|
Bhodi posted:If you think people would use it, you should stick it into a book and add in a bunch of connector recipes as well because making pipes is tedious as hell - I really dig the angled pipe recipes in the one below. If you're looking for more work, did you integrate an integrated sub-setup that will keep the inserters and conbinators powered enough to insert more fuel? The separate power grid you are talking about (usually consisting of a few solar panels and accumulators) is mainly used to stop brownouts from causing a negative feedback loop. It is useful if your moderation scheme is timer based. Your circuit running more slowly means your timer runs more slowly. That means the next set of fuel cells is inserted late, which means further brownouts, which makes your timer run even slower, and so on. A separate power grid can also be useful if the design has a lot of pumps, for the same reason. Pumps not getting enough power chokes the water/steam, which results in even less power, and so on. A moderation scheme based around counting output fuel cells doesn't really need a separate power grid. Sure, the circuit and inserter arm runs a little slower, but the actual power difference will be pretty minimal. You might insert a second late, compared to the timer based setup inserting REALLY late. The timer setup is a little more fragile than counting, given that it requires a separate power grid. It does let you avoid the dry time that purely counting based setups have though. With fast inserters, better counting setups can minimize dry time to one turn of an inserter arm, which is about half a second. Given that a full burn is 200s, that means you are operating for 200s every 200.5s, so your maximum run time will be 99.75%. A timer setup can make the decision to load right before the previous fuel cells come out, resulting in a 100% run time. The other consideration with moderation schemes is syncing. Both timer and counting based approaches can be made to sync the reactors with appropriate circuitry. Syncing with counting is particularly easy if you use something like secondary steel chest storage for spent fuel cells. The chest method adds an extra inserter arm rotation of dry time, but it removes the need for a counter and a latch. It is probably the easiest synced moderation method to understand and implement. If you want to start building your own moderated plant and aren't a circuit expert, that is where I would suggest starting. Bhodi posted:There was also a book I used to have but I deleted it(?) of an expandable setup that I really dug, with a series of blueprints starting with single plant going up to 8 that you could layer and build without having to do a rip-and-replace once you outgrow your current plant. I found that really helpful since you're often really limited on materials when you initially want to switch over to nuclear. https://factorioprints.com/view/-L2Vbi3Ft814dlqBiGXe The one design choice I tend to like on my plants is single-sided hookup. I dislike needing to landfill out the middle of a lake to lay one down, given that robots can't do it. Player time is valuable. I feel like just needing to edge a lake off is much less painful. Filthy Monkey fucked around with this message at 18:43 on Feb 20, 2018 |
#
¿
Feb 20, 2018 18:16
|
|
|
|
| # ¿ May 12, 2024 06:06 |
|
|
So people who have made LED displays have done some pretty amazing work. DaveMcW's blueprint here is particularly impressive for a very efficient numerical display with digits you can chain together. You just paste down as many digits as you need, and they connect themselves into a functioning display. https://forums.factorio.com/viewtopic.php?f=193&t=19825&start=60#p282391 It is dead easy to use. You hook up your signal A to the right-most digit, and the display just works. I thought it was a little hard to see though, so I increased the size by a factor of four. I don't quite understand the magic computation that decides if a particular lamp should be lit, but I can at least take the lamp in each position and make it four lamps. I put the display on a moderated eight reactor plant to display the percentage of steam storage that is currently full. Running below full load, you can see it drop down to 26% where it then loads fuel cells, and watch the number go back up. .  I even had it changing the display color to green when the reactors were loaded and white when on steam, but I found the green just made the display harder to read. The reactor has single-sided hookup. There are more water hookups (14) than necessary (10), as the extra throughput lets the reactor run without any pumps. The moderation scheme counts empty fuel cells modulo 8 to sync the reactors and uses a latch to keep the load state. With no timer circuit and no pumps, it should be pretty resilient to brownouts, and definitely doesn't need any sort of protected electric network. It does produce a full 1.1 GW over a 10 hour torture test, though it likely to be a hair short of the theoretical 1120 MW due to not using a timer. I wouldn't be surprised if the steady state were around 1115 MW. Admittedly, it did just occur to me that rotating the plant is likely to make that display kind of hard to read. It would be easy enough to make a version that is flipped upside down, anyway. https://pastebin.com/5QvxHfAU Filthy Monkey fucked around with this message at 16:13 on Feb 23, 2018 |
|
#
¿
Feb 23, 2018 06:33
|
|
|
I updated that eight reactor with steam display to also check for fuel availability in the requester chests. If there is a chest without fuel, it will throw an alert in the corner and play a non-global audio alarm. I figured the constant global audio alarm might be annoying as the plant is getting built. It also won't load the reactors unless each of the chests have fuel. I added a flipped version with a southern hookup too, which needs a separate blueprint because of the display. https://factorioprints.com/view/-L63IKLwHrj1bn0Lgr8r From a logic perspective this is definitely the fanciest reactor I've built.
