Factorio Forums

Can someone explain to me, why some of boiler lines have only around 65C, while rest is close to 100C ?
I redid "low temperature" lines, without success.
To me this looks like bug, but maybe i'm overlooking something.

First i noticed that, when instead of 30MW, 6 lines were not enough to supply 24MW (since 2 lines were at 2/3 of nominal output due temperature)

note, this is vanilla save (not single mod).

You're problem is that you're not forcing the water to route through the boilers, but giving it a path around. Boilers are completely flow-through, and don't actually consume water (that's what the steam engines do). Change the lines to the setup below, and things are just fine.



That said, I'm honestly confused that you're not experiencing this problem on all the lines, but eliminating the water bypass fixes the problem easily enough.

Mehve wrote:You're problem is that you're not forcing the water to route through the boilers, but giving it a path around. Boilers are completely flow-through, and don't actually consume water (that's what the steam engines do). Change the lines to the setup below, and things are just fine.



That said, I'm honestly confused that you're not experiencing this problem on all the lines, but eliminating the water bypass fixes the problem easily enough.

i'm using this setup as preliminary change to what comes in 0.15, since according my testing, you have 100C water either you draw water from boiler line sidewise (new setup, that will be required in 0.15, but works well in 0.14), or pass trough (regular current setup).

since you checked my savegame, you know all lines are setup same way (eg. 0.15 like), yet top lines have nearly 100C and 5/6 line has only 62-63C

i'm not asking for FIX but for explanation, there is no reason, that all lines should not behave same way.

btw. i have encountered similar issue even before with normal "passtrough" boiler line, but redoing line fixed it (eg. some quirk in liquid handling system), in this case redoing line doesn't automagically fix the issue.

Yeah, I suspected that 0.15 was the reason for the setup

Alright, I might have something. It's weird, but it seems pretty repeatable. It's been documented in the past that the order you lay pipe down has a tiny, miniscule effect on straight-through flowrates. Well, I completely erased and redid the bypass lines on the problem lines, making sure to lay down the pipe in the direction of water flow. The result? Of the three <70C lines, two of them are fixed by simply replacing and re-laying the bypass pipe in the "correct" direction. The third one was fixed after replacing all the piping, including the boilers.

I know this game assigns an ID number to everything that's laid down, so it's quite possible that the order of the ID number influences the priority that fluid is transferred to/from?

Mehve wrote:Yeah, I suspected that 0.15 was the reason for the setup

Alright, I might have something. It's weird, but it seems pretty repeatable. It's been documented in the past that the order you lay pipe down has a tiny, miniscule effect on straight-through flowrates. Well, I completely erased and redid the bypass lines on the problem lines, making sure to lay down the pipe in the direction of water flow. The result? Of the three <70C lines, two of them are fixed by simply replacing and re-laying the bypass pipe in the "correct" direction. The third one was fixed after replacing all the piping, including the boilers.

I know this game assigns an ID number to everything that's laid down, so it's quite possible that the order of the ID number influences the priority that fluid is transferred to/from?

sounds, well weird. but i suspect there is a quirk in fluid system.

like i said, i noticed this even in past and redoing fixed that.

guess i'll try to redo complete bottom part of boilers 1 line by 1 line, beginig with broken line and will see.

and if it's reproducible (your guess might be correct), then i will report that as a bug, because it's more bug, than just "quirk".

ok, seem your theory about "build order" is correct.

following build orders gave different temperatures ("output pipes" are those sidepipes from boilers)
pump - pipe - boilers - "output pipes" - pipe to generators - temperature is 100C
pump - pipe - pipe to generator - "output pipes" - boilers - temperature is 92C
pump - pipe - boilers - part of "output pipes" - pipe to generators - another part of "output pipes" - temperature is roughly 62C

while fluid system has some quirks i noticed few times already, this seems kinda ridiculous and imo should be considered as bug.

i made a vid to illustrate how can build order affect temperature - quality is not best, but should be visibile - messed up setup is at end of vid.
https://www.dropbox.com/s/1notmujlgto8g ... 3.flv?dl=0

It's a known issue:
viewtopic.php?f=41&t=31917

kovarex wrote:The thing is, that the water simulation depend on the order of updatable entities, which is related to chunks and the order in which it has been built.
And also the water flow logic is kind of a "magic box" that "kind of works" and we are afraid to touch it.

Someone might do a proper solution someday.

