Bro thank you. I've been looking for a solution to the first thing your demonstrated for a looong time. So cool! Your other methods looks great as well.
Oddly, I have found valves are good at creating pressure in long pipes. Minimizes sloshing in the pipes. I had 30 fuel generators arranged in two rows of 15, the entire pipeline was saturated before they were turned on, pumps on everything. After a few days one of my generators wasn't getting enough fuel and began sputtering. I simply added a single valve halfway up the pipeline and it began working without a hitch. While pumps generally work, sometimes valves are a no power cost alternative.
@@magepunk7711 correlation vs causation. Valves don't stop back flow. Basically any time anyone uses valves and 'fixes' an issue, time or something else they fiddled with. basically any system that works with a valve will work w/o them
@@magepunk7711 I believe that works because it stops the backflow before the it reaches a section of consistently full pipes. It'll probably still, eventually, cause some production slowdowns at the extractor once the pipes are sufficiently full, but that should only prevent the pipes from ever filling completely rather than reducing efficiency at the production machines. As in, it moves the problem of reduced flow to the last part of the pipe, which doesn't require the full 600/s for optimal efficiency.
Interesting. Never had this issue. Whenever there was any issues with refineries going yellow, the issue always was somewhere else. Like forgetting to upgrade to Mk.2 pipe or something backing up down the line. Never the oil input. Weird.
Same for me. I can’t think of a single time my problem was that oil wasn’t going fast enough. Always something with a manifold, a part of the production chain not being correctly optimized, something like that
Had issues with pipes in the past, most of it was headlift, but I've learned a few things that help things want to stay stabilized. First, if you're flat and within the 10m natural headlift, you dont need to gravity feed from above, but if you are building taller, add a pump and give your pipe a 4m bump at the top before it comes back down to where you need it. Second, if your machines divide the input evenly, theres no reason not to feed from one end of the the pipe or the other, adding into the middle invites feedback issues. Third, try letting the input pipes fill up entirely before you connect the refineries to power, bonus points if you do this just before the extractor comes to rest, but itll work from cold start full pipes. This last point holds doubly true for coal generators, always let them fill with water entirely before the first nugget enters the line. Hope this helps =)
if you loop the ends of your pipe manifold back to your feed pipe it does teh same thing without any extra bp or valves or anything. the loop will allow fluid to flow into the manifold of pipes from multiple directions without worrying about valves or "load balancing" - you are basically doing the same thing with your bp, but in a more complex and unnecessary way. Also, VOLUME (full pipe) does NOT affect FLOW (throughput m3/min) - you *want* full pipes with full volume to prevent backflow. Full pipes = happy pipes. Simply loooping the pipes i have multiple 600m3/min systems without anything so complex as this
@@siriuscowking in a pipe loop, backflow only happens when the connected production machines are not able to consume all of the incoming flow. So, while you're, technically correct, it doesn't matter because if you built enough production machines, no backflow will happen. Edit: Oh and even if it did, both directions would need to be simultaneously blocked for it to cause the extractor to accumulate fluid to it's buffer.
@@siriuscowking you dont need to prevent backflow. If the system is full, and looped, and you are using the same amount that is being produced, you will have no "effective" backflow. (it will still occur through your system between every segment of pipe because pipes are bi directional, and fluid will always flow toward where there is none - which will result in fluid flowing both ways (backflow) out of all segements anytime a machine gulps down fluid (which it does far faster than pipes can fill back up) the internal buffer on machines is a *good* thing and it should be full, because it helps smooth out small variances in the way liquid flows back and forth down pipe systems, letting sometimes more liquid is in that section than needed and sometimes less. with 100% full (Volume) pipes slosh/backflow is not a problem. Fill your pipes before turning on machines, turn machines on in sequence letting pipes fill up again in between, and you will not have issues with backflow at all.
Ever since pipes were introduced I have been having to help friends fix their pipe networks and in every case they had their network set up in an overly complex manner and purely simplifying it fixed all of their problems. Typically, they have a bunch of valves or pumps set up in an attempt to fix the issue of the overly complex network and are frustrated. I do as Lynx said, I loop half my liquids to the far end of my manifold and feed half from each end and have had zero issues since the beginning in my own networks. I have never used valves and only have used a pump in the rare case I need more head-lift, which is rare since I always try to build my fluid networks on ground level.
Pipes are still such a hot topic ^^ I just redid my rocket fuel factory after 6h of troubleshooting my previous one in the past. The main difference? I'm using exclusively MK1 pipes. Supposedly, MK2 pipes have been fixed, but only after this I was able to get a fully working factory. And in this new setup I'm not using a single valve or buffer. It just works.
@@Miglow Nope, just normal pumps right after the water extractors. By now, I have experimented a bit and I think the main issue is that MK2 pipes can't reliably transport 600/min, while MK1 pipes can reliably transport 300/min. I don't know what the threshold is after which MK2 pipes "break", but I'm successfully using them sparingly in places where 600/min are not needed.
I always just run fluids up to a storage tank that is at a higher elevation than any of the sinks. I do also use the elevated "bumps" to create overflow gates. For valves, what I've heard is that you should only use valves with equal limit values connected to a given node.
Nice video and explanation of fluid dynamics. I noticed a distinct lack of vertical pipe junctions.... Much of what you did can be simplified by using a vertical junction, either by attaching to an existing pipe or placing on temp walls (use nudge or walkway for wall placement/alignment). A vertical junction has 3 levels of priority/flow. Feed machines using overhead manifolds with vertical junctions to prevents backflow and force sequential filling (same as using smart splitters with overflow along a manifold). Placing an inverted U to act as a literal fluid overflow helps prevent backflow too.