|
|
#
¿
Feb 24, 2018 00:05
|
|
|
ionn posted:Yeah, full throughput of one offshore pump is about 11 heat exchangers (almost 12). But isn't one pipe supposed to be able to flow more than that, given enough pumpage (if I read the numbers right)? I guess that is what limits how long an unbroken chain of heat exchangers can be, how much water you can push in at either side. But then again it's easy enough to leave a 1-tile gap for a water side-feed, until you run out of heatpipe capacity instead. Fluid flow is basically black magic, decreasing with distance. I find it is best to just build in some wiggle room into the design, given how hard it is to accurately measure. For example, I made this plant, which rides the line at 11 heat exchangers per offshore hookup. https://factorioprints.com/view/-L5VKZO5Id46w92x5SF9 11 exchangers per offshore means the fluid needs to be moving at nearly full offshore pump speed to the exchangers, which demands a lot of fluid pumps given the distance. Without them, the far exchangers go dry. The plant works at full power, but at the same time I kind of dislike fluid pumps in a power plant design unless they are running off of a separate power grid. On your main grid fluid pumps will pump less if you have a brownout, which results in even less power, which results in less pumping, and so forth. Negative feedback loops aren't good. Safer, I think, is running something more like eight heat exchangers per offshore pump. That gives you a little more room for fluid slowndown due to distance. I made this plant with the eight exchangers to one offshore ratio, and it doesn't need fluid pumps to operate at full power. The extra wiggle room is good enough. The https://factorioprints.com/view/-L63IKLwHrj1bn0Lgr8r The tanks in that design are right next to the heat exchangers too, to try and take advantage of the negative pressure pulling the steam out. Trying to make a single giant steam buffer without fluid pumps will often result in the steam pooling in unfortunate ways. I also don't like plants that require you to landfill out the middle of a lake. It is a pain in the butt, since robots can't do it. Filthy Monkey fucked around with this message at 18:50 on Mar 5, 2018 |
|
#
¿
Mar 5, 2018 18:48
|
|
|
Boilers and steam generators? Nah. I'll leave up steel smelters until after I get nuclear power. Once the nuke plant comes online, then it is time for a refit and expansion of the smelting lines with electric.
|
|
#
¿
Apr 4, 2018 15:58
|
|
|
I would advise against all the mods when you are new. There is PLENTY to learn about in the vanilla game. Really, I am not a big fan of the big mods which add extra complexity. I don't mind the quality of life or feature-specific mods. Mods like resource spawner overhaul, bottleneck, creative mode, upgrade planner, long reach, squeak through, and such are all fine. Even then, I usually don't use much except creative mode for testing. Also, you have no idea what insane is.
|
|
#
¿
Apr 10, 2018 16:26
|
|
|
I agree that going too nuts on net blueprints isn't a good idea. There are some net blueprints I do use. Namely modular railroad tracks and balancers. Actually verifying that an 8 to 8 balancer works isn't something I really want to do. The rest of my blueprints are personally designed. I am probably most proud of my nuclear power plants. https://imgur.com/4hjMZQl https://imgur.com/agmTP7Q
|
|
#
¿
Oct 3, 2018 23:20
|
|
|
katka posted:What does that grid in the 2nd one read? How full the steam tanks are? Vernal posted:Justifiably, the latter one looks splendid! https://factorioprints.com/view/-L63IKLwHrj1bn0Lgr8r
|
|
#
¿
Oct 4, 2018 11:54
|
|
|
I've never done anything special with furnace fuel. Fuel goes a much longer way than the ore itself. A single yellow belt of coal can support 592 steel furnaces, which is 12 smelting columns when using red belts for the ore/plates. https://kirkmcdonald.github.io/calc.html#furnace=steel-furnace&items=iron-plate:f:592 If you start using red belts and solid fuel, you can keep over 3700 steel furnaces running. https://kirkmcdonald.github.io/calc.html#furnace=steel-furnace&fuel=solid-fuel&belt=fast-transport-belt&items=iron-plate:f:3703 By the time fuel throughput would become an issue (which is basically never), you would have long since switched over to electric smelting anyway. A balancer on the ore input is a good idea, particularly once you start pulling it off of trains. I typically don't do anything too fancy with the starting smelting columns either, like trying to leave room for electric smelters in the starting belt layout. This pic isn't mine, but something similar to this is pretty standard for starting out. I usually run the fuel line perpendicular to the ore though, and pull of the same way to feed multiple columns. 48 stone furnaces can max a yellow belt, and 48 steel can max a red.  By the time you are switching to blues and electric, you'll have bots to do all of the teardown and rebuild. Filthy Monkey fucked around with this message at 17:12 on Oct 20, 2018 |
|
#
¿
Oct 20, 2018 16:55
|
|
|
The common way of making a counter is to wire an arithmatic combiner's outback back to its input, so that you hold your value from tick to tick. If I want to count to 40, I could always wire it to an artihmatic combiner set to modulo 40.  The above will count the uranium moved by the inserter modulo 40. It starts at 0. Once 40 have been moved, it loops back to 0.
