This video was discussed on Reddit, including detailed responses from Sytharin and JinkyRain: www.reddit.com/r/SatisfactoryGame/comments/1dgs4gg/fluid_byproducts_i_tested_the_vip_junction/
@@johnyoung5392 Sinking Water in a Coal Generator requires Coal, which is a somewhat valuable resource - sure, you get Power, but once you've moved beyond Coal Generators for Power, that Coal might be more useful for something else. (Stealing this recommendation from the Reddit thread ...) Using Wet Concrete to sink Water will produce Concrete and is a highly-efficient use of Limestone (and overflow Concrete can be sent via Smart Splitter to an Awesome Sink, so there is no danger of a backup.)
@@tristen9736Many of the greatest minds(especially theorists and mathematicians) in history were German, and Factory games are science-y and mostly math-ey, so it kind of just fits
"FICSIT Does Not Waste", after all. Also my favorite production loop is the one that makes aluminiminiminium without involving silica, recycling its own water and only requiring half a pump of fresh input... but due to me not using priority pipelines, you must sink any excess ingots if storage fills, because if the smelters back up into the refineries, _everything will permanently and irreversibly clog._
for a channel with just 325 subscriber, this is incredible production value. Well done! The point was brought across expertly, with some entertainment along the way.
Amazing content and presentation in a brief format explaining the dark art of fluid managment. I think the only joke he missed was saying, "..buffer equilibrium appears to occur at approximately 1803.70 cubic metres..." I'm going to have to subscribe.
I love your amazing advice here "Get to the fun stuff faster" you're not concerned how people play the game, you're showing us different ways to play the game and that's really cool
Very cool video. I've been playing quite a while and my solution to output water is...limestone. Make it into concrete and either use it or sink it. I really like the Headlift Reset idea and I will be using it going forward. Thanks for this!
@@XyrillPlays I really liked your video dude, I watched the whole thing. I just wanted to mention, couldn't you use a smaller liquid buffer? Literally just by lifting it up 2-4 meters, making it as tall or taller than the industrial sized buffer. Or is there something I am missing, like the larger buffer making the pressure rise more/quicker?
just as a note: the ficsit plumbing manual is a bit outdated. Update 7 fixcet alot of the pipe problems, but in update 8 the problems were back again. lets wait till 1.0 to see what the pipes will do then. Great video thow, i liked watching it and loved your solutions.
@@Pikawarps No really my experience. I now have a setup where I let output water flow in from two sides, with a large buffer on one side and an unpowered pump of the inflow pipe for the fresh water. That seems to have stabilized it due to the fact that at a certain point the pressure from the buffer an the inflow stabilize enough to not cause overfill.
I don't even play or watch Satisfactory videos. Somehow, this came on my RUclips recommended, and your video editing for a channel with almost 900 subs is incredibly impressive. Great work, sir. I wish you luck on your endeavors.
You should get the game if you enjoy this kind of thing, it may be slightly expensive at 30$ and having a price hike to 40$ soon but its worth it and a great game that has sucked up way too much of my life.
For the fluid buffer height requirement, iirc it's one of the unintuitive things. You could have a vertical pipe the same height as the fluid buffer and it would serve the same purpose. The fluid buffer is more convenient to place instead of creating a vertical pipe and looks nicer. It's kind of like how water towers just need a vertical pipe. For the priority system, the priority pipe is counter intuitively supposed to be the upper pipe. Liquid flowing downward seems to have a small priority over sideways. It also allows the waste water to back flow into the fresh water pipe, which will eventually cause the water pumps to shut off. In this configuration, it helps to put a safety valve/pump on the downward pipe so that fresh water can't flow into the waste water line. Finally, if the waste water and machines are all full, the system is completely clogged and requires at least the waste water line to be flushed. Just a hunch, but flushing the waste water line after the missing inputs were restored might have gotten the system working again. Overall, I love the video. You made an amazing demonstration of multiple solutions and have great comedy throughout the video!
A better solution is to put a valve on the wastewater pipe preventing backflow. In the event of an input underflow that should result in the wastewater pipe emptying at least partially, which should prevent the jam.
Amazing post! Thank you. I think the most valuable thing here was the whiteboard. Ahem, I mean the test setup, where you identified what would cause stalls and then verified them. I had no idea that input starvation would behave differently than output clogging. Neat. FYI: I invariably use the boring method since the pipes behave so differently from reality and they're so unfriendly in terms of debugging what's happening.
The tests are shown in the same order in which I did them, and I was certainly very happy that the Missing Input test was decisive in some way. A conclusion of "they all work" would have been quite boring (and not in the good way).
@@XyrillPlays Yeah it would have been boring in the "How the heck am I supposed to make this into a video" way. There are 2 other methods you didn't test tho, Gravity Priority (have the refineries high up and let the byproduct water fall down into a junction that is also fed from below by the water extractors, using pumps to raise the water to the input of the refineries), and "Break the loop" (use the Wet Concrete recipe to dispose of the byproduct water instead of recycling it, which gives you the option of either sinking or recycling the... concrete, which is a lot easier than doing the same with the Water).
Small YT channel producing quality content? I love to see it. Ive never played or ever wanted to play Satisfactory ever. But quality video and made me interested.
I'm surprised the simplest and most reliable option wasn't a contender here: gravity. Just have one water extractor feed into a junction from below, feed to refineries from the sides via pumps that have enough headlift and let the recycled water fall in from the top. As long as the refineries are high enough above the junction, the virgin water source can't fill up the pipe coming down from top, but the inputs are kept full. The system never gets deadlocked and has instant recovery when the solids are moving again.
Great idea! I've not dealt with waste water yet but the theory seems sound and exactly how I would try to approach this problem. I hate finicky unstable 'solutions' and build for endurance, adaptability and self-managed recovery if possible.
For the VIP I think the issue is that when you take away the bauxite,fresh water still being pumped in. There is no height buffer limiting how much the pumps can fill. There is no valve limiting fresh water intake either so the fresh water pump is pumping a full 600m^3 a minute as opposed to 240m^3 a minute. Had you waited longer before restarting, I think the merged valve setup might have locked up too. Also one of my best waste water setups was in a nuclear plant. The waste water was 240m^3 per minute and the sulfuric acid required 240m^3 per minute. So once I had filled the pipes and buffer I could disconnect the pump entirely. That was in Update 7 though. I'm not sure how much the recipes changed in update 8.
This is exactly the problem, and you can solve it by putting a valve on the byproduct pipe. This prevents fresh water from filling that pipe and avoids a jam when the input underflows.
Deutsche Qualität !!! Thanks man, finally someone who dug deep into that infamous topic !! you will def help many people. Thanks again and really good job man
So I went to apply mergless because I've been struggling with this for a while in my alumina factory and I just wanted to be done - but I discovered something because my factory's larger (twice as large in fact!) 9:44 - If you double the size of this factory you can use the output to feed 1/3rd of the input. You'll need 480 water in to 4, but the combined output is exactly what the other 2 need. So no need to build "an extra" building (assuming the double doesn't count as extra)
Hello. I haven't play Satisfactory for 2 years now but I find this video fascinating :) Got a question: Could we use just a 12m high vertical pipe instead of Industrial Fluid Buffer in the Head Lift Reset solution 7:20 ? Or normal buffer on 4m high elevation. (I can't test it myself cause I'm waiting for v1.0)
That was my immediate thought, the regular Fluid Buffer should work just fine as long as you give it a base that lifts it up a few meters. I don't think just a vertical pipe would be enough, since pipe capacity is so limited, making it a very weak buffer. Best case scenario: it works but slowly.
Hah, when I showed this to some friends as a test audience, they had the same question immediately. Given that fluid buffers are just bulky pipes, I think the answer is "yes", with the caveat that @Denkkar also gave, but I have not explicitly tested it.
Vertical pipe works just fine, I tested it a while back when pondering about this idea. It makes sense - its only purpose is to raise "pressure" in the system which depends ONLY on the water column height. Pipe heght only needs to be higher than lower priority input headlift, plus some room for fluctuations (I think), but it can be as tall as you want just to be safe. Mine was pretty high, like 50m above the junction. And whats pretty cool about this design is its entirely realistic. You can build a scale model with some check valves and bottles of water at different heights, and if you build it right to accout for bernoulli's principle, it will work - one bottle will empty before the other will even begin to drain. Because of this, this junction design might work even with no extra vertical pipes/buffers at all, as long as theres a difference in input "pressure" - although I haven't tested it and am not sure if the game will model this correctly.
That's how our whole factory is currently build -- but I will now switch to the fluid buffer instead, that looks really nice and actually serves as a small input buffer.
I do with the intent to add pillars later, and sometimes I'll put the pillars in before I start building on the factory, but yes, this does make it easier. For me it's mostly cantilevered overhangs off a cliff, sometimes with pillars added later.