This is a very important concept and your method offers a far better solution than how I tried to fix it. Try to minimize the amount of time a pipe is at max flow rate. The small pipe leading to a machine will always flow backwards at the first junction due to work pressure. Allowing room for fluctuations to occur is vital in pipe balancing. My method was to feed both ends of my manifold and have multiple fluid compensators in series to ensure a consistent 600 pressure. I would add a single compensator to the end of your blueprint, but splitting at the beginning is elegant enough
Also. Valves aren’t really broken at lower capacity, they just behave poorly in general. A full pipe has a work pressure of 100%, which allows maximum throughput. So a mk2 pipe with 600 fluid running through it will only push that 600 when the pipe is full. A half full pipe will push less. (Not sure if it’s linear so don’t want to say 300). A valve is weird where it both acts and doesn’t act like a pipe. A valve will only push the amount of liquid it’s limited to if and only if the pipe feeding it is full. However, considering the use cases for attempting this, the input pipe will almost never be full. So, in short, don’t use valves as limiters lol.
Not sure if I have a use for this exact layout, but I think I see the principle behind it's behavior. I recently built a modest aluminum ingot factory (1200/m) and have been struggling super hard getting fluids moving correctly. At this point I'm 90% functional and that was good enough, but I may go back and have another look with some of this in mind.
@@thehuggz-i9k for aluminum I have a very simple setup in my other video, the safety strat is to have multiple small network of pipes and pumping in only the water needed to make sure the water doesn't clog in the machine
I’ve never had a problem getting my oil to go at 600/min except the normal stuff like running out of headlift or a buggy pipe connection. Didn’t realize this was a thing people had to work at
@@drachenchunter yeah thats very useful to quick fix a setup manually, I use it to test my systems, as long as the numbers no longer go up I know there's no accumulation somewhere
This seems over complicated. A key rule for reliable pipe flow is to ensure that the manifold into the machines is always higher than the machine input and that the head pressure point is higher than the manifold. That's it, not a lot else is required. Take the machines around 09:30. These have a manifold that is the same level as the inputs and as a result this will cause hell with the manifold as there will be backflow from the machines. To fix this setup, all that is needed is to raise the manifold with a single step on a pipeline holder and then run the machine pipes in at a downward angle rather than horizontal. In this scenario the supply pipe coming across the gap is higher than the manifold therefore the head pressure point should be higher than the manifold (nothing will work otherwise). I have 600/s supply pipes all over the place. For example, 5 water extractors all connected to a low level manifold with a single fluid pump which pumps the fluid up as high as it can and from there it's down to a manifold and fed into machines. As long as the high point of the fluid pump is higher than the manifold everything will run smoothly and a setup requiring 600/s will run just fine every time.
You don't necessarly need the valves. The important part is that liquids get liftet at least once higher than the machine input. Also as far as i'm aware the pipe can't handle the full 600 from a number representation issue. It will take many hours / a view days. But unless they fixed it for 1.0 its 599,9996 or something like that.
I have had the game running on this save file for over 300h now in 1.0 and it's still running at 600 with the setup shown in the video. I know I "shouldn't" need the valves, but it works with them. Let's remember that this is a video game, not real physics, sometimes you need these things for them to work properly
@@siriuscowking Oh, cool. Thanks. Then they fixed it to 0.8, and now. The lift above was not necessarily because of real physics. The algorithm sees everything as a volume, and each liquid has a dynamic number. If a pipe network is on the same level, a fraction of the volume moves from the higher to the lower level. If you filled a tank with water and made a huge splitter with like 9-27 pipes coming out and merging into a new tank it used to swap back and forth basically forever. If you have an elevation change it fills every volume on the same level first. As machines consume and have a buffer they are basically never full. The valve and pumps cut the backflow completely, so they serve the same purpose for the most part.
have 600 without an issue... if you have an issue just put a high bend to ensure the lowest fills... I have 2 600 oil from near desert to the spire coast at a steady 600... its a long haul.. but i have been playing for last 4 years and have figured out the issues.. which are a lack of understanding that set up right you will never run low, you won't have sloshing, and will always have full reserves. as for your video you won't feel the back pressure due to the valve which isn't needed.. place a high spot before the inlets and problem is solved
@@deadcell66 full pipes has no impact on wether they are working well or not, a water tower is the equivalent of a very large pipe and has no impact either. Any source other than a pure oil node maxed out will give you less than 600 per min, just don't have 600 per min in a pipe unless you have to
@@siriuscowking Righty. Fluid mechanics are still confusing to me. This system does seem to work amazingly! So thanks for that. Now I'm just wondering if it would be wise to use more simplified versions of this on other machines that provide fluids as well, like refineries, mixers, fluid storage. Or is this only applicable when you're dealing with full pipes?
