Volcano's coolant keeps evaporating when off-screen?

[my_hat_stinks]

The water I'm using to cool a copper volcano keeps evaporating at random, but only when off-screen. When I load an autosave and play it back nothing goes wrong, when playing at normal or 3x speed. Since I don't have a chance to see what actually goes wrong I have no idea how I'd go about fixing it.

After playing through the last cycle again:

I originally had only one aquatuner, lead radiant pipes, and clean water in the pipes. The first time the issue cropped up some of the pipes melted, so it must have somehow reached over 300 degrees.

I switched to polluted water for the extra temperate range and copper pipes which seemed to help for a while, but that didn't last.

With the second aquatuner the volcano now easily resets to below -5°C between eruptions, spiking to a little over 70° when the copper is conveyored through the metal tiles. This lasted 40 cycles this activity period without breaking, plus more from before it went dormant.

 

The same cooling loop is used to cool both the volcano and the steam turbine, but the temperature resets with about half a cycle to spare between eruptions so I really doubt that's the issue. The steam room bounces between around 140° to 180°.

The conveyor loop is slightly too short so some hot copper does sit in the loader for a bit. The loader has yet to overheat so that's probably not an issue, it's only the pipes that seem to break.

I've attached save files if anyone else wants to take a stab at it. One save from after the pipes blew, the autosave before that, one after replaying that cycle, and one older save before I reworked the cooling (more prone to breaking). save_files.7z

The only thing I can think of is that heat is calculated differently if it's not on screen and that's for some reason occasionally causing my pipes to blow.

[mathmanican]

When you have two gasses in a room, weird things can, and do, happen.  It would not surprise me if a single blob of steam gets surrounded by hydrogen while over the high pressure vent you have there, and then gets deleted in the next tick (I've seen stuff like that, even if the vent doesn't have anything coming out of it - @Lifegrow illustrated this back around launch if I recall correctly.) Try deleting the vent you have in there, and see if it fixes things. 

Another issue could be the conveyor bug with conveyors of screen (was this fixed yet). 

Personally, I would probably remove all the hydrogen, and just use water and steam, nothing else. That way you have a monogas environment, which is much more predictable.    

[PhailRaptor]

Without knowing exactly how much water is there, it's hard to say for certain, but at north of 2 kg/tile Hydrogen, I'd have to agree with mathmanican's assessment, the water is evaporating to steam, and the pressure differential between the hydrogen and the steam is deleting the steam.

[my_hat_stinks]

There's no water in the chamber until the pipes burst, it's the boiled polluted water that was in the pipe.

[mathmanican]

1 hour ago, my_hat_stinks said:

There's no water in the chamber until the pipes burst, it's the boiled polluted water that was in the pipe.

When your volcano erupts, there will be a huge temperature difference between the hot metal and the pipes.  This will instantly cause the pipes to shoot up in temp, and then they will boil the liquid inside.

You need a thermal buffer on the lower level of your volcano. An entire layer of water (1000kg/tile) would help.  Without this, your pipes will keep breaking. A liquid layer in constact with liquid metal receives a 625 multiplier on thermal transfer, so you definitely want a liquid layer to help buffer the hot metal. Increasing the hydrogen content will be nowhere near as effective as adding a layer of water. 

I thought the water you had on the ground was this buffer, and your buffer kept vanishing.  Since that isn't the problem, then just add a thermal buffer (layer of water).You can make it two tiles deep, and up to 500kg on the third level, without affecting the volcano flow. 

[my_hat_stinks]

The temperature drops below 0 between eruptions so water wouldn't work, but I guess I could try crude oil.

I really don't see what difference it would make though? I can disable the cooling system, stop the pipes flowing, or both for a full eruption cycle and it doesn't overheat so I'm not sure how adding some liquid would help. Again, the pipes only overheat off-screen, and not reliably. As long as I'm looking at the volcano it lasts indefinitely, peak temperature is around 70° and the temperature completely resets between eruptions with plenty of time to spare.