Nope. I can’t wrap my head around floating objects that defy gravity. Even my rail must have 2m blocks as support and bridges contain girder and/or pillars
I used to put support on my platforms, but building takes so long even with blueprints that I just couldn't be bothered. No one is gonna look at my stuff anyways. I stopped using walls, even.
Great video! As a long time Satisfactory player > 3000 hrs, pipe mechanics have been a bit of an enigma and I've always tried to keep it simple. The optimized fluid buffer approach that you demonstrated is urging me to be more creative with the pipe mechanics.
@Xyrill Plays - 14:16 For the VIP junction, I don’t know about anybody else personally, but when I use a junction on a pipe, I delete the excess pipe at the dead end to avoid sloshing (which becomes possible since adding a junction splits the pipe into two separate entities).
The VIP Junction basically works by blocking the upper pipes as long as the lower pipe kept flowing. The trouble starts once the byproduct water production was disrupted and water from upper pipe clogs the output, essentially turning the intended function on its head.
I don't think this is the case I have a vip junction for my baux setup and the input pipes are always 100 full while giving priority to draining the output pipes into the input ones. It always seems to give priority to the lower pipe which is the redirected output pipes.
The actual reason it jams is that the fresh water fills the byproduct pipe itself, which prevents the factory from putting new water into that pipe. You can prevent the jam with a valve controlling the flow direction on the byproduct pipe, which prevents the fresh water from filling it.
After numerous failures I've adopted a sledgehammer approach. I package everything then unpackage it directly into the machine. This is the only way I've ever found to make machine runs that never fail or lose efficiency. It costs power and machines but it saves my sanity and for anything that MUST NEVER FAIL its really quite effective. My understanding is that under the right situation all of these except for the split method will eventually fail under the right circumstances.
Water is pretty much the only fluid that is used as both ingredient and byproduct (aside from sulfuric acid for batteries, but to solve that problem we have the "Classic Battery" recipe still, I think?). Anyways, there's a neat and handy way to solve this without all the packagers. Stop Recycling, make more concrete, use more limestone, sink the excess concrete. The "Wet Concrete" alt recipe is what you need to enable this. Byproduct water stays in it's own pipe, never interacting with input water. Instead of recycling the water by ANY method, we "use it for something else instead", namely to make Wet Concrete. A handy feature of the Wet Concrete recipe is that it is VERY water hungry, and does NOT produce a liquid byproduct. If you want to recycle the concrete, you can, but it's not mandatory. If you do, it's as simple as a priority splitter set to "Overflow" one way, with that way leading to an AWESOME sink. And just like that, the water's GONE! Isn't that nice?
@@collinkaufman2316 Maybe that's what it's designed for, but I find it to also be the best concrete recipe anyways, all you need is water and limestone, meaning that if you move the limestone TO the water source, you don't need to transport huge volumes of water. 600 m3/min of water (1 mk2 pipe) runs 6 refineries making wet concrete, which produces A LOT of concrete, either 600 or 780 a minute (I forget if it's 100 or 120 concrete/min output per refinery, but I know it's one of those), meaning that unlike the normal concrete recipe in a constructor, you get MORE concrete out than you put in limestone. There aren't many other recipes that come close to that, and the ones that do require a rarer resource than concrete or water to make work (like quartz in the form of silica, for one of them). Point is, for the way I play, Wet Concrete is the best way to get rid of water AND the best way to acquire concrete.
"My understanding is that under the right situation all of these except for the split method will eventually fail under the right circumstances." That's largely my thoughts as well, with the exception people mentioned that just making wet concrete and sinking it is another possibility. I haven't tried the industrial storage buffer method, but I used to use the valve method when I first started playing. I say "used to" because now I use only the split method or the wet-concrete-sink method - I don't know how or why, but the valve method will eventually break and lock up. Maybe it takes a lot longer to occur than normal, maybe it takes reloading games involved, but it definitely failed on me multiple times, on different playthroughs. The packager method sounds interesting for nuclear processing, simply because I try to actually limit the amount of radioactive garbage floating about which means load-balancing the buildings - Something which is not really possible with the split method (It will still 'work', but loads in way more radioactive material into buildings than it should). At least with packagers, I could use a priority merger (stack both lines on top of each other, put splitters all over the 'priority' line, and connect both sides to mergers on the 'input' line with lifts - gets close to 99.9% of just using the priority line at 6 splitters), and sinking the 0.1% rounding error coming out of the end of the priority line
I love that this game can be played simply, or it can be played as damn complicated as you want, potentially giving yourself a seemingly endless array of complex but still solvable engineering problems depending on what specific limitations you put on yourself, whether on accident or on purpose. Yet solving those problems is where a huge amount of the enjoyment lies. Really genius game design
As a fellow min-maxing german: I have never been so proud. (Also this channel actually deserves more attention, let's hope the YT algorithm does its thing)
It absolutely does. Both with this and the previous video, the first 100 or so views were majorly from social media (Reddit, Discord, etc.), and then the algorithm got attuned to it and started recommending it on people's homepages.
@@XyrillPlays Data point checking in - I have been watching Satisfactory demonstrations and abominations for a long time now, but only just been recommended your channel. Keep up the (Satisfactory) work, pioneer!
I didn't test the Head Lift Reset method in U8, but I did test it in 1.0. I don't know if anything changed between the two versions but for 1.0 there's a VITAL component/detail missing from the description in this video that I discovered after much experimentation and reading some comments on the Reddit post referring to this video. Backstory: I never liked pipes that got in the way of me walking around a platform, so tend to design with the pipes UNDER the refineries. So first I tested the head-lift reset method with my usual standard: pipes down into pipe holes, and under my refineries' foundations, and back up again. The buffer and unpowered pump's input and output were all at the same level as the refinery's input, and I thought this would be enough (since this is where the head-lift on the fresh water input would be 0) to work. It took a couple hours to lock up, it but it locked up. Thinking that the byproduct pipes going down under things and coming back up again might be the problem, I then tested the pipes being lifted about 4m up above the foundations the refineries were on (so just enough room to walk under the pipes) where possible, but still returning back down to the 0-height of the unpowered pumps, buffer, and refinery input/output where they needed to return to the buffer tank and the inputs for the refineries. It also locked up in an hour or two. I then clipped pipes through belts and such just to get everything PERFECTLY HORIZONTAL, resenting the fact that I had to try it, because I hate pipes that block my ground level views. It didn't lock up. Then, based on some conversation on Reddit I looked at for 10 seconds... I simply added about eight junctions to the byproduct pipe, changing nothing else, which divided the byproduct pipe into multiple shorter segments. The whole system locked up in about THREE MINUTES. Not an hour or two, *three* *minutes*. It was the fastest it had ever locked up, by far. It turns out that if you want to lower the priority of a source, simply make it move through more segments of pipes. So I tried that: I removed my earlier junctions returning the byproduct pipe to very long unbroken pipes, and then I added about seven junctions to the fresh water pipe prior to it reaching the unpowered pump... ...and then deleted the buffer tank entirely. It's been running for about ten minutes without so much as a hitch with NO BUFFER. There should be an excess of 60m³/s in the pipes, but nothing is backing up, even without the buffer. Time will tell if it ever locks up, but signs seem to suggest it won't. I'll probably come back and reply to this comment with more updates if anything changes or if further testing shows anything else interesting.
Follow-up: It's been running for hours with no issues. No buffer, just extra segmentation in the fresh water input. I've removed the extra segments to see if that results in any change. Buffer is still gone.
After removing the buffer, the line has continued to work without issues. I rechecked my math, and I should still be seeing an excess of 60 m³/s in my system, and yet the byproduct from refineries is draining, and the water extractors are backing up and shutting off. So I counted pipe segments. Basically, I used the deconstruction tool and highlighted/counted each length of pipe over what I could find would be the longest (in count) distance water would need to travel from a water extractor to a refinery, and refinery byproduct to refinery. If a pipe was broken up by a pump, floor hole, or junction, that would be two pipes. They each highlight separately. There are 13 segments between my water extractor and the furthest refinery producing alumina solution. There are 11 between byproduct refineries making scrap and the refinery making alumina solution. So I added more segments on the byproduct line, and things STILL didn't back up. This wasn't making much sense, so I looked closely and realized that while everything else in my pipe network was a Mk2 pipe, the pipes from my water extractors to the main fresh water Mk2 pipe was a Mk1 pipe. I have a theory that, despite 120 m³/s flowing through a pipe rated for 300 m³/s being nothing unusual, that somehow this is acting as some sort of valve or pressure limiter. I've replaced those segments with Mk2 pipes to see what will happen.