@@Holistra it takes 5 min to make, just place a foundation with a junction on it, build 2 more junctions, build the valves, connect the pipes to everything, delete the foundation
I was planning on a rocket fuel VS ionized fuel video, I could include a simple setup in it to show the process, this of course would include turbo fuel and diluted fuel
@@siriuscowking Ionized is actually fun, I am building a setup that will output 1200 ionized fuel per minute only using SAM, water, oil, coal and nitrogen. Alt recipes ftw haha (oops will be more like 2400 since I'll be boosting my final stage with sloops)
While the method used here would definitely work and prolly the safest one, the other method was pumps but it use power despite its just 4 to 8MW, the powerless method I propose was just put valve to the direction of the pipes that would get filled first in the manifold that way if you have extractor that was lower than the factory, the filed one will be full and wouldnt be able to back flow because of the valve and the extractor would just continue to fill the other side, you could also put valve on all 3 side if youre worried, another thing is to fill the input from above but it would sacrifice the uplift make sure you have less than 10m, youd always want the liquid to flow in one direction so avoid filling them from the the center like the vid shows there as much as possible, if not then just use one of the method shown or what I proposed there...
@@key462 I recently tested this. I built a pipe connected to water extractor and a pipe that went up 12m from the centerpoint of the water extractor's output. Then I tested it with both valve and unpowered pump. With Valve the pipe filled fully while with unpowered pump, it only filled to about 60%. Though, I did not test,yet, what happens if the valve or pump were to be placed at 10-12m height.
@@joeljoonatankaartinen3469 Interesting tbh ive never use pumps as valve before, if i need them to flow somewhere I just use valve instead since they have smaller footprint, maybe it was that, since pumps have longer body most of the pipes it occupy were mostly occupied by pump body length the and continuity of liquid was probably prevented by that pump body or something, hard to judge unless you see the games code
In the first bit with the blueprint. It appears you made a flow compensator. The other bit at alum fac is a high and low priority splitter using gravity. Let me know. if right or wrong.
No unlike belts fluids will flow in any direction and accumulate at the lowest point. This is intentional, as it's low compute approximation of what liquids in pipes would do.
Doesn't carrying the liquid up to a higher position than the manifold line by pumps then splitting it and connecting it to the both ends of the manifold fix the issues with 600m³? I mean do we really need this much?
gas doesn't actually flow, it just sort of fills the entire system at the same rate, so as long as you have enough flow capacity you will never have any problem
The core of the solution seems to be to not use pipes for 100%, right? Even with verticality and valve, that by itself would not prevent the stoppage from going up to the oil output. The principle of transmission is "can't flow forward", not "flowing backward"
The problem with mk2 pipes isnt sloshing, its in how the calculates flow with floating point values. Theres nothing you can do to make a mk2 flow 600 perfectly forever. The best solution is to never design your factories in such a way that youre counting on a pipe to carry 600
yet there is. They fixed the floating point values. I have multiple oil extractors which run at 600 cubic meters at 100% efficiency with 0 trouble if you remove sloshing and have a good setup.
There are issues with how the code rounds up/down floating point numbers. Eventually it breaks perfect systems that use max 600m3 pipes. You can easily fix it by dropping the usage to 599m3 with an oil extractor working on 600m3 speed. That extra 1m3 fixes it. Everything else is a cope
I always have fluid input manifolds Above the connection point and looped instead of strait line and equal or below connection point. that way gravity is never working against flow. This way my input manifolds are essentially fluid buffers that gravity feed. Never had any issues with flow since starting to do it this way and always utilize max yield from oil nodes. And always make sure to prime your pipes, let them fill to max including input and output buffers of factories and mines before turning on production lines. If you turn on a production line without maxing fill capacity of the pipeline, you will never get correct flow even if you're producing enough
Hi! I watched this video about a few days ago and I was amazed by these designs. However, today I saw another video: ruclips.net/video/ZwO-F82sYE4/видео.html And I feel like the first design and the one you're using as an example at the aluminum build are the same. To my surprise, that's the one that failed the missing item test, while a valve surprisingly worked just fine. And even though I don't know how that second design works, it performed very well compared to all the other options. Is there a difference between the two examples or would your system have the same issue if it lacks the items for some time? Because in case it does have the same problem, it might be better to use a different method.
@@nebula600 the method I suggest in my basic liquid tutorial is the simplest way to deal with liquid byproduct, just throw it back in the loop. As long as there is no power shortage, there will never by a missing item input. On that same note, you should always have an awesome sink to deal with the overflow. If you are the kind of player who doesn't keep an eye on their power usage and also don't have a reserve battery, then you might want to consider using the liquid on a 2ndary production. That being said, making more power plants isn't difficult considering the amount of rocket fuel you can make for even just 150 crude oil, see my next video.
@@siriuscowking Oh don't worry, I would give ANYTHING to make my power chart 2 straight lines. Technically just for my OCD, but of course the good part is if something isn't at 100% efficiency, I can read that off from the power chart which would be AMAZING, if not for those things that consume/produce power in an oscillating way.😡
The valves are not helping with fluid sloshing. The proper way to deal with the issue is to build a second pipe above the the one supplying refineries, parallel to it, and connect it to the ends of the feeding pipe. This way any sloshing you get will be above the input line, until it fills up, and while its happening the lower pipe will always be full anyway. If refinery gets blocked in the meantime, excess oil from the upper pipe will always act as a buffer.