Choumiko wrote:It's a known issue:
viewtopic.php?f=41&t=31917

kovarex wrote:The thing is, that the water simulation depend on the order of updatable entities, which is related to chunks and the order in which it has been built.
And also the water flow logic is kind of a "magic box" that "kind of works" and we are afraid to touch it.

Someone might do a proper solution someday.

i will add link to this there. since this will need to be fixed prior 0.15, otherwise it might screw up boiler setups, considering additional steam pipe requirement.

imagine, you will start for example with 2 boilers, but make long enough line to fit additional 12 later (if ratio remains same), then added boiler will not only have "lower" effectivity, but if placed in wrong order, they might even decrease effectivity of your whole setup.

I'm not sure how much of a bug it can be considered. You're trying to make a setup that will accommodate a different boiler that outputs something that it doesn't currently (steam). Yes, we'll have to reconfigure things come 0.15, but that's just part of the game sometimes, and likely won't be for several months yet. In the meantime, your setup is allowing the incoming cold water to completely bypass the boilers. Nothing is keeping you from keeping the spacing and pipe offset, so you can just add a few pipes later on to make things keep working.

If it makes you feel any better, my main map has a 2GW steam assembly that I've built so compactly that I KNOW I won't be able to simply change the pipes around to keep it working.

Mehve wrote:I'm not sure how much of a bug it can be considered. You're trying to make a setup that will accommodate a different boiler that outputs something that it doesn't currently (steam). Yes, we'll have to reconfigure things come 0.15, but that's just part of the game sometimes, and likely won't be for several months yet. In the meantime, your setup is allowing the incoming cold water to completely bypass the boilers. Nothing is keeping you from keeping the spacing and pipe offset, so you can just add a few pipes later on to make things keep working.

If it makes you feel any better, my main map has a 2GW steam assembly that I've built so compactly that I KNOW I won't be able to simply change the pipes around to keep it working.

if you look at topic choumico linked, it can even happen with "straight" line of boilers (no sidepipes), which i already observed few times, but replacing boilers always fixed it.

if you take in account people might (and do) use bots and blueprints, then you don't have linear placement of component in blueprint. so depending whether fluid system is built upon placement of BP or upon placement of actual part, might also affect flow. and if "disordered" placement of straight pipe parts can affect flow, then something is wrong with fluid system ... it's bug, not just quirk.

The quote from Klonan "it kinda works" describes the current fluid system best. Build order, orientation, position and possibly unit_number all seem to play into update order of flow, pressure and temperature.
When testing GotLags Reactor mod it drove me nuts how some would constantly output 100° while others only 98.98°. Turned out it was build order of pumps. When building across 2 chunks you can either loose or gain massive amounts on hot water that way too. I'd be surprised if it was any different with base boilers.

I wonder with the introduction of non-flow through boilers, new pumps, and fluid train wagons in vanilla we will see a better implementation of the fluid flow and how it works. What worries me is the introduction of 'heat' pipes as the workaround to re-working fluid to have it make sense...Any update on this front would be nice.

slaphappypap wrote:I wonder with the introduction of non-flow through boilers, new pumps, and fluid train wagons in vanilla we will see a better implementation of the fluid flow and how it works. What worries me is the introduction of 'heat' pipes as the workaround to re-working fluid to have it make sense...Any update on this front would be nice.

i doubt new system will fix issue with disordered pipe placement and therefore out of order update of fluid level. if anything it will bring more headache, rather than solving problem.

as long, as fluid is processed and updated with each piece of pipe (instead of one fluid box per chunk for pipe), this issue will repeat, although there is enough topics on forum covering this issue, so people at least can find how to "fix" it.

mexmer wrote:

slaphappypap wrote:I wonder with the introduction of non-flow through boilers, new pumps, and fluid train wagons in vanilla we will see a better implementation of the fluid flow and how it works. What worries me is the introduction of 'heat' pipes as the workaround to re-working fluid to have it make sense...Any update on this front would be nice.

i doubt new system will fix issue with disordered pipe placement and therefore out of order update of fluid level. if anything it will bring more headache, rather than solving problem.

as long, as fluid is processed and updated with each piece of pipe (instead of one fluid box per chunk for pipe), this issue will repeat, although there is enough topics on forum covering this issue, so people at least can find how to "fix" it.

I believe fluid flow (as well as belts) are on the docket to be worked on for 0.15, so I think we might see some fixes to the current oddities of the fluid flow system, but I understand it's not an easy problem to fix, especially to fix in a manner that will be efficient enough for it to remain efficient at the scales this game allows for.