@@Moleculor Flow rate is based on how full a pipe is. So a Mk.1 pipe that is half full can only flow at 150m3/s max. However, a half full Mk.2 pipe can flow at 300m3/s. The other thing to take note of is that pipe segments have volumes. You can watch each segment to see the volume rise and fall as fluids pass through. While dropping a pipe through the floor doesn't affect headlift, it does increase the volume of that pipe and the rate at which it will fill and empty due flow being related to volume.
One thing I will note about the VIP is that it needs to be built as close to the picture as possible. The failure scenario comes about when the volume of pipe between the pump and the junctions are different.
This was a great video, both the content and the editing. You are a good speaker with a sense of humor so I found it very entertaining. I also love the rail design idea, going up when empty and down when full. Smart, but I don't know if that makes a difference mechanically in the game. Given the 1.0 announcement I wonder if any of these will change, or if they would ever consider adding a fluid intake to the AWESOME Sink?
It's been a while since I did the Aluminium line in Satisfactory. I think the way I got around the water waste was based on the headlift limit. My plant was up-hill compared to the water source, so the headlift of the water pump could only fill a water storage half way. With the other open pipe connector of the water storage on the same level as the chemical plants, I just fed the output back to the input, with the storage also on the input. It just worked. I guess this is like your headlift reset example, but this was back in Update 3, before pumps existed, so the fluid buffer is doing the same job as your unpowered pump.
This is really well done! I love your humor style and excellent thorough explanations. Fluid byproduct management has been my biggest struggle in satisfactory and I think I can finally get my head wrapped around how to handle it. Thank you for all the hard work you put into this video!
7:00 Would the smaller fluid buffer work if it were elevated? You can build it right above the cross-connectors, elevating it 4+ meters, and get a more compact footprint. It seems to work in my testing so far, but I don't yet know if it'll fail eventually.
I love your humor, and really interesting content too. Thanks for making this, it should help me a lot soon, now that I'm working my way through oil processing!
Seriously the best guide ever, especially for calling the control for what it is: so easy it's boring... baffling why this isn't the go-to for everyone except for those who seek the extra challenge, but 🤷♀️
I don't know if something changed in 1.0 but I can't get the headlift reset method to work at all. It works initially, then after a short time (~15minutes maybe) the buffer fills to a point and then the system gets stuck. I've tried configuring it almost exactly like it is in the video (basically the same minus the fact that mine isn't on a platform up in the air but at ground level on the water), as well as ~4 different variations. I wanted to go with it, because it seemed intuitive and reliable, though I just can't figure out how to make it work right. But the VIP junction worked fine right out of the box. I've been thinking about how that junction must work, and the only thing I can come up with is that the game isn't a 100% accurate fluid simulation, and instead it's calculating the lower pipes first and then filling them with the upper pipes, perhaps as just some byproduct of how the simulation is written. Or as someone else put in the comments, it's more like a bottle than it is an actual pressurized fluid system. Perhaps from a technical standpoint the game is assigning a height value to the pipes (which it would have to do for gravity) and then calculating lower pipes on the system and then working it's way up. This would cause it to calculate the byproduct pipes first, filling them as much as possible, and then calculate the upper ones and filling the remainder. It would be interesting to see the actual programming behind the fluid simulation to get an idea of how it actually works.
@XyrillPlays, I did not check that linked reddit stuff, however might some useful info. I also struggled with the fluid by-products but got it somehow managed as most of the time you can somehow even the numbers between "by-product" and "fresh" line and then you are fine. I have all 5 phases completed but did not yet run some uranium power setup, so I decided to setup some uranium power. For sure good prepared, but this also means that I had a lot of "fresh" (much too much), coming from above with huge headlift. Means for my encased uranium cell production, my "fresh" sulfuric acid coming from above was huge in comparison to the "by-product" I re-sent to this blender (as I always start with a "small" scalable test setup) in production - so really huge in comparison to the 600/min possible with a MK2 pipeline. So, I thought that this #2 headlift-reset starting at 6:33 in your video sounds really great, but this unfortunatelly did not work for me. I think because when I see your setup in your vid, your "fresh" is coming from below, so I assume that your "by-product" has more headlift as your "fresh". However, headlift reset is what I need, so thx for that info with that unpowered pump - did not yet know this. Whatever, my general setup for sure would start with this #2 headlift reset. As a next step to fix my problem, I added valves to ensure the flow on specific locations. This helped only temporarily, as soon as the industrial fluid buffer was full and the "by-product" pipes were also full, "by-product" began to stuck outflowing. So how to solve? (when "by-product" doesn't have more headlift than "fresh" has) Add a "powered" pump for the "by-product" at 6 o'clock (6:33). This will lead to that a headlift is applied to the "by-product" flow. Then "by-product" has the highest headlift and the fluid buffer will always be filled by the "by-product" because of the headlift-reset of the "fresh" fluid. assumption: "work-preasure" seems to be some kind of related to the max (or current?) flow throughput between 2-X flow lines when the flows do hit in the junction. E.g.: crossing 2 flow lines.. #1: 20 in , #2 10 in , #3 15 out - may result in that #1 10 and #2 5 into #3 -- only assumption) I think this mainly should be taken in consideration in case flow lines do have the same headlift - at least in regards to the target. If you don't want that flow line 1 is influenced by flow line 2 here (as flow lines can influence each other), place valves (where a pump - powered or unpowered - does probably fulfill same function as a valve for that "work-preasure" case)
for 1.0: I have no fancy setups and my plant works completely fine (at 100%), I have just a simple loop from waste water back to production + extra water from extractors. the only noticeable thing about my setup is that my water extractors sits lower than where the excess water is produced. So I suspect higher remaining head lift at a junction has priority for filling the pipeline. my setup in detail: 2 water extractors at 300/min few meters below main pipeline, and capable of fully saturating the whole pipeline, and alumina solution production. Main pipeline... 2 refineries producing 200/min of alumina solution each... into 1 refinery consuming 400/min alumina solution, with simple flat pipeline for waste water recycling. Waste water connects to the same junction as water extractors.
Really high quality material, for video to capture interst of a person wich was never interested in satisfatory or hydrodynamics (me) truly speaks about author skill. Thanks for your work.
13:55 My guess would be that it can manage to go back to production when output clogged because at an output clogg the pressure is "released" from the back of the production chain when output is not clogged anymore. So when input is halted it wants to start up again the second refinery isn't the first one to run, but the first one is but it can't output because the loop is already fully pressurised, and the input from the fresh water have catched up and is also completely pressurised. I'm pretty sure that you could merge the VIP and HLR methods so that the head pressure of the output recycled loop is higher than the feeding from the freshwater pump. But then you might aswell use the HLR method.
VIP worked great for me in release 6-8, thought I would try the HLR after this video in 1.0 and it failed miserably. Back to VIP and it corrected itself and has been good ever since.
14:04 i think the issue is the distance to the joining section. since the bottom pipe has that extra distance added instead of just going straight to the join. couldn't explain why but thats the only think i could think of as an issue.
7:22 cant you just build the small tank a little bit higher up to so it doesnt fill ? in that case cant you just build a large pipe upwarts ? or is the volume needed for the tank ?
what I recommend is that you make yourself a hybrid design, having one or two refineries running backflow, and rest going though pump to fresh water. if you also priority input of ores for that machine, it behaves best in test where you have less input than ideal. 3/4 of those designs would fail if you would have partial input. expecially if you just run normal spliters. fluid buffer version & headlift reset can be made with just a vertical pipe, or smaller tank if you raise it 4/6m. MIGHT be easier to hide. Issue with just vertical pipe, is that you might have low throuput, slowing down recovery at times. but hey, multiple vertical pipes do same as just tank.
Not on fluid management, but I have one video that is in the research stage, and I have one very vague idea. But I won't be making videos just to fill a schedule. There will be a video whenever I have something to say that I find worthy of the audience's time.
Awesome video! Entertaining and very informative! you speak well and the information was displayed and organized in an easy to understand way. Hope to see more like this from you!
Hey man, as a Gold Coffee cup owner I've come back to the game to do it again for 1.0. Been watching some videos to jog my memory. i really enjoyed your video.
Interesting, I usually go with the valve, easy, clean and it works just fine. Just one nitpick pipes don't have pressure, not at all. What Darren Plays explained this pretty well, we need to start think those pipes as bottles and that is why the VIP is working that way. The bottom one is a bottle that will receive water from top when "there is space for it". If the bottom bottle is full => the top will not provide any water, if the bottom have space the top bottle will star adding water to the bottom one. Easy as that. On why this stops is kinda tricky, but I suspect this is a deadlock. You can't consume, so you can't add more water and the water itself is just clogged. If you remove the any water from either input or output it will probably restart and get back to 100%. Additionally I suspect the HLR will suffer the same problem, just wait until the buffer is full and you will have a deadlock as well. You can check that as well by looking at the buffer before the input removal and after, then wait a while... I suspect the buffer will be nearly the same amount and will never drop to zero again. Is just a delayed deadlock probably. Hope this helps!