I'm going to say this once and probably TWICE, STOP THE BACKFLOW with PUMPS. More sloshing means more backflow, that means the extractor knows that something is up and doesn't run at 100%. This is an issue with people that barely know pipe mechanics and they're a bit buggy but still. Learn to feed from above and keep things from going around and sloshing; key word i'm going to say... Valves are cancer in the game and I suggest just using pumps and gravity feeding things. I like the ingenuity but my personal engineering skills (and game experience) want to screech with how much you made the thing 10x more complex with slapping valves down for liquids. (Yes valves do reset the headlift and that's why they're rarely used)
First, valves don't reset headlift. They used to, but not anymore. Second, I completely agree that this is a way too complex of a solution. I just finished an oil refinery setup using up 600/s of crude oil from a pure node using just Mk2 pipes last week. No pumps, no valves. Also, the refineries are actually several meters *above the oil extractor* and the loop is actually vertical because I supply both end of the refinery pipeline from below. The trick? Make a loop out of the pipe that feeds the refineries. As in, don't connect it to the middle, but junction the pipe that comes from the extractor and connect it to both ends of the refinery pipeline. Loops are very effective at making backflow a non-issue because in a loop there is no wall to reflect back from. The only time a loop would start disrupting the flow is when the refineries aren't actually pulling in the full 600/s.
@@tabascosauce6490 RUclips either is being slow or ate my previous reply, so I'll try again with some expanded testing results. I have 5 different test results here. The last two of them don't make too much sense, especially combined together, but are interesting regardless. 1. I tested a water extractor connected to junction followed by a valve and 10 meters of pipe straight up followed by a short horizontal pipe. Both filled up 100%. 2. I built an unpowered pump to the second port of the junction, followed by the same pipe structure as in test 1. The first part only filled up to 72%. 3. I powered up the pump, saw that the pipes filled fully, so I continued the pipe up to the maximum length possible. The last part filled up to 20l. 4. I then added a valve after the pump right after the 90 degree turn up. The fill rate at the topmost pipe dropped to around 15l. 5. Just for fun, I then added another valve a bit above the first one. To my suprise, the water level at the topmost pipe shot up to 100% , which is 70.4l. I tried adding one more pipe, but it only got like 0.4l after that. Though, when I removed the second valve and flushed the 70.4l pipe and rebuilt it, it only filled back up to 15l and stayed there. It only went back to 70.4l after I flushed it, so this is probably some kind of a bug. So, it does seem like valves can adversely affect headlift when placed after pumps, but I don't think it's correct to say that they reset it.
@joeljoonatankaartinen3469 oh fun lol. Yeah I was using 400 out of 600 of a mk2 pipe in two different locations. Went to check the pipe and the area using 300/s was getting 600 flow rate somehow even after letting it set for a while... So tried using a valve to restrict flow rate and seen it stopped flowing up the hill so had to add pumps. But my pipes were flowing atleast 40 or 50m up after the valve. Which was still way lower than the source it came from. Eventually just had to split the mk2 pipe into two mk1s to fix the issue. Fluids are never fun to deal with
They don't fix the problems. Tanks are just big pipes. Your pipes already function as tanks. In fact, those specific tanks fudge the flow direction and headlift, so I wouldn't use them
You don't need any valves or pumps. You're doing something very wrong here. Valves also don't stop back flow, fluid in front can still knock back fluid behind the valve even if the fluid units don't cross.
@@siriuscowking you certainly said some things, but it's perfectly feasible to get 600 from 1 node w/o valves and usually without pumps. I honestly don't know what you're doing to your pipes to make this be an issue but you're free to get some pipe trouble shooting on the discord
Orrrrrrrrrrr, you set the two other valves at 300 each, problem solved without the extra pipe intersection....If you need a 200/400 devide, set 1valvue to 200, the other to 400. Not that difficult.
Bro thank you. I've been looking for a solution to the first thing your demonstrated for a looong time. So cool! Your other methods looks great as well.
Forget the valves, just use a pump directly after the extractor.
Oddly, I have found valves are good at creating pressure in long pipes. Minimizes sloshing in the pipes.
I had 30 fuel generators arranged in two rows of 15, the entire pipeline was saturated before they were turned on, pumps on everything. After a few days one of my generators wasn't getting enough fuel and began sputtering.
I simply added a single valve halfway up the pipeline and it began working without a hitch.
While pumps generally work, sometimes valves are a no power cost alternative.
@@magepunk7711 correlation vs causation. Valves don't stop back flow. Basically any time anyone uses valves and 'fixes' an issue, time or something else they fiddled with.
basically any system that works with a valve will work w/o them
the only difference between a valve and a pump is headlift, they have the same functionality otherwise
@@cobaltofdoom Yes, valves do stop backflow. Literally the only thing I did was add another valve.
@@magepunk7711 I believe that works because it stops the backflow before the it reaches a section of consistently full pipes. It'll probably still, eventually, cause some production slowdowns at the extractor once the pipes are sufficiently full, but that should only prevent the pipes from ever filling completely rather than reducing efficiency at the production machines. As in, it moves the problem of reduced flow to the last part of the pipe, which doesn't require the full 600/s for optimal efficiency.
Interesting. Never had this issue. Whenever there was any issues with refineries going yellow, the issue always was somewhere else. Like forgetting to upgrade to Mk.2 pipe or something backing up down the line. Never the oil input. Weird.
Same for me. I can’t think of a single time my problem was that oil wasn’t going fast enough. Always something with a manifold, a part of the production chain not being correctly optimized, something like that
Same I was about to say just hook a pipe to it
Had issues with pipes in the past, most of it was headlift, but I've learned a few things that help things want to stay stabilized. First, if you're flat and within the 10m natural headlift, you dont need to gravity feed from above, but if you are building taller, add a pump and give your pipe a 4m bump at the top before it comes back down to where you need it. Second, if your machines divide the input evenly, theres no reason not to feed from one end of the the pipe or the other, adding into the middle invites feedback issues. Third, try letting the input pipes fill up entirely before you connect the refineries to power, bonus points if you do this just before the extractor comes to rest, but itll work from cold start full pipes. This last point holds doubly true for coal generators, always let them fill with water entirely before the first nugget enters the line.