|
|
#
¿
Oct 26, 2018 04:17
|
|
|
I knew it quickly because I use something similar on my nuclear power plants to sync the reactors. On my eight reactor plant, for instance, I count the empty fuel cells removed modulo 8 via that method. One of the conditions for fuel insertion is that the count is equal to 0. If it is anything else, I know that all eight cells haven't come out yet.
Filthy Monkey fucked around with this message at 05:26 on Oct 26, 2018 |
|
#
¿
Oct 26, 2018 04:45
|
|
|
Finally started to decentralize some of my base in my current game. Now I have separate smelting stations that take in trains of ore, and will output trains full of iron/copper/steel. So one set of trains go from the ore outposts to the smelting outposts, and other trains go from the smelting outposts to the main base. The trains are all 3 locomotive 8 cargo. I pull a belt from each side of each cargo wagon, so 16 belts. Every time something goes into or comes off a train, it goes though a 16 to 16 balancer. I might go and make my own little rail blueprint book for personal use, as all of the ones I've tried so far have little annoyances. I always find it the most fun to make my own stuff in this game anyway.
|
|
#
¿
Apr 21, 2019 02:16
|
|
|
RVWinkle posted:I'm generally terrible with signaling so I have been using these blueprint books and they're pretty good. The spacing is real tight but it overuses big electric poles. The stations are pretty large but it's meant to reduce traffic jams. The only major downside is that there's no 4 way intersections. I searched around for intersections and I found a thread on the Factorio forums with dozens of 4 ways but they were all awful. I just went ahead and made a 'main base' type dropoff station meant to mesh with the pieces in that blueprint book. It uses the book's onramp and offramp pieces.  https://pastebin.com/2DyeehcS Each train output is fed into a 100% throughput 8 to 8. The dimensions of this particular 8 to 8 are actually beneficial here. Filthy Monkey fucked around with this message at 05:33 on Apr 21, 2019 |
|
#
¿
Apr 21, 2019 05:18
|
|
|
I have a question. I pretty much use my own blueprints for everything except balancers, which are arcane wizardry to me. I only ever use balancers for train loading and unloading ever since priority splitters fell from heaven. When do I need to care about whether a balancer is throughput limited or not? Do I need to care at all?
|
|
#
¿
Apr 23, 2019 02:08
|
|
|
Guess I need to scour the net for the best unlimited balancers then. I downloaded the belt balancer tool to verify function and throughput. https://github.com/d4rkc0d3r/FactorioSimulation/releases A lot of the balancers that people post don't actually work right, are output limited, or are unnecessarily large. The standard 4 to 4 balancer everybody has used is indeed full throughput. No reason to use too much else. I find I don't have too much use for 4 to 4 balancers any more after priority splitters. After all, trains are mostly going to output 8 or 16 lanes depending on the size. Looking at 8 to 8 balancers, I found two of interest. One came from here. Not the original post, but RedditNamesAreShort in the comments (who made the balancer analyzer tool as well). https://www.reddit.com/r/factorio/comments/6evxn8/8_to_8_inline_balancer_full_throughput_8x18/ He only posted a picture of his balancer though, and not the actual blueprint. I recreated it from the picture, and tested it in the tool. It does seem to be the smallest full throughput 8x8 I've been able to find. https://pastebin.com/TU0ayWBd code:https://www.reddit.com/r/factorio/comments/a65wwz/88_throughput_unlimited_2_gap_balancer_10x15_tiles/ It also tests as being full throughput. code:https://gist.github.com/Bilka2/aeec4ff7123ff5544cb9a80cf1046a06 code:https://factorioprints.com/view/-KtY4yWQFzZb06O6oIi6 code:Filthy Monkey fucked around with this message at 03:48 on Apr 23, 2019 |
|
#
¿
Apr 23, 2019 03:46
|
|
|
The 8 and 16 balancers seem like the most useful, as most trains are either 4 or 8 wagons, pulling two blue belts off each wagon. You can very easily pull three blue belts off a wagon, evenly from all the chests. The problem is balancing it afterward. With 4 wagons that means you have 12 belts, and with 8 wagons it means 24 belts. In either case. you don't have a power or 2. I think you could just loop back four belts in a full throughput 16 to 16 to make a full throughput 12 to 12. You actually can pull four blue belts off of each wagon with some splitter tricksery. That makes balancing no problem. All of the solutions I've seen will pull from the chests unevenly though, which isn't particularly desiriable. See the following. https://www.reddit.com/r/factorio/comments/avfk21/belt_loading_tricks_with_the_new_017_belt_speed/ https://www.reddit.com/r/factorio/comments/ayj5i5/more_on_belt_unloaders_in_017/ quote:The "splitter trick" is arranging a stack inserter to drop items onto a splitter from the side (not the back or front), and then combing the two outputs from the splitter (either all on the same lane, or on both lanes). Image example. It allows the inserter to drop items onto the belt more quickly, because it's effectively dropping items onto both splitter inputs. Quantitatively, the period of an inserter dropping onto a plain belt is 52 ticks, but dropping onto the prioritized side of a splitter is only 37 ticks. (Non-prioritized splitters used to be 40 when I wrote the OP, but I'm seeing 38 in my timing sandbox map right now, so... ¯\(ツ)/¯) The splitter trick allows filling a belt with only 3 inserters. 