You mentioned there was a conveyor bug, do you have any details about that? The hot copper is looped through metal tiles until a thermal sensor detects it's cold enough to leave (I have this set to 20° a the moment), but since the loop is slightly too small there's a bit of hot copper behind a bridge and in the loader that has to wait for the loop to clear.

[mathmanican]

1 hour ago, my_hat_stinks said:

but I guess I could try crude oil

You will end up with sour gas and petro eventually in very small quantities. I'll open your save soon and look.

[mathmanican]

17 hours ago, my_hat_stinks said:

what actually goes wrong

This could be your problem. Can't say for 100% sure (as it never broke on me during several cycles off screen or on screen). 

This gap of liquid keeps recurring. If it occurred at the wrong time in the wrong spot, then the radiant pipe would have no buffer to prevent it from raising temp quite rapidly.  The next bit of water that enters could vaporize? I don't know if the flaking mechanic applies to pipes, but conduction could still possibly do it. 

You also have intermittent power outages, causing backup periodically on your aquatuner. This doesn't cause water to stay in the loop, so I don't think that is the issue.  I could not reproduce the breakage though, so could not dissect the problem. Sorry. 

 

[my_hat_stinks]

That could definitely be it. There's an alternate passive route for cold water that feeds back into the main loop, I had to account for the chance only the first aquatuner would be active so there's a shortcut to ensure there's no gaps. Since I was using two bridges there I didn't properly measure the long route; the bridges would cover any gaps anyway.

There's up to a 4-pipe gap there which normally isn't an issue, but if the water switches to the active route at just the right time there wouldn't be any water to fill the gap, and since the route is slightly longer the gap reaches the main loop. It's definitely possible that a 4-tile gap appears in the main loop and hits the volcano at just the wrong time.

I've adjusted the passive route so it's the exact same length as packets running through both aquatuners, this should ensure there's no gap. We'll see if it explodes again over the next hundred cycles or so.

 

You're right about the brown-outs. I hooked up natural gas generators for the first time then linked all my power systems (previously powered by hamster wheels) just in time for all three geysers to go dormant at the same time. I went a little overboard with electrolysers so there's always fuel for the hydrogen generators, plus some hamster wheels for emergencies. The aquatuners are offline a lot of the time anyway so it's not a big deal if they lose power for a moment; if it backs up the water keeps flowing through the passive route so at least the heat doesn't concentrate on one tile.

Edit:

Well that didn't work, it exploded while I was tabbed out writing that post.

Here's the latest two autosaves, from before and after it blew. Again, I see no issues if I play again from the previous autosave. Maybe it's a bug related to the game not being in focus? My screen wasn't over the volcano and I was definitely tabbed out since I was writing this post. auto_save.7z

[mathmanican]

15 minutes ago, my_hat_stinks said:

Well that didn't work, it exploded while I was tabbed out writing that post.

If the tuner backs up, and backs up your overflow, then your pipes will definitely burst. You could remove the entire pipe section to the left, put in a liquid reservoir, and have more than enough liquid inside to make sure your loop stays 100% full. My best guess while you did the write-up is that your entire system got stuck (grid lock), which could easily happen.  A single liquid reservoir instead of the pipe spaghetti to the left, would resolve this (and let us reject it as the problem).

Another option, since your have over 2kg/tile of hydrogen, is to just put a layer or two of 1000kg polluted water (left over pee).  It will buffer the output of the volcano, and even empty pipe segments won't result in bursting. Grid lock would still eventually cause a problem, but you would have several cycles to see the gridlock, rather than a few seconds.   One time, @Saturnus ran a build for 50+cycles with no hitches.  He saved the file, shared it with me, and I opened to grid lock because of piping on tuners.  Since then, there have been many threads on how to properly prevent gridlock. Have you examined them? Brownouts will stutter your setup. 