Hope this helps =)
bro thank you so much. i spent 2 weeks on my alternate turbo fuel refinary and still have some refinaries in the yellow...will try some of these
Nice video! I like how you lay out the basics of fluids. I actually learned a lot from your explanation of backflow.
if you loop the ends of your pipe manifold back to your feed pipe it does teh same thing without any extra bp or valves or anything. the loop will allow fluid to flow into the manifold of pipes from multiple directions without worrying about valves or "load balancing" - you are basically doing the same thing with your bp, but in a more complex and unnecessary way.
Also, VOLUME (full pipe) does NOT affect FLOW (throughput m3/min) - you *want* full pipes with full volume to prevent backflow. Full pipes = happy pipes.
Simply loooping the pipes i have multiple 600m3/min systems without anything so complex as this
that does not prevent backflow at the first junction
@@siriuscowking in a pipe loop, backflow only happens when the connected production machines are not able to consume all of the incoming flow. So, while you're, technically correct, it doesn't matter because if you built enough production machines, no backflow will happen. Edit: Oh and even if it did, both directions would need to be simultaneously blocked for it to cause the extractor to accumulate fluid to it's buffer.
@@siriuscowking you dont need to prevent backflow. If the system is full, and looped, and you are using the same amount that is being produced, you will have no "effective" backflow. (it will still occur through your system between every segment of pipe because pipes are bi directional, and fluid will always flow toward where there is none - which will result in fluid flowing both ways (backflow) out of all segements anytime a machine gulps down fluid (which it does far faster than pipes can fill back up)
the internal buffer on machines is a *good* thing and it should be full, because it helps smooth out small variances in the way liquid flows back and forth down pipe systems, letting sometimes more liquid is in that section than needed and sometimes less. with 100% full (Volume) pipes slosh/backflow is not a problem.
Fill your pipes before turning on machines, turn machines on in sequence letting pipes fill up again in between, and you will not have issues with backflow at all.
Ever since pipes were introduced I have been having to help friends fix their pipe networks and in every case they had their network set up in an overly complex manner and purely simplifying it fixed all of their problems. Typically, they have a bunch of valves or pumps set up in an attempt to fix the issue of the overly complex network and are frustrated. I do as Lynx said, I loop half my liquids to the far end of my manifold and feed half from each end and have had zero issues since the beginning in my own networks. I have never used valves and only have used a pump in the rare case I need more head-lift, which is rare since I always try to build my fluid networks on ground level.
Pipes are still such a hot topic ^^
I just redid my rocket fuel factory after 6h of troubleshooting my previous one in the past. The main difference? I'm using exclusively MK1 pipes. Supposedly, MK2 pipes have been fixed, but only after this I was able to get a fully working factory. And in this new setup I'm not using a single valve or buffer. It just works.
Using a water tower?
@@Miglow Nope, just normal pumps right after the water extractors. By now, I have experimented a bit and I think the main issue is that MK2 pipes can't reliably transport 600/min, while MK1 pipes can reliably transport 300/min. I don't know what the threshold is after which MK2 pipes "break", but I'm successfully using them sparingly in places where 600/min are not needed.
rocket fuel is a gas.
I always just run fluids up to a storage tank that is at a higher elevation than any of the sinks. I do also use the elevated "bumps" to create overflow gates. For valves, what I've heard is that you should only use valves with equal limit values connected to a given node.
Nice video and explanation of fluid dynamics.
I noticed a distinct lack of vertical pipe junctions....
Much of what you did can be simplified by using a vertical junction, either by attaching to an existing pipe or placing on temp walls (use nudge or walkway for wall placement/alignment).
A vertical junction has 3 levels of priority/flow.
Feed machines using overhead manifolds with vertical junctions to prevents backflow and force sequential filling (same as using smart splitters with overflow along a manifold).
Placing an inverted U to act as a literal fluid overflow helps prevent backflow too.
Good video and great solution - for upcoming problems, not there yet but planning ahead! Great comments too, Thanks much!
This is a very important concept and your method offers a far better solution than how I tried to fix it. Try to minimize the amount of time a pipe is at max flow rate. The small pipe leading to a machine will always flow backwards at the first junction due to work pressure. Allowing room for fluctuations to occur is vital in pipe balancing. My method was to feed both ends of my manifold and have multiple fluid compensators in series to ensure a consistent 600 pressure. I would add a single compensator to the end of your blueprint, but splitting at the beginning is elegant enough
Also. Valves aren’t really broken at lower capacity, they just behave poorly in general. A full pipe has a work pressure of 100%, which allows maximum throughput. So a mk2 pipe with 600 fluid running through it will only push that 600 when the pipe is full. A half full pipe will push less. (Not sure if it’s linear so don’t want to say 300). A valve is weird where it both acts and doesn’t act like a pipe. A valve will only push the amount of liquid it’s limited to if and only if the pipe feeding it is full. However, considering the use cases for attempting this, the input pipe will almost never be full. So, in short, don’t use valves as limiters lol.
Not sure if I have a use for this exact layout, but I think I see the principle behind it's behavior. I recently built a modest aluminum ingot factory (1200/m) and have been struggling super hard getting fluids moving correctly. At this point I'm 90% functional and that was good enough, but I may go back and have another look with some of this in mind.