@@bencoad8492 You know what uses more water than a coal gen (and therefore takes up less space), uses Limestone instead of coal (which is much more abundant) and produces Concrete (a lot easier to recycle into something useful than the pesky water)? Refineries. Not the aluminum refineries tho, those don't use Limestone. Refineries using the Wet Concrete alt recipe!. Each one uses a whole 100 m3/min Water, and 60 Limestone/min, to output a whopping 120 Concrete/min! That's a LOT of concrete! Sure concrete's not worth much, but you can recycle it a lot easier into the feeds supplying your Encased Industrial Beam and Heavy Modular Frame production, or just put it into an AWESOME Sink if that's too much trouble. Concrete is worth like 3 points per item I think? Sure, not a lot, but when you've got T5 belts running at 780/min, that's 2340 points per minute per full T5 belt of concrete you're sinking, and you only need half that much Limestone (and 600 m3/min water, or a full Mk2 pipe) to keep that rate up infinitely. Considering that 2 T5 belts of Aluminum Ingots output takes 3 T5 belts of Bauxite Input, and that you output less than 2 Mk2 pipes of Water (as a byproduct) for that rate of production, just shoving the byproduct water into the wet concrete recipe makes a lot of sense to me. Sure it takes power, but coal power plants don't really give you a whole lot of power for your water investment anyways, they're your "starter fully automated" power production after all, even when overclocked.
I used the HLR design on a huge aluminum factory I made a few years ago, and back then the instructions showed an upside down U shape in the pipe just before the buffer, but the thing is it failed horribly for me many times. I had to go empty the buffers around the whole factory every few hours or it would jam up. I'll have to go back to that save file and see if I can fix it sometime
There's a simpler solution to the water output. *Use the water in another process* Since You are alreary making complex teir C,D, or higher items by the time You need to solve this output- chances are that another item in a chain would require water. In Your case, aluminum outputs water at some point. Which means You'll likely need copper within a recipe, or a copper substitute. You can make Copper ingots with water in a Refinery and get higher yeilds. This helps during the Aluminum Sheets chain. I think there is a water and coal in a refinery alt recipe, but I've never rolled it in RNG. Or use iron ingots and water in a Refinery, along with The Iron Wire Recipe. Which can become useful in the Turbo Motor Chain. In the Nuclear and Plutonium Chain, You'll have several ways to use the Output water along the way. Especially if Your build is vertical not horizontal. Sending the water down a few floors to the simpler items chain, then belting the ingredients up, eliminates the potential sloshy in the pipe
Really enjoyed this video! I was curious if you'd be willing to explain your strategy for train logistics in more detail. Im working on preparing my strategy for my 1.0 save, looked at your save and was curious how you divide up resources collected/produced in large quantities in one location out to several receiving factories. It's something ive never been able to figure out how best to do in my head, and would really appreciate your insight. Thanks!
I am not too bothered about learning the intracicies of how fluid or fluid machines work in Satisfactory (As long as they work without fail!) but this video was really refershing and wonderful. A great properly scientific test, with great results!
I was unable to get the HLR method to behave as shown in 1.0. When only the supply water through the unpowered pump is connected, it correctly only fills the buffer to 10% or so. However, when the byproduct water is hooked up also, the buffer over the course of two hours fills to 95% or so (probably the HL limit of the byproduct-prodcing building), then jams the whole system up. Looking at the pipes, it seems to behave as though the HL from the byproduct building now also applies to the water coming from the unpowered pump, and the flows compete with each other. The level in the buffer swings wildly, but there's a small net increase across many oscillations.
Followup: I solved this by implementing a suggestion by someone earlier in the comments. I eliminate the dead pump and buffer entirely, and merge my supply water into the byproduct pipe, with the byproduct buildings between the fresh water and the consumption buildings. It works shockingly well. Fast startup and recovery, no rate limiting, and no balancing. In fact, have the extractor give more water than the system needs, and it Just Works. Byproduct water builds up inside the buildings' internal buffers, but it only hangs around for a few seconds before downstream consumption makes room for it, and it fully drains into the line, with supply water filling any remaining capacity. It happens so quickly that I never even see a void appear in the pipes. They stay full at all times, and just have a brief surge in flow.
Awesome analysis! I'll definitely be revisiting this when I replay the game. I found the VIP junctions fine enough and small when I played in U6 but looks like I should revisit. Valve is very tempting!
I got to Aluminum just last week. I'll write this before I watch the whiteboard part. The system I built outputed enough water to cover 1/3 of the total needed at the start of the production line. So I cycled it back and put a Valve on both the fresh water and the return water. I set the max flow rate on the fresh water valve to 2/3 of the total needed. Now Im gonna continue watching
I've tried the JustAValve method before looking anywhere else for help, as I assumed that it would solve the problem. For some reason it didn't, so I tried the VIP method...which also didn't work for me. Perhaps I already had too much pump pressure in my main line (I needed a pump somewhat close to my refineries in my layout). I ended up making a very simple junction which has the byproduct (lower flow rate) water coming in on the Refinery level, and the main line coming in (at the same height) to a 4-way that faces up. So the main line bends up and then straight down into the 4-way. That hasn't failed me yet, and doesn't require the area or materials of the VIP setup.
A German explains efficiency in Satisfactory. Sometimes the jokes write themselves. Joking. Thanks. Nice to know my “just a valve” method is legitimate.
This video was discussed on Reddit, including detailed responses from Sytharin and JinkyRain: www.reddit.com/r/SatisfactoryGame/comments/1dgs4gg/fluid_byproducts_i_tested_the_vip_junction/
Diluted Packaged Fuel is usefull recipe
There is a very reliable way to sink water. Toss it into coal generators. They are always on and will take whatever water you give them.
@@johnyoung5392 Sinking Water in a Coal Generator requires Coal, which is a somewhat valuable resource - sure, you get Power, but once you've moved beyond Coal Generators for Power, that Coal might be more useful for something else.
(Stealing this recommendation from the Reddit thread ...) Using Wet Concrete to sink Water will produce Concrete and is a highly-efficient use of Limestone (and overflow Concrete can be sent via Smart Splitter to an Awesome Sink, so there is no danger of a backup.)
@@dondumitru7093awesome
A German explaining Satisfactory, you know this is going to be good. Great vid man, I can't wait for 1.0 release!
I defnitely did not have to reshuffle any productions because of the 1.0 release date. *looks around nervously*
idk what it is with germans and factory games.
@@tristen9736 MAXIMALE EFFIZIENZ
@@tristen9736Many of the greatest minds(especially theorists and mathematicians) in history were German, and Factory games are science-y and mostly math-ey, so it kind of just fits
@@jcrafterz lol man that was really funny
"Let's just sink them." Spoken like a true German engineer. Efficency at its finest! Great video, happy to sub.
"FICSIT Does Not Waste", after all. Also my favorite production loop is the one that makes aluminiminiminium without involving silica, recycling its own water and only requiring half a pump of fresh input... but due to me not using priority pipelines, you must sink any excess ingots if storage fills, because if the smelters back up into the refineries, _everything will permanently and irreversibly clog._
What are you sinking about ?
The alternative use to German production at large scale - historically - is less preferred. At least according to our neighbors.
The satisfactory community is so good that we got a lecture irl for playing the game. Great vid man, keep it up!
for a channel with just 325 subscriber, this is incredible production value. Well done! The point was brought across expertly, with some entertainment along the way.
Holy shit this channel grew
@@jonvanmaarenyeah no idea how he hasn't before, but glad that number is finally starting to move!
3500 now just 2 months later
Also compare the views this video over the others on the channel. Holy cow
Amazing content and presentation in a brief format explaining the dark art of fluid managment. I think the only joke he missed was saying, "..buffer equilibrium appears to occur at approximately 1803.70 cubic metres..."
I'm going to have to subscribe.
I love your amazing advice here "Get to the fun stuff faster" you're not concerned how people play the game, you're showing us different ways to play the game and that's really cool
Peak German efficiency.
Very cool video. I've been playing quite a while and my solution to output water is...limestone. Make it into concrete and either use it or sink it. I really like the Headlift Reset idea and I will be using it going forward. Thanks for this!
Yes, I realized too late that I should have referenced that option in the script, and so ended up only putting it in the video description.
@@XyrillPlays I really liked your video dude, I watched the whole thing. I just wanted to mention, couldn't you use a smaller liquid buffer? Literally just by lifting it up 2-4 meters, making it as tall or taller than the industrial sized buffer. Or is there something I am missing, like the larger buffer making the pressure rise more/quicker?