Filthy Monkey fucked around with this message at 05:02 on Apr 23, 2019 |
|
#
¿
Apr 23, 2019 04:43
|
|
|
Went back to loving with 3-8 trains. They have a good bit more acceleration and braking power than a 1-4, and carrying more stuff is always good. Designing stations, I've found that they are best off with one locomotive in front and two in back. That lets the back two locomotives hang out on a curve, saving a little space. My current gigastation for train delivery looks like this. Bays for four 3/8 trains. Each train gets 16 belts pulled off, which is put through one of the unlimited throughput 16 to 16 balancers I mentioned on the previous page. I found the gap design was a better fit than the inline. Total this is 64 belts of fun.  I actually need to stretch the length vertically by one more track. The very tip of the back car is hitting a train signal I don't want it to, even though the trains actually fit. Filthy Monkey fucked around with this message at 22:59 on Apr 23, 2019 |
|
#
¿
Apr 23, 2019 22:56
|
|
|
The Locator posted:What is the scenario where a nuclear plant would be shut down by a cascading power failure once fuel for the plants has been produced in significant quantity (and 1 level 3 assembler without beacons or anything can build up a huge stockpile of fuel in pretty short order)? Protected electric networks are generally used in nuclear plants that have the potential for negative feedback loops from insufficient satisfaction. There are two main circumstances that can cause that. 1) Pumps. Unpowered pumps act like valves. The less power it gets, the less flow it gets. If you have reduced satisfaction, your pump moves liquid more slowly, which can result in even less power, which means even slower pumping, so on and so forth. 2) Timer circuits. If you are using a timer circuit to insert fuel, then the circuit running more slowly means fuel gets inserted late, which means even less less power, which means your circuit runs even more slowly, and so on. I still mostly just use this design that I made over a year ago. No pumps, and no timer. It is overkill on the steam tanks, but I wanted it to be able to completely store a fuel cycle as steam even if it was disconnected from any power consumers at the time of fuel loading. I know it could have fewer tanks and store some of it as heat, but I can't read heat with circuits, and I wanted to have fun with the lamp display. https://factorioprints.com/view/-L63IKLwHrj1bn0Lgr8r I know that the whole steam storage thing isn't a huge deal from a resource perspective, as keeping enough fuel around is trivial. Half of the fun of factorio is overengineering solutions though, and steam is fun to play with. Filthy Monkey fucked around with this message at 16:57 on May 4, 2019 |
|
#
¿
May 4, 2019 16:50
|
|
|
Percentage of of steam tank capacity used. The reactors load when the steam capacity is down to around 26%. If you are under full load you can watch it load, watch the steam storage percentage go up as it burns the fuel cell, watch the percentage go down as it runs on the stored steam, and then see it reload the reactors midway through 26% again.
|
|
#
¿
May 4, 2019 21:10
|
|
|
|
| # ¿ May 12, 2024 06:06 |
|
|
I noticed nuclear plant numbers are way different in krastorio 2. A single reactor's heat output is 250 MW, compared to 40 MW in the base game. Heat exchangers are also nearly doubly as powerful, but also suck down double the fluid. All of the ratios are thrown off, and a a plant design from the base game isn't going to come anywhere close to operating effectively. A two reactor plant can generate a maximum of 626 MW (with 62.5 turbines), which is more than a four reactor plant could generate in the base game. A four reactor plant can generate 1.5 GW (with 150 turbines). I went ahead and rejiggered one of my old two reactor designs to Krastorio 2 standards. It does have steam storage, and will operate losslessly no matter what your power draw is. Though a two reactor design has a maximum potential output of 626 MW, this one can dip down to about 618 MW due to the mechanism I used to reload. Not a huge deal, all things considered. One of these will probably be good enough to get people through the mid-game and into the higher tech power options. https://factorioprints.com/view/-M4DR_tmCVxzIIxQqvLG
|
|
#
¿
Apr 6, 2020 08:50
|
|