@@thehuggz-i9k for aluminum I have a very simple setup in my other video, the safety strat is to have multiple small network of pipes and pumping in only the water needed to make sure the water doesn't clog in the machine
I’ve never had a problem getting my oil to go at 600/min except the normal stuff like running out of headlift or a buggy pipe connection. Didn’t realize this was a thing people had to work at
You can just grab the excess and trash it? 500 hours in and i didnt know that
@@drachenchunter yeah thats very useful to quick fix a setup manually, I use it to test my systems, as long as the numbers no longer go up I know there's no accumulation somewhere
Just cause I recognize your accent and we come from the same place, I'll give you a like
Where is it from? I'm not sure
montreal, canada
This seems over complicated. A key rule for reliable pipe flow is to ensure that the manifold into the machines is always higher than the machine input and that the head pressure point is higher than the manifold. That's it, not a lot else is required.
Take the machines around 09:30. These have a manifold that is the same level as the inputs and as a result this will cause hell with the manifold as there will be backflow from the machines. To fix this setup, all that is needed is to raise the manifold with a single step on a pipeline holder and then run the machine pipes in at a downward angle rather than horizontal. In this scenario the supply pipe coming across the gap is higher than the manifold therefore the head pressure point should be higher than the manifold (nothing will work otherwise). I have 600/s supply pipes all over the place. For example, 5 water extractors all connected to a low level manifold with a single fluid pump which pumps the fluid up as high as it can and from there it's down to a manifold and fed into machines. As long as the high point of the fluid pump is higher than the manifold everything will run smoothly and a setup requiring 600/s will run just fine every time.
The simple solution is to use downflowing feedpipes into any consumer of fluid.
that's not always an option, besides you'll most likely have an issue at the first junction either way if you only count on that
You don't necessarly need the valves. The important part is that liquids get liftet at least once higher than the machine input. Also as far as i'm aware the pipe can't handle the full 600 from a number representation issue. It will take many hours / a view days. But unless they fixed it for 1.0 its 599,9996 or something like that.
I have had the game running on this save file for over 300h now in 1.0 and it's still running at 600 with the setup shown in the video. I know I "shouldn't" need the valves, but it works with them. Let's remember that this is a video game, not real physics, sometimes you need these things for them to work properly
@@siriuscowking Oh, cool. Thanks. Then they fixed it to 0.8, and now. The lift above was not necessarily because of real physics. The algorithm sees everything as a volume, and each liquid has a dynamic number. If a pipe network is on the same level, a fraction of the volume moves from the higher to the lower level. If you filled a tank with water and made a huge splitter with like 9-27 pipes coming out and merging into a new tank it used to swap back and forth basically forever.
If you have an elevation change it fills every volume on the same level first. As machines consume and have a buffer they are basically never full.
The valve and pumps cut the backflow completely, so they serve the same purpose for the most part.
bro, thx, UR so damn cool. As a beginner, I connected the pipes from below and my production stopped. Thx
have 600 without an issue... if you have an issue just put a high bend to ensure the lowest fills... I have 2 600 oil from near desert to the spire coast at a steady 600... its a long haul.. but i have been playing for last 4 years and have figured out the issues.. which are a lack of understanding that set up right you will never run low, you won't have sloshing, and will always have full reserves.
as for your video you won't feel the back pressure due to the valve which isn't needed.. place a high spot before the inlets and problem is solved
So why wouldn't I want full pipes?, also, would yu say use the splitter you made for water as well? And what abut adding a water tower?
@@deadcell66 full pipes has no impact on wether they are working well or not, a water tower is the equivalent of a very large pipe and has no impact either. Any source other than a pure oil node maxed out will give you less than 600 per min, just don't have 600 per min in a pipe unless you have to
@@siriuscowking Righty. Fluid mechanics are still confusing to me. This system does seem to work amazingly! So thanks for that.
Now I'm just wondering if it would be wise to use more simplified versions of this on other machines that provide fluids as well, like refineries, mixers, fluid storage. Or is this only applicable when you're dealing with full pipes?
This is pretty cool.
Do you have somewhere that you are sharing this blueprint ?
@@Holistra it takes 5 min to make, just place a foundation with a junction on it, build 2 more junctions, build the valves, connect the pipes to everything, delete the foundation
Are there any engineers in the crowd? I'm curious as to the hydrodynamic modeling the game is using.
@@Shalmaneser1 there was an engineer in the crowd earlier, basically the game mechanics definitely does not reflect real life haha
Would a fluid buffer do the same thing? Or is this better for some reason?
@@Piecesxiii a fluid buffer simply holds a lot of liquid like a big pipe, the limit of the flow is still 600 per min, it does not solve the problem
i use pump + junction + valve + pump to create manifolds, it works great,
right know i'm using it on 3 places in my 2400 turbo fuel energy plant
Do you have a tutorial on turbo fuel not making it so much but a setup that works well?