Wet concrete was my solution as well. Send it into dimensional storage and any overflow is redirected to a sink with a smart splitter.
Ah, but coal gen is the best, only 1 machine to use, and you need coal for aluminum
Underrated channel
A funny german, a high quality channel, with clear explanation on complicated things in satisfactory? WOW subscribed, you are so much underrated!
just as a note: the ficsit plumbing manual is a bit outdated. Update 7 fixcet alot of the pipe problems, but in update 8 the problems were back again. lets wait till 1.0 to see what the pipes will do then. Great video thow, i liked watching it and loved your solutions.
Well update, I had the VIP set up today..... and it clogged 🙈🙈
In 1.0 I’m pretty sure pipes prioritize output water over fresh water so they will never overfill and fail because of a full water output
@@Pikawarps No really my experience. I now have a setup where I let output water flow in from two sides, with a large buffer on one side and an unpowered pump of the inflow pipe for the fresh water. That seems to have stabilized it due to the fact that at a certain point the pressure from the buffer an the inflow stabilize enough to not cause overfill.
This is unintentionally the most German video on RUclips. Excellent video keep it up!
I don't even play or watch Satisfactory videos. Somehow, this came on my RUclips recommended, and your video editing for a channel with almost 900 subs is incredibly impressive. Great work, sir. I wish you luck on your endeavors.
You should get the game if you enjoy this kind of thing, it may be slightly expensive at 30$ and having a price hike to 40$ soon but its worth it and a great game that has sucked up way too much of my life.
For the fluid buffer height requirement, iirc it's one of the unintuitive things. You could have a vertical pipe the same height as the fluid buffer and it would serve the same purpose. The fluid buffer is more convenient to place instead of creating a vertical pipe and looks nicer. It's kind of like how water towers just need a vertical pipe.
For the priority system, the priority pipe is counter intuitively supposed to be the upper pipe. Liquid flowing downward seems to have a small priority over sideways. It also allows the waste water to back flow into the fresh water pipe, which will eventually cause the water pumps to shut off. In this configuration, it helps to put a safety valve/pump on the downward pipe so that fresh water can't flow into the waste water line. Finally, if the waste water and machines are all full, the system is completely clogged and requires at least the waste water line to be flushed. Just a hunch, but flushing the waste water line after the missing inputs were restored might have gotten the system working again.
Overall, I love the video. You made an amazing demonstration of multiple solutions and have great comedy throughout the video!
A better solution is to put a valve on the wastewater pipe preventing backflow. In the event of an input underflow that should result in the wastewater pipe emptying at least partially, which should prevent the jam.
Amazing post! Thank you. I think the most valuable thing here was the whiteboard. Ahem, I mean the test setup, where you identified what would cause stalls and then verified them. I had no idea that input starvation would behave differently than output clogging. Neat.
FYI: I invariably use the boring method since the pipes behave so differently from reality and they're so unfriendly in terms of debugging what's happening.
The tests are shown in the same order in which I did them, and I was certainly very happy that the Missing Input test was decisive in some way. A conclusion of "they all work" would have been quite boring (and not in the good way).
@@XyrillPlays
Yeah it would have been boring in the "How the heck am I supposed to make this into a video" way.
There are 2 other methods you didn't test tho, Gravity Priority (have the refineries high up and let the byproduct water fall down into a junction that is also fed from below by the water extractors, using pumps to raise the water to the input of the refineries), and "Break the loop" (use the Wet Concrete recipe to dispose of the byproduct water instead of recycling it, which gives you the option of either sinking or recycling the... concrete, which is a lot easier than doing the same with the Water).
Small YT channel producing quality content? I love to see it. Ive never played or ever wanted to play Satisfactory ever. But quality video and made me interested.
I'm surprised the simplest and most reliable option wasn't a contender here: gravity. Just have one water extractor feed into a junction from below, feed to refineries from the sides via pumps that have enough headlift and let the recycled water fall in from the top. As long as the refineries are high enough above the junction, the virgin water source can't fill up the pipe coming down from top, but the inputs are kept full. The system never gets deadlocked and has instant recovery when the solids are moving again.
+1 to the Gravity Priority Valve idea (that's what I call it).
Or a water tower. Make a upside down "U" pipe well over head lift, pump wastewater to top of pipe and let it fall over on the other side.
Great idea! I've not dealt with waste water yet but the theory seems sound and exactly how I would try to approach this problem. I hate finicky unstable 'solutions' and build for endurance, adaptability and self-managed recovery if possible.
This is one of the most high-quality Satisfactory (but also any game) videos I've ever seen. I love the format! Can't wait to see more of this.
this was too entertaining for a 3k sub youtuber. Excellent editting and acting. 10/10
For the VIP I think the issue is that when you take away the bauxite,fresh water still being pumped in. There is no height buffer limiting how much the pumps can fill. There is no valve limiting fresh water intake either so the fresh water pump is pumping a full 600m^3 a minute as opposed to 240m^3 a minute. Had you waited longer before restarting, I think the merged valve setup might have locked up too.
Also one of my best waste water setups was in a nuclear plant. The waste water was 240m^3 per minute and the sulfuric acid required 240m^3 per minute. So once I had filled the pipes and buffer I could disconnect the pump entirely. That was in Update 7 though. I'm not sure how much the recipes changed in update 8.
This is exactly the problem, and you can solve it by putting a valve on the byproduct pipe. This prevents fresh water from filling that pipe and avoids a jam when the input underflows.
Deutsche Qualität !!! Thanks man, finally someone who dug deep into that infamous topic !! you will def help many people. Thanks again and really good job man
So I went to apply mergless because I've been struggling with this for a while in my alumina factory and I just wanted to be done - but I discovered something because my factory's larger (twice as large in fact!)
9:44 - If you double the size of this factory you can use the output to feed 1/3rd of the input. You'll need 480 water in to 4, but the combined output is exactly what the other 2 need. So no need to build "an extra" building (assuming the double doesn't count as extra)
Hello. I haven't play Satisfactory for 2 years now but I find this video fascinating :) Got a question:
Could we use just a 12m high vertical pipe instead of Industrial Fluid Buffer in the Head Lift Reset solution 7:20 ? Or normal buffer on 4m high elevation.
(I can't test it myself cause I'm waiting for v1.0)
That was my immediate thought, the regular Fluid Buffer should work just fine as long as you give it a base that lifts it up a few meters.
I don't think just a vertical pipe would be enough, since pipe capacity is so limited, making it a very weak buffer. Best case scenario: it works but slowly.
Hah, when I showed this to some friends as a test audience, they had the same question immediately. Given that fluid buffers are just bulky pipes, I think the answer is "yes", with the caveat that @Denkkar also gave, but I have not explicitly tested it.
Vertical pipe works just fine, I tested it a while back when pondering about this idea. It makes sense - its only purpose is to raise "pressure" in the system which depends ONLY on the water column height. Pipe heght only needs to be higher than lower priority input headlift, plus some room for fluctuations (I think), but it can be as tall as you want just to be safe. Mine was pretty high, like 50m above the junction.
And whats pretty cool about this design is its entirely realistic. You can build a scale model with some check valves and bottles of water at different heights, and if you build it right to accout for bernoulli's principle, it will work - one bottle will empty before the other will even begin to drain. Because of this, this junction design might work even with no extra vertical pipes/buffers at all, as long as theres a difference in input "pressure" - although I haven't tested it and am not sure if the game will model this correctly.
That's how our whole factory is currently build -- but I will now switch to the fluid buffer instead, that looks really nice and actually serves as a small input buffer.
Its wild that when unconcerned with necessity for support, practically everyone starts building sky islands for their industry
I do with the intent to add pillars later, and sometimes I'll put the pillars in before I start building on the factory, but yes, this does make it easier. For me it's mostly cantilevered overhangs off a cliff, sometimes with pillars added later.
@@Uriel238 keep the AC6 aesthetic
Nope. I can’t wrap my head around floating objects that defy gravity. Even my rail must have 2m blocks as support and bridges contain girder and/or pillars
I used to put support on my platforms, but building takes so long even with blueprints that I just couldn't be bothered. No one is gonna look at my stuff anyways. I stopped using walls, even.
Actually, I founded my first 'skybase' under the map: exactly 1/5 of a unit just above the killzone.😅
Great video! As a long time Satisfactory player > 3000 hrs, pipe mechanics have been a bit of an enigma and I've always tried to keep it simple. The optimized fluid buffer approach that you demonstrated is urging me to be more creative with the pipe mechanics.
@Xyrill Plays
- 14:16
For the VIP junction, I don’t know about anybody else personally, but when I use a junction on a pipe, I delete the excess pipe at the dead end to avoid sloshing (which becomes possible since adding a junction splits the pipe into two separate entities).
