of the largest drains on UPS is fluid calculations so one of the best strategies is to limit entities moving fluids to the minimum required to do the job. As Chindraba mentioned one off shore pump plus 1 set of pipe to grounds (so you have space for belts or rails) that connect directly to you first heat exchanger in the row is best for fluid calculations while providing maximum water. Adding redundant tanks to store steam will just increase fluid calculations. Additionally, dumping all of the output from each line of heat exchangers into one large tank system that is then fed into separate banks of turbines can create back flow that will reduce the amount of steam that reaches the farthest turbine in the row reducing your potential power output.
I've done the calculations and 1 off shore pump plus 1 set of pipe to grounds feeding directly into a row of 11 heat exchangers that use 1 set of pipe to grounds (or better no pipe to grounds) will produce 1,200 steam/sec that can be consumed exactly by a single row of 20 turbines provided that you have no branches that can cause back flow and provided that all the heat exchangers in the row are at or above 500 C.
Fuel consumption is constant and the power potential of the turbines is static if each turbine is receiving the required amount of steam/sec whether you use that power or not. With that in mind...
2) For vanilla limiting nuclear fuel consumption is not necessary. Not only does it require additional calculations for the circuit network required, it increases the number of entities required to make the same system work, which uses resources that could have gone elsewhere. Not a big deal in the entire scheme of things but worth a note.
Additionally, Katherine of Sky did the calculations at one point in time. Roughly speaking, a 1 million uranium ore patch with pre-space productivity researches can provide enough fuel cells to power a 1 rocket per minute base for 720 real time hours (a full real time month) of continuous use. If you could keep your frame rate at 60 FPS until you used up all of your copper and iron deposits on the entire map you would still have uranium patches left over (if you set up default resource distribution at map creation).
My own personal experience is creating my most recent 5,500 science per minute base. It consumes more than 24 GW of nuclear power (plus additional solar power). I tapped 24 ore patches each of copper and iron and tapped only 4 uranium patches and still reached a point where my centrifuges backed up with material and started idling. My frame rate was 55 at that point so I am not able to expand further.
The TL:DR of caveat #2 is that you don't have to conserve nuclear fuel cells. You won't run out. Conserving UPS is more important at that point in a post rocket launch game.
One offshore pump will supply the 1200 water/sec. Keeping the pipes short to make sure that flow is available to the heat exchangers is a good goal. If we take it that the 1200 water/sec does make it to the bank of heat exchanges, that's fine. Now, the problem: 11 heat exchanges will produce a maximum of 1134 (and minor change) steam/sec. [Exact figure if needed is 1134 2/97]
Well, not exactly. Fully consumed would be correct. 20 steam turbines can exactly consume 1200 steam/sec. 11 heat exchanges can only supply 1134 steam/sec, which will almost satisfy 19 steam turbines, which can consume 1140 steam/sec.
You most definitely need to redo all your calculations. Everything which flows from the 11 heat exchanger premise/value, is completely in error.
The regulated vs unregulated reactor setup is probably an eternal debate which will only be settled once the universe attains heat death.
Nevertheless, UPS is often a reason, though not always, for considering the idea. There are UPS benefits, and costs, for regulating the reactors. Using regulation it's possible to reduce the use of inserters - one of the biggest issues in the UPS battle. It's not just the inserters handling the fuel cells in and out of the reactor, but every one involved in the manufacture, reprocessing, and transportation of those cells. The train traffic is also somewhat involved, though that's usually less UPS-significant than the inserters. The circuits to regulate the thing can be well designed, thrown together, or works of art - I've seen all three. The well designed, and well thought out, can help the UPS, being less computation than is saved with the inserters mostly, while the thrown together versions often cost more than they save.
The biggest consideration, before even considering whether or not to regulate, is the differential between supply and demand on the electrical network. The closer to capacity the system is the less benefit there will be from regulation.
Regulation uses the fact that the reactors always produce heat while running, whether or not the rest of the system (primarily the turbines) need that heat. If you have 8 cores making 1.28 GW and your factory is running at 800 MW then the system could be making nearly 5000 steam/sec extra. Store that in tanks for the duration of on fuel cell burn and when the cells stop burning you will have 2 minutes of power stored in the steam tanks and the reactors can not use fuel for that 2 minutes. {Actually there's also the cool-down and reheating of the reactors to factor in, but the concept remains the same.} If the electric grid is only using 400 MW (tiny for a base, but whatever) then every burn cycle gets over 7 minutes of stored power. Or, if the factory is using 1200 MW then the steam reserves are only about 13 seconds of stored power.
In short, the closer to actual needs that you can produce power the less regulation will matter, or help. Therein lies one of the benefits of narrow print tile designs. If the blueprint is for a pair of reactors, to be added onto a Nx2 reactor group, each print gets an extra 160 MW of power. That is easy to keep the power needs and power supply so close that regulation is pointless. My latest print is 4.48 GW per tile. Much harder to make supply and demand match. Stand-alone prints, usually 4-core or higher, suffer the same problem, with a 4-core design making 480 MW or a 6-core making 640 MW. Still and all, if supply and demand can be kept close, with supply always higher of course, the regulation becomes much less helpful, if at all. Using solar to "fill the gap", with switching to make it secondary rather than primary, can help with that as well - at it's own UPS costs of course.