I was planning on a rocket fuel VS ionized fuel video, I could include a simple setup in it to show the process, this of course would include turbo fuel and diluted fuel
@@siriuscowking Ionized is actually fun, I am building a setup that will output 1200 ionized fuel per minute only using SAM, water, oil, coal and nitrogen. Alt recipes ftw haha (oops will be more like 2400 since I'll be boosting my final stage with sloops)
While the method used here would definitely work and prolly the safest one, the other method was pumps but it use power despite its just 4 to 8MW, the powerless method I propose was just put valve to the direction of the pipes that would get filled first in the manifold that way if you have extractor that was lower than the factory, the filed one will be full and wouldnt be able to back flow because of the valve and the extractor would just continue to fill the other side, you could also put valve on all 3 side if youre worried, another thing is to fill the input from above but it would sacrifice the uplift make sure you have less than 10m, youd always want the liquid to flow in one direction so avoid filling them from the the center like the vid shows there as much as possible, if not then just use one of the method shown or what I proposed there...
the only difference between a valve and a pump is headlift, they have the same functionality otherwise
@@siriuscowking Do pumps increase slight headlift or smt when not powered? I do know they act like a valve when not powered
@@key462 I recently tested this. I built a pipe connected to water extractor and a pipe that went up 12m from the centerpoint of the water extractor's output. Then I tested it with both valve and unpowered pump. With Valve the pipe filled fully while with unpowered pump, it only filled to about 60%. Though, I did not test,yet, what happens if the valve or pump were to be placed at 10-12m height.
@@joeljoonatankaartinen3469 Interesting tbh ive never use pumps as valve before, if i need them to flow somewhere I just use valve instead since they have smaller footprint, maybe it was that, since pumps have longer body most of the pipes it occupy were mostly occupied by pump body length the and continuity of liquid was probably prevented by that pump body or something, hard to judge unless you see the games code
In the first bit with the blueprint. It appears you made a flow compensator. The other bit at alum fac is a high and low priority splitter using gravity. Let me know. if right or wrong.
Why is there backflow if the system is never saturated? Won't the fluid only go in one direction?
even if the pipe is saturated, the liquid will flow in both directions unless going down a slope
@@siriuscowking Do you consider this a bug or more of a feature of the fluid system?
No unlike belts fluids will flow in any direction and accumulate at the lowest point. This is intentional, as it's low compute approximation of what liquids in pipes would do.
Incroyable, ça fonctionne merci beaucoup :D
Doesn't carrying the liquid up to a higher position than the manifold line by pumps then splitting it and connecting it to the both ends of the manifold fix the issues with 600m³? I mean do we really need this much?
that's not always an option or fit the design of what you're doing, if anything I'm doing that on a small scale
@@siriuscowkingYou may be right. I always design my factories around the feeding from the top principle so I wouldn't know
Do you have to solve the same backflow issue with gas ?
gas doesn't actually flow, it just sort of fills the entire system at the same rate, so as long as you have enough flow capacity you will never have any problem
With gases it is all about pressure. So, no.
The core of the solution seems to be to not use pipes for 100%, right?
Even with verticality and valve, that by itself would not prevent the stoppage from going up to the oil output. The principle of transmission is "can't flow forward", not "flowing backward"
a pure oil node maxed out is exactly 600, that is the only situation where it can't be avoided
what colour did you use for your factory
greyge: #D1B99D and #8B7A61
The problem with mk2 pipes isnt sloshing, its in how the calculates flow with floating point values. Theres nothing you can do to make a mk2 flow 600 perfectly forever. The best solution is to never design your factories in such a way that youre counting on a pipe to carry 600
The pipe coming out of your extractor needs to hit 600.
yet there is. They fixed the floating point values. I have multiple oil extractors which run at 600 cubic meters at 100% efficiency with 0 trouble if you remove sloshing and have a good setup.
@@joshuasteffanie947 coffee stain posted a video less than a week ago basically saying it will never be fixed.
@@FraterAlephyou got a link? Because this specifically says its fixed ruclips.net/video/ASfKo0dKBbU/видео.htmlsi=nIXOuorHdAImGZTH
@@FraterAleph which video says that because i've heard several times in stream that they specifically said it is fixed.
There is no issue with pipes, only ppl that doesn't understand how they work.
Fr. Heard about this "bug" for a while and was concerned until it happened with me...but with 300m^3/s in mk1 pipes. This same issue will happen
There are issues with how the code rounds up/down floating point numbers. Eventually it breaks perfect systems that use max 600m3 pipes. You can easily fix it by dropping the usage to 599m3 with an oil extractor working on 600m3 speed. That extra 1m3 fixes it. Everything else is a cope
j'aime ton split , pi la balance en de sous
I always have fluid input manifolds Above the connection point and looped instead of strait line and equal or below connection point. that way gravity is never working against flow. This way my input manifolds are essentially fluid buffers that gravity feed. Never had any issues with flow since starting to do it this way and always utilize max yield from oil nodes. And always make sure to prime your pipes, let them fill to max including input and output buffers of factories and mines before turning on production lines. If you turn on a production line without maxing fill capacity of the pipeline, you will never get correct flow even if you're producing enough
Hi! I watched this video about a few days ago and I was amazed by these designs. However, today I saw another video: ruclips.net/video/ZwO-F82sYE4/видео.html And I feel like the first design and the one you're using as an example at the aluminum build are the same. To my surprise, that's the one that failed the missing item test, while a valve surprisingly worked just fine. And even though I don't know how that second design works, it performed very well compared to all the other options. Is there a difference between the two examples or would your system have the same issue if it lacks the items for some time? Because in case it does have the same problem, it might be better to use a different method.
@@nebula600 the method I suggest in my basic liquid tutorial is the simplest way to deal with liquid byproduct, just throw it back in the loop. As long as there is no power shortage, there will never by a missing item input. On that same note, you should always have an awesome sink to deal with the overflow. If you are the kind of player who doesn't keep an eye on their power usage and also don't have a reserve battery, then you might want to consider using the liquid on a 2ndary production. That being said, making more power plants isn't difficult considering the amount of rocket fuel you can make for even just 150 crude oil, see my next video.