The VIP Junction basically works by blocking the upper pipes as long as the lower pipe kept flowing.
The trouble starts once the byproduct water production was disrupted and water from upper pipe clogs the output, essentially turning the intended function on its head.
I don't think this is the case I have a vip junction for my baux setup and the input pipes are always 100 full while giving priority to draining the output pipes into the input ones. It always seems to give priority to the lower pipe which is the redirected output pipes.
The actual reason it jams is that the fresh water fills the byproduct pipe itself, which prevents the factory from putting new water into that pipe.
You can prevent the jam with a valve controlling the flow direction on the byproduct pipe, which prevents the fresh water from filling it.
14:40 can I just use a small tank but place it 8m up from the "ground level"?
Haven't tested it, but since a fluid buffer is pretty much just a bulky pipe, it ought to work.
After numerous failures I've adopted a sledgehammer approach. I package everything then unpackage it directly into the machine. This is the only way I've ever found to make machine runs that never fail or lose efficiency. It costs power and machines but it saves my sanity and for anything that MUST NEVER FAIL its really quite effective. My understanding is that under the right situation all of these except for the split method will eventually fail under the right circumstances.
Plus the bonus feature of no pumps needed as you can put the unpackager as high up as you'd need it so that gravity would feed everything below.
Water is pretty much the only fluid that is used as both ingredient and byproduct (aside from sulfuric acid for batteries, but to solve that problem we have the "Classic Battery" recipe still, I think?).
Anyways, there's a neat and handy way to solve this without all the packagers.
Stop Recycling, make more concrete, use more limestone, sink the excess concrete.
The "Wet Concrete" alt recipe is what you need to enable this.
Byproduct water stays in it's own pipe, never interacting with input water. Instead of recycling the water by ANY method, we "use it for something else instead", namely to make Wet Concrete. A handy feature of the Wet Concrete recipe is that it is VERY water hungry, and does NOT produce a liquid byproduct.
If you want to recycle the concrete, you can, but it's not mandatory. If you do, it's as simple as a priority splitter set to "Overflow" one way, with that way leading to an AWESOME sink.
And just like that, the water's GONE!
Isn't that nice?
@@44R0Ndini believe water concrete is for water sinking and concrete being i a byproduct
@@collinkaufman2316
Maybe that's what it's designed for, but I find it to also be the best concrete recipe anyways, all you need is water and limestone, meaning that if you move the limestone TO the water source, you don't need to transport huge volumes of water. 600 m3/min of water (1 mk2 pipe) runs 6 refineries making wet concrete, which produces A LOT of concrete, either 600 or 780 a minute (I forget if it's 100 or 120 concrete/min output per refinery, but I know it's one of those), meaning that unlike the normal concrete recipe in a constructor, you get MORE concrete out than you put in limestone. There aren't many other recipes that come close to that, and the ones that do require a rarer resource than concrete or water to make work (like quartz in the form of silica, for one of them).
Point is, for the way I play, Wet Concrete is the best way to get rid of water AND the best way to acquire concrete.
"My understanding is that under the right situation all of these except for the split method will eventually fail under the right circumstances."
That's largely my thoughts as well, with the exception people mentioned that just making wet concrete and sinking it is another possibility.
I haven't tried the industrial storage buffer method, but I used to use the valve method when I first started playing. I say "used to" because now I use only the split method or the wet-concrete-sink method - I don't know how or why, but the valve method will eventually break and lock up. Maybe it takes a lot longer to occur than normal, maybe it takes reloading games involved, but it definitely failed on me multiple times, on different playthroughs.
The packager method sounds interesting for nuclear processing, simply because I try to actually limit the amount of radioactive garbage floating about which means load-balancing the buildings - Something which is not really possible with the split method (It will still 'work', but loads in way more radioactive material into buildings than it should).
At least with packagers, I could use a priority merger (stack both lines on top of each other, put splitters all over the 'priority' line, and connect both sides to mergers on the 'input' line with lifts - gets close to 99.9% of just using the priority line at 6 splitters), and sinking the 0.1% rounding error coming out of the end of the priority line
dude just two mins in, im hooked, great production. I am already in love with this, talk about taking a game seriously! Please keep at it.
"Do what works for you and get to the fun stuff faster."
Genius advice. Looking forward to digging into your other videos!
7:00 it's funny to me, because unpowered pumps function precisely as I first intuitively thought valves worked
i cannot believe you only have 4000 subs. please dont stop these types of videos i loved everything about it
I love that this game can be played simply, or it can be played as damn complicated as you want, potentially giving yourself a seemingly endless array of complex but still solvable engineering problems depending on what specific limitations you put on yourself, whether on accident or on purpose. Yet solving those problems is where a huge amount of the enjoyment lies. Really genius game design
As a fellow min-maxing german: I have never been so proud. (Also this channel actually deserves more attention, let's hope the YT algorithm does its thing)
It absolutely does. Both with this and the previous video, the first 100 or so views were majorly from social media (Reddit, Discord, etc.), and then the algorithm got attuned to it and started recommending it on people's homepages.
@@XyrillPlays Data point checking in - I have been watching Satisfactory demonstrations and abominations for a long time now, but only just been recommended your channel. Keep up the (Satisfactory) work, pioneer!
0:20 Easy, give it to me, I drink a lot of water
Love your editing style dude and all the little jokes and stuff hope more people find this!
I didn't test the Head Lift Reset method in U8, but I did test it in 1.0. I don't know if anything changed between the two versions but for 1.0 there's a VITAL component/detail missing from the description in this video that I discovered after much experimentation and reading some comments on the Reddit post referring to this video.
Backstory: I never liked pipes that got in the way of me walking around a platform, so tend to design with the pipes UNDER the refineries.
So first I tested the head-lift reset method with my usual standard: pipes down into pipe holes, and under my refineries' foundations, and back up again.
The buffer and unpowered pump's input and output were all at the same level as the refinery's input, and I thought this would be enough (since this is where the head-lift on the fresh water input would be 0) to work. It took a couple hours to lock up, it but it locked up.
Thinking that the byproduct pipes going down under things and coming back up again might be the problem, I then tested the pipes being lifted about 4m up above the foundations the refineries were on (so just enough room to walk under the pipes) where possible, but still returning back down to the 0-height of the unpowered pumps, buffer, and refinery input/output where they needed to return to the buffer tank and the inputs for the refineries. It also locked up in an hour or two.
I then clipped pipes through belts and such just to get everything PERFECTLY HORIZONTAL, resenting the fact that I had to try it, because I hate pipes that block my ground level views.
It didn't lock up.
Then, based on some conversation on Reddit I looked at for 10 seconds... I simply added about eight junctions to the byproduct pipe, changing nothing else, which divided the byproduct pipe into multiple shorter segments. The whole system locked up in about THREE MINUTES. Not an hour or two, *three* *minutes*. It was the fastest it had ever locked up, by far.
It turns out that if you want to lower the priority of a source, simply make it move through more segments of pipes.
So I tried that: I removed my earlier junctions returning the byproduct pipe to very long unbroken pipes, and then I added about seven junctions to the fresh water pipe prior to it reaching the unpowered pump...
...and then deleted the buffer tank entirely.
It's been running for about ten minutes without so much as a hitch with NO BUFFER. There should be an excess of 60m³/s in the pipes, but nothing is backing up, even without the buffer. Time will tell if it ever locks up, but signs seem to suggest it won't.
I'll probably come back and reply to this comment with more updates if anything changes or if further testing shows anything else interesting.
Follow-up: It's been running for hours with no issues. No buffer, just extra segmentation in the fresh water input.
I've removed the extra segments to see if that results in any change. Buffer is still gone.
After removing the buffer, the line has continued to work without issues. I rechecked my math, and I should still be seeing an excess of 60 m³/s in my system, and yet the byproduct from refineries is draining, and the water extractors are backing up and shutting off.
So I counted pipe segments. Basically, I used the deconstruction tool and highlighted/counted each length of pipe over what I could find would be the longest (in count) distance water would need to travel from a water extractor to a refinery, and refinery byproduct to refinery.
If a pipe was broken up by a pump, floor hole, or junction, that would be two pipes. They each highlight separately.
There are 13 segments between my water extractor and the furthest refinery producing alumina solution.
There are 11 between byproduct refineries making scrap and the refinery making alumina solution.
So I added more segments on the byproduct line, and things STILL didn't back up.
This wasn't making much sense, so I looked closely and realized that while everything else in my pipe network was a Mk2 pipe, the pipes from my water extractors to the main fresh water Mk2 pipe was a Mk1 pipe. I have a theory that, despite 120 m³/s flowing through a pipe rated for 300 m³/s being nothing unusual, that somehow this is acting as some sort of valve or pressure limiter. I've replaced those segments with Mk2 pipes to see what will happen.