All the elements of the debate between the two which relate to resources - any resources - is pointless. At the point that UPS is a concern resources are of zero practical concern. During the building phase, when resources can matter, it's just a question of goals and priorities, both of which are personal choices and opinions. There are "facts" involved, but the weight for those facts is still personal and outside the scope of a debate.
Micheal hendricks have made some good cases for regulation. In his challenge runs he's sometimes handcrafted a few fuel cells, put them in the reactor and then just wanting them to last as long as possible. I can't remember why he couldn't set up automation for it, but I do remember that it made kind of sense. For any normal run however just set up automation for your fuel cells.
Also power isn't the only use for uranium. Most megabases will have trains, that run on uranium, so even if you save 30% processing for the reactors that's maybe 10% of your total processing. The rounding error for having a discrete number of centrifuges and miners might be bigger.
Also you could have this:
https://steamcommunity.com/sharedfiles/filedetails/?id=2971474406 My uranium operation is still tiny, just a small patch near spawn cut by water generating over a big part of it.
If ever do a regulated build it will be to solve, with as elegant a solution, or an as obtuse as possible method (or both) because I've decided to tackle that particular puzzle. It's sure to have nothing to do with resources, UPS, space-saving, or any other in-game considerations.
I've had a 56-core reactor running for slightly over 250 hours, having used 257k of the 278k cells made, along with making 2.1k atomic bombs and 6.9k nuclear fuels. The, so far, one tapped uranium deposit started at 23 M, it's now down to 17 M, Another 700 hours and I'll have to find a new deposit of uranium ore. Lucky for me there's over 2 dozen inside my walls to pick from (not counting the ones I've built something on top of already).
And, just to keep it from being sited as too low, the nuclear fuel is in limited use. Most of the trains are still using solid fuel, with a few still on coal. The nuclear fuel is for my personal shuttles, construction trains and the arty trains, which suck on standby and unused.
No, it is not a failure in any way to consider proportions. Not in any fashion. I've considered proportions, ratios and all the math quite thoroughly. If you've read some of my posts, in this thread alone, you'll notice that I'm quite conscious of the math involved in much of the game. I might ignore it in my builds, for reasons of my own, but that's by choice not ignorance.
Just to jump ahead, since it is also a failure to consider proportions, correctly, on your part:
The U-235 used in making one reactor run for 200 seconds - the only option available - could also power ten locomotives for 200 seconds. And that's 200 seconds of travel time which is on average between half and two-thirds of the total time. At the top end of two-thirds 200 seconds of reactor time would be 3000 seconds of locomotive time. Worse, the making of the nuclear fuel requires about one-quarter of the inserters needed to make, use, and reprocess the fuel cells. Now we're at the point where the processing of fuel cells, relative to nuclear fuel is in the range of 95% of the U-235 usage. Your "rounding error" is slightly more rounded, and in error, than you realize.
And, before you decide to dig deeper, yes, that is on a pure 1:1 comparison, which is unlikely to exist in a megabase still using nuclear power. Unless the base in built by a lover, and abuser of trains, (pointing the finger at myself) the ratio is likely more like 6:1 rather than 60:1, or around 80%-85% of the processing.
I'll give you half-credit here. A growing base is likely to have extra processing, of everything, scattered about. A stable base, however, shouldn't. If a stable base does have such, that's a situation which needs addressing and could have greater impact on performance than the reactor controls. The catch is, unless it's built as a showcase, and then ignored, stable bases, of any size, are a myth. With a growing base the extra processing will become fully utilized, and then become insufficient. Reducing the extra stuff is only making more work to rebuild it in a couple hours. Reducing the performance impact of the nuclear power plant, especially if the system is in-built to the blueprint(s), is a one-time effort which will continue to pay off as long as the base is operational. The proportion you failed to consider here is the operational life of the base versus the operational life of the power plant.
All that aside, I'm still in the group who choose not to regulate my nuclear power plants in such a fashion. If I am in a position where regulation has an impact on performance, which also means that every rounding error is significant as well, I'd rather control the building of reactors to keep the supply near to the demand. A power plant fully utilized needs no regulation and I've saved the build time and processing time associated with the circuits to regulate the fuel usage as well as the resources, for those who count them at that point, of adding all the steam storage, which in extreme cases could amount to 9 iron ore (4 iron plates and 1 steel plate) per reactor. (Hardly a big expense in resources.)
I'm not sure what your screenshot is intending to display, but I'll accept that your uranium operation is small. Mine is not. https://steamcommunity.com/sharedfiles/filedetails/?id=3222172913
Not only is it not small, if you look to the lower section you'll see it has the ability already built in to double production.