@@siriuscowking Oh don't worry, I would give ANYTHING to make my power chart 2 straight lines. Technically just for my OCD, but of course the good part is if something isn't at 100% efficiency, I can read that off from the power chart which would be AMAZING, if not for those things that consume/produce power in an oscillating way.😡
The next Albert Einstein
French Canadian accent > French accent, amirite boys?
@@deeteeohbee6794 montreal
The valves are not helping with fluid sloshing. The proper way to deal with the issue is to build a second pipe above the the one supplying refineries, parallel to it, and connect it to the ends of the feeding pipe. This way any sloshing you get will be above the input line, until it fills up, and while its happening the lower pipe will always be full anyway. If refinery gets blocked in the meantime, excess oil from the upper pipe will always act as a buffer.
I'm going to say this once and probably TWICE, STOP THE BACKFLOW with PUMPS. More sloshing means more backflow, that means the extractor knows that something is up and doesn't run at 100%. This is an issue with people that barely know pipe mechanics and they're a bit buggy but still. Learn to feed from above and keep things from going around and sloshing; key word i'm going to say... Valves are cancer in the game and I suggest just using pumps and gravity feeding things.
I like the ingenuity but my personal engineering skills (and game experience) want to screech with how much you made the thing 10x more complex with slapping valves down for liquids. (Yes valves do reset the headlift and that's why they're rarely used)
First, valves don't reset headlift. They used to, but not anymore.
Second, I completely agree that this is a way too complex of a solution. I just finished an oil refinery setup using up 600/s of crude oil from a pure node using just Mk2 pipes last week. No pumps, no valves. Also, the refineries are actually several meters *above the oil extractor* and the loop is actually vertical because I supply both end of the refinery pipeline from below.
The trick? Make a loop out of the pipe that feeds the refineries. As in, don't connect it to the middle, but junction the pipe that comes from the extractor and connect it to both ends of the refinery pipeline. Loops are very effective at making backflow a non-issue because in a loop there is no wall to reflect back from. The only time a loop would start disrupting the flow is when the refineries aren't actually pulling in the full 600/s.
the only difference between a valve and a pump is headlift, they have the same functionality otherwise
@@joeljoonatankaartinen3469 Yes valves still reset headlift. I had one do it two days ago
@@tabascosauce6490 RUclips either is being slow or ate my previous reply, so I'll try again with some expanded testing results.
I have 5 different test results here. The last two of them don't make too much sense, especially combined together, but are interesting regardless.
1. I tested a water extractor connected to junction followed by a valve and 10 meters of pipe straight up followed by a short horizontal pipe. Both filled up 100%.
2. I built an unpowered pump to the second port of the junction, followed by the same pipe structure as in test 1. The first part only filled up to 72%.
3. I powered up the pump, saw that the pipes filled fully, so I continued the pipe up to the maximum length possible. The last part filled up to 20l.
4. I then added a valve after the pump right after the 90 degree turn up. The fill rate at the topmost pipe dropped to around 15l.
5. Just for fun, I then added another valve a bit above the first one. To my suprise, the water level at the topmost pipe shot up to 100% , which is 70.4l. I tried adding one more pipe, but it only got like 0.4l after that. Though, when I removed the second valve and flushed the 70.4l pipe and rebuilt it, it only filled back up to 15l and stayed there. It only went back to 70.4l after I flushed it, so this is probably some kind of a bug.
So, it does seem like valves can adversely affect headlift when placed after pumps, but I don't think it's correct to say that they reset it.
@joeljoonatankaartinen3469 oh fun lol. Yeah I was using 400 out of 600 of a mk2 pipe in two different locations. Went to check the pipe and the area using 300/s was getting 600 flow rate somehow even after letting it set for a while...
So tried using a valve to restrict flow rate and seen it stopped flowing up the hill so had to add pumps. But my pipes were flowing atleast 40 or 50m up after the valve. Which was still way lower than the source it came from.
Eventually just had to split the mk2 pipe into two mk1s to fix the issue. Fluids are never fun to deal with
I can get 600 out of a node by just putting a pump after the extractor
oui oui
Issue has just been sent down the line. Adding head height above the intake is what solves it.
I would just use tanks
They don't fix the problems. Tanks are just big pipes. Your pipes already function as tanks. In fact, those specific tanks fudge the flow direction and headlift, so I wouldn't use them
@@AfePooq never had issues with oil, getting the full capacity. Tanks are needed to balance the flow.
You don't need any valves or pumps. You're doing something very wrong here.
Valves also don't stop back flow, fluid in front can still knock back fluid behind the valve even if the fluid units don't cross.
I explain in the video how the pipe being backed up will be felt behind the valve
@@siriuscowking you certainly said some things, but it's perfectly feasible to get 600 from 1 node w/o valves and usually without pumps. I honestly don't know what you're doing to your pipes to make this be an issue but you're free to get some pipe trouble shooting on the discord
I just use a buffer high up. Works for me.
I hate fluid in this game!! 😂
they are pretty simple once you understand the basics, check out my other video for that
Orrrrrrrrrrr, you set the two other valves at 300 each, problem solved without the extra pipe intersection....If you need a 200/400 devide, set 1valvue to 200, the other to 400. Not that difficult.
Ve says that he's had issues with valves when set to any intermediate value.
if your pipes are ever not full (VOLUME not FLOW) behind the valve, it will *not* send 300 m3 through the valve.
this is not about setting specific values, it's about making sure a full 600 flows out