[my_hat_stinks]

I've stopped the flow before by removing liquid bridges while trying to diagnose the issue, if the water's not moving at all it can still survive one eruption (though probably not two) without bursting the pipe. Liquid reservoir's a good idea though, I should've thought of that. Definitely going to use that.

[mathmanican]

6 hours ago, my_hat_stinks said:

You mentioned there was a conveyor bug, do you have any details about that?

Using a google search such as "site:forums.kleientertainment.com off screen conveyor" should get you lots more. 

The fix could be as simple as moving your bridge to be INSIDE the metal tile, instead of outside of it. 

[my_hat_stinks]

Could be related, but I'm not sure it's the exact same bug. The northeast/southwest thing could match since that's where most of my base is, but they mentioned their bug prevents heat transfer and my issue is too much heat. Still, it's pretty close so I'll re-arrange my conveyors and get back to you if it breaks again.

[mathmanican]

20 minutes ago, my_hat_stinks said:

but they mentioned their bug prevents heat transfer and my issue is too much heat.

The bug doesn't prevent heat transfer. It changes where the transfer occurs, and all of the heat transfers in one spot. Head down a tad in that thread and you see one person complaining about how their crude changed to petro because of the bug. Basically, imagine what would happen if all of the heat from your entire rail were concentrated in one spot, rather than spread out.  If that one spot is a gas tile of low pressure hydrogen, then it could get really hot really fast. The start point of your broken pipes seems to be the second time your pipes pass the tiles attached to your bridge.  

This is why putting the bridge in a metal tile would help mitigate the but.  Heat transfer is really fast in metal tiles.  Add a tempshift plate to spread the heat out more rapidly, even better.  Or put the bridge under a layer of 1000kg of polluted water, and you won't have any problems. If you want to confirm it is the conveyor bug, try the metal tile approach and let us know. If you just want to get it working and not worry anymore nor track it down, then adding a large thermal buffer of polluted water will help (though you will probably see the water convert to regular water over time. Toss the hydrogen, add regular water (which converts to steam and back to water if needed), and adjust your temp sensors to 15 instead of -5, and you should never have an issue, even with the conveyor bug.

[Artorias36]

Maybe this setup can help you. I always had two issues with iron or copper vulcano, they either create a lot of heat and the refined metal is always too hot to build something in your base. With this setup all that heat is turned into energy (but is not energy free and needs to be connected to a real power grid) and your refined ores can come out at max 30-40 degrees.

You need steel or everything will break. You also need to analyze the vulcano so that you know when it will erupts. You cannot build while it erupts or all will break. After filling the first square high of water, use a water lock to create a vacuum. Vacumm is important or this will not work properly. Behind the vulcano there are 3 tempshift plates of diamonds, so that, in vacuum, the iron will get solid immediately. The upper valve at the left of the turbine is in vacuum too and is set on 1000 grams, to avoid the  phase changes of water / steam.

Typical cooling loop. You can use water too but you need to set the thermo sensor to 20 degrees to be sure, with polluted water you can go down a bit more.

 

The conv behind the vulcano are made of steel, not sure if other material can do, but i prefered to not risk. There is a timer of 40 seconds to let drop 20kg of metals per time. The timer must be adjusted based on how much metal per second the vulcano shoots.

 

A simple filter gate to block the aquatuner in case the temp is reached.

The setup works as follow. The vulcano erupts iron at around 300-400 degres. It should take around 2 or max 3 eruptions to turns all the water into steam. The iron will go into the serpentine and already lose a lot of degrees by touching the 4 metal tiles. Then it will be cooled by going into the turbine where to cooling loop is. When the water turns into steam, the turbine will release just 1000 grams of water, so the water will slowly cooldown the metal tiles and won't break the pipes.

If you want more power, leave the iron inside by adjusting the timer, but as i said before this is just to get normal temp refined metal and not to create exceeded power. The first metals may come out at around 70 degrees, but the others will go as down as 30, or even more if you use polluted water.