@@Moleculor Flow rate is based on how full a pipe is. So a Mk.1 pipe that is half full can only flow at 150m3/s max. However, a half full Mk.2 pipe can flow at 300m3/s.
The other thing to take note of is that pipe segments have volumes. You can watch each segment to see the volume rise and fall as fluids pass through. While dropping a pipe through the floor doesn't affect headlift, it does increase the volume of that pipe and the rate at which it will fill and empty due flow being related to volume.
One thing I will note about the VIP is that it needs to be built as close to the picture as possible. The failure scenario comes about when the volume of pipe between the pump and the junctions are different.
This was a great video, both the content and the editing. You are a good speaker with a sense of humor so I found it very entertaining. I also love the rail design idea, going up when empty and down when full. Smart, but I don't know if that makes a difference mechanically in the game. Given the 1.0 announcement I wonder if any of these will change, or if they would ever consider adding a fluid intake to the AWESOME Sink?
Congrats on a well produced video. Content is good too.
It's been a while since I did the Aluminium line in Satisfactory.
I think the way I got around the water waste was based on the headlift limit. My plant was up-hill compared to the water source, so the headlift of the water pump could only fill a water storage half way. With the other open pipe connector of the water storage on the same level as the chemical plants, I just fed the output back to the input, with the storage also on the input. It just worked.
I guess this is like your headlift reset example, but this was back in Update 3, before pumps existed, so the fluid buffer is doing the same job as your unpowered pump.
This is really well done! I love your humor style and excellent thorough explanations. Fluid byproduct management has been my biggest struggle in satisfactory and I think I can finally get my head wrapped around how to handle it. Thank you for all the hard work you put into this video!
Oh man, such a great video. I love this kind of approach, so informative and with a researcher mindset. I would really love to watch more like these.
The level of production from a person with 1.6k subs is amazing. Looking forward to more content!!
7:00 Would the smaller fluid buffer work if it were elevated? You can build it right above the cross-connectors, elevating it 4+ meters, and get a more compact footprint. It seems to work in my testing so far, but I don't yet know if it'll fail eventually.
I love your humor, and really interesting content too. Thanks for making this, it should help me a lot soon, now that I'm working my way through oil processing!
The amount of effort you put into this really shows lol. For what is just a... game video describing some functions. I'm quite pleasantly surprised!
Editing style and presentation are so friggin' charming, great job!
terrible jokes aside, it is criminal that this guy only has 5k subscribers. Thanks for the video homie!
Seriously the best guide ever, especially for calling the control for what it is: so easy it's boring... baffling why this isn't the go-to for everyone except for those who seek the extra challenge, but 🤷♀️
Great video! I've wrestled with byproduct water in the past and what I came up with was nowhere as clean as these solutions.
I don't know if something changed in 1.0 but I can't get the headlift reset method to work at all. It works initially, then after a short time (~15minutes maybe) the buffer fills to a point and then the system gets stuck. I've tried configuring it almost exactly like it is in the video (basically the same minus the fact that mine isn't on a platform up in the air but at ground level on the water), as well as ~4 different variations.
I wanted to go with it, because it seemed intuitive and reliable, though I just can't figure out how to make it work right. But the VIP junction worked fine right out of the box. I've been thinking about how that junction must work, and the only thing I can come up with is that the game isn't a 100% accurate fluid simulation, and instead it's calculating the lower pipes first and then filling them with the upper pipes, perhaps as just some byproduct of how the simulation is written. Or as someone else put in the comments, it's more like a bottle than it is an actual pressurized fluid system. Perhaps from a technical standpoint the game is assigning a height value to the pipes (which it would have to do for gravity) and then calculating lower pipes on the system and then working it's way up. This would cause it to calculate the byproduct pipes first, filling them as much as possible, and then calculate the upper ones and filling the remainder.
It would be interesting to see the actual programming behind the fluid simulation to get an idea of how it actually works.
@XyrillPlays, I did not check that linked reddit stuff, however might some useful info.
I also struggled with the fluid by-products but got it somehow managed as most of the time
you can somehow even the numbers between "by-product" and "fresh" line and then you are fine.
I have all 5 phases completed but did not yet run some uranium power setup, so I decided to setup some uranium power.
For sure good prepared, but this also means that I had a lot of "fresh" (much too much),
coming from above with huge headlift.
Means for my encased uranium cell production, my "fresh" sulfuric acid coming from above
was huge in comparison to the "by-product" I re-sent to this blender (as I always start
with a "small" scalable test setup) in production - so really huge in comparison to the
600/min possible with a MK2 pipeline.
So, I thought that this #2 headlift-reset starting at 6:33 in your video sounds really great,
but this unfortunatelly did not work for me. I think because when I see your setup in your vid, your "fresh" is
coming from below, so I assume that your "by-product" has more headlift as your "fresh".
However, headlift reset is what I need, so thx for that info with that unpowered pump - did not yet know this.
Whatever, my general setup for sure would start with this #2 headlift reset.
As a next step to fix my problem, I added valves to ensure the flow on specific locations.
This helped only temporarily, as soon as the industrial fluid buffer was full and the
"by-product" pipes were also full, "by-product" began to stuck outflowing.
So how to solve? (when "by-product" doesn't have more headlift than "fresh" has)
Add a "powered" pump for the "by-product" at 6 o'clock (6:33). This will lead to that a headlift
is applied to the "by-product" flow. Then "by-product" has the highest headlift and the fluid
buffer will always be filled by the "by-product" because of the headlift-reset of the "fresh" fluid.
assumption:
"work-preasure" seems to be some kind of related to the max (or current?) flow throughput between
2-X flow lines when the flows do hit in the junction.
E.g.: crossing 2 flow lines.. #1: 20 in , #2 10 in , #3 15 out - may result in that #1 10 and #2 5 into #3 -- only assumption)
I think this mainly should be taken in consideration in case flow lines do have the same headlift - at least in regards
to the target.
If you don't want that flow line 1 is influenced by flow line 2 here (as flow lines can influence each other), place valves
(where a pump - powered or unpowered - does probably fulfill same function as a valve for that "work-preasure" case)
for 1.0:
I have no fancy setups and my plant works completely fine (at 100%), I have just a simple loop from waste water back to production + extra water from extractors. the only noticeable thing about my setup is that my water extractors sits lower than where the excess water is produced. So I suspect higher remaining head lift at a junction has priority for filling the pipeline.
my setup in detail: 2 water extractors at 300/min few meters below main pipeline, and capable of fully saturating the whole pipeline, and alumina solution production.
Main pipeline... 2 refineries producing 200/min of alumina solution each... into 1 refinery consuming 400/min alumina solution, with simple flat pipeline for waste water recycling.
Waste water connects to the same junction as water extractors.
Really high quality material, for video to capture interst of a person wich was never interested in satisfatory or hydrodynamics (me) truly speaks about author skill. Thanks for your work.
13:55 My guess would be that it can manage to go back to production when output clogged because at an output clogg the pressure is "released" from the back of the production chain when output is not clogged anymore. So when input is halted it wants to start up again the second refinery isn't the first one to run, but the first one is but it can't output because the loop is already fully pressurised, and the input from the fresh water have catched up and is also completely pressurised. I'm pretty sure that you could merge the VIP and HLR methods so that the head pressure of the output recycled loop is higher than the feeding from the freshwater pump. But then you might aswell use the HLR method.
VIP worked great for me in release 6-8, thought I would try the HLR after this video in 1.0 and it failed miserably. Back to VIP and it corrected itself and has been good ever since.
My godness, what a great guide! How am I finding this Chanel only now?
You deserve a sub, tis is amazingly well and easily digestable information
We need more of this!
Ich bedanke mich bei dir :)
Thank you so much for putting this together. I just ran into this problem today setting up aluminum with the hope of recycling the byproduct water
your video is by far the best I have seen to get my head aroud fluid dynamics in this game
Just found your channel and I must say I love your content and your humor!
14:04 i think the issue is the distance to the joining section. since the bottom pipe has that extra distance added instead of just going straight to the join. couldn't explain why but thats the only think i could think of as an issue.
7:22 cant you just build the small tank a little bit higher up to so it doesnt fill ? in that case cant you just build a large pipe upwarts ? or is the volume needed for the tank ?
This honestly reminds me of old youtube videos, I definitely missed the raw feeling this video evokes.
17:35 "So this is what's its like to play Factorio."
More true words have never been spoken!
Sick video bro. Diggin your style.
The "there is no intercom, you are adding this in post" got your my sub :)
Love the detail you've gone into here, bravo!
You don't even know how much stuff I cut out to keep this below 20 minutes. :) It's so easy to just get lost in this game.
what I recommend is that you make yourself a hybrid design, having one or two refineries running backflow, and rest going though pump to fresh water.
if you also priority input of ores for that machine, it behaves best in test where you have less input than ideal.
3/4 of those designs would fail if you would have partial input. expecially if you just run normal spliters.
fluid buffer version & headlift reset can be made with just a vertical pipe, or smaller tank if you raise it 4/6m. MIGHT be easier to hide. Issue with just vertical pipe, is that you might have low throuput, slowing down recovery at times. but hey, multiple vertical pipes do same as just tank.
this is a very good video and channel deserves lots of subscribers excellent quality
Will you be making more Satisfactory videos? This is incredibly informative and your world looks amazing from what we've seen
Not on fluid management, but I have one video that is in the research stage, and I have one very vague idea. But I won't be making videos just to fill a schedule. There will be a video whenever I have something to say that I find worthy of the audience's time.
@@XyrillPlays Makes perfect sense, looking forward to it already :)
Thanks for all the effort and time! Gonna save this to reference for my next build!
You earned my sub when I saw the big pipes at the start of the video
As a fan of the mergeless design, I was pleased to see my build has been rated Boring. Thank you for the great video!
Love how you included the pump storage plant Niederwartha in the intro! Love from Radebeul :D
Awesome video! Entertaining and very informative! you speak well and the information was displayed and organized in an easy to understand way.
Hope to see more like this from you!
this is exactly the type of satisfactory content im looking for
Hey man, as a Gold Coffee cup owner I've come back to the game to do it again for 1.0. Been watching some videos to jog my memory. i really enjoyed your video.
I really like watching your vids, it feels like im working toward a diploma in Satisfactory engineering
Holy, dude literally edited satisfactory video like its some TV show on National Geographic
Interesting, I usually go with the valve, easy, clean and it works just fine.
Just one nitpick pipes don't have pressure, not at all. What Darren Plays explained this pretty well, we need to start think those pipes as bottles and that is why the VIP is working that way.
The bottom one is a bottle that will receive water from top when "there is space for it". If the bottom bottle is full => the top will not provide any water, if the bottom have space the top bottle will star adding water to the bottom one. Easy as that. On why this stops is kinda tricky, but I suspect this is a deadlock. You can't consume, so you can't add more water and the water itself is just clogged. If you remove the any water from either input or output it will probably restart and get back to 100%. Additionally I suspect the HLR will suffer the same problem, just wait until the buffer is full and you will have a deadlock as well. You can check that as well by looking at the buffer before the input removal and after, then wait a while... I suspect the buffer will be nearly the same amount and will never drop to zero again. Is just a delayed deadlock probably.
Hope this helps!
I go with, sink the water into a coal gen lol, easier if u have spare coal..
@@bencoad8492 You know what uses more water than a coal gen (and therefore takes up less space), uses Limestone instead of coal (which is much more abundant) and produces Concrete (a lot easier to recycle into something useful than the pesky water)?
Refineries. Not the aluminum refineries tho, those don't use Limestone.
Refineries using the Wet Concrete alt recipe!.
Each one uses a whole 100 m3/min Water, and 60 Limestone/min, to output a whopping 120 Concrete/min!
That's a LOT of concrete! Sure concrete's not worth much, but you can recycle it a lot easier into the feeds supplying your Encased Industrial Beam and Heavy Modular Frame production, or just put it into an AWESOME Sink if that's too much trouble.
Concrete is worth like 3 points per item I think? Sure, not a lot, but when you've got T5 belts running at 780/min, that's 2340 points per minute per full T5 belt of concrete you're sinking, and you only need half that much Limestone (and 600 m3/min water, or a full Mk2 pipe) to keep that rate up infinitely.
Considering that 2 T5 belts of Aluminum Ingots output takes 3 T5 belts of Bauxite Input, and that you output less than 2 Mk2 pipes of Water (as a byproduct) for that rate of production, just shoving the byproduct water into the wet concrete recipe makes a lot of sense to me. Sure it takes power, but coal power plants don't really give you a whole lot of power for your water investment anyways, they're your "starter fully automated" power production after all, even when overclocked.
The second I heard the german accent I knew this was going to be very informative with fun humor
14:00 one possible improvement would be to have the pumps at different levels and add a standpipe on top, that would reverse the priorities though
I used the HLR design on a huge aluminum factory I made a few years ago, and back then the instructions showed an upside down U shape in the pipe just before the buffer, but the thing is it failed horribly for me many times. I had to go empty the buffers around the whole factory every few hours or it would jam up. I'll have to go back to that save file and see if I can fix it sometime
Hey man. Just wanted to say that this is a perfect video! Good job!
Thank you for your hard work
There's a simpler solution to the water output.
*Use the water in another process*
Since You are alreary making complex teir C,D, or higher items by the time You need to solve this output- chances are that another item in a chain would require water.
In Your case, aluminum outputs water at some point.
Which means You'll likely need copper within a recipe, or a copper substitute.
You can make Copper ingots with water in a Refinery and get higher yeilds. This helps during the Aluminum Sheets chain.
I think there is a water and coal in a refinery alt recipe, but I've never rolled it in RNG.
Or use iron ingots and water in a Refinery, along with The Iron Wire Recipe.
Which can become useful in the Turbo Motor Chain.
In the Nuclear and Plutonium Chain, You'll have several ways to use the Output water along the way.
Especially if Your build is vertical not horizontal.
Sending the water down a few floors to the simpler items chain, then belting the ingredients up, eliminates the potential sloshy in the pipe
Really enjoyed this video! I was curious if you'd be willing to explain your strategy for train logistics in more detail. Im working on preparing my strategy for my 1.0 save, looked at your save and was curious how you divide up resources collected/produced in large quantities in one location out to several receiving factories. It's something ive never been able to figure out how best to do in my head, and would really appreciate your insight. Thanks!
I can really feel the germyness in here. Love it.
I have never seen a video of you, but your editing is very funny!
I am not too bothered about learning the intracicies of how fluid or fluid machines work in Satisfactory (As long as they work without fail!) but this video was really refershing and wonderful. A great properly scientific test, with great results!
I was unable to get the HLR method to behave as shown in 1.0. When only the supply water through the unpowered pump is connected, it correctly only fills the buffer to 10% or so. However, when the byproduct water is hooked up also, the buffer over the course of two hours fills to 95% or so (probably the HL limit of the byproduct-prodcing building), then jams the whole system up. Looking at the pipes, it seems to behave as though the HL from the byproduct building now also applies to the water coming from the unpowered pump, and the flows compete with each other. The level in the buffer swings wildly, but there's a small net increase across many oscillations.
Followup: I solved this by implementing a suggestion by someone earlier in the comments. I eliminate the dead pump and buffer entirely, and merge my supply water into the byproduct pipe, with the byproduct buildings between the fresh water and the consumption buildings. It works shockingly well. Fast startup and recovery, no rate limiting, and no balancing. In fact, have the extractor give more water than the system needs, and it Just Works. Byproduct water builds up inside the buildings' internal buffers, but it only hangs around for a few seconds before downstream consumption makes room for it, and it fully drains into the line, with supply water filling any remaining capacity. It happens so quickly that I never even see a void appear in the pipes. They stay full at all times, and just have a brief surge in flow.
I never knew of this pipeline manual and have 1000 hours in the game. Thank you friend!
Awesome analysis! I'll definitely be revisiting this when I replay the game. I found the VIP junctions fine enough and small when I played in U6 but looks like I should revisit. Valve is very tempting!
I need to use signs like you do to keep track of my throughput numbers. I'm always forgetting that stuff when I come back for stuff later.
Love your content style and i am a big satisfactory fan. Thought about making videos like this back in the day.
I got to Aluminum just last week. I'll write this before I watch the whiteboard part. The system I built outputed enough water to cover 1/3 of the total needed at the start of the production line. So I cycled it back and put a Valve on both the fresh water and the return water. I set the max flow rate on the fresh water valve to 2/3 of the total needed. Now Im gonna continue watching
I've tried the JustAValve method before looking anywhere else for help, as I assumed that it would solve the problem. For some reason it didn't, so I tried the VIP method...which also didn't work for me. Perhaps I already had too much pump pressure in my main line (I needed a pump somewhat close to my refineries in my layout). I ended up making a very simple junction which has the byproduct (lower flow rate) water coming in on the Refinery level, and the main line coming in (at the same height) to a 4-way that faces up. So the main line bends up and then straight down into the 4-way. That hasn't failed me yet, and doesn't require the area or materials of the VIP setup.
This is a really high quality video for such a small channel, damn
A German explains efficiency in Satisfactory. Sometimes the jokes write themselves. Joking. Thanks. Nice to know my “just a valve” method is legitimate.