|
Dunno-Lars posted:In a previous game, I placed a block of diamond tiles under the rockets which I then circulated hydrogen through into a steam chamber. I had a bypass valve to keep the temperature under 200*C inside the chamber. Free power and "cooled" down all the tiles to a nice and cool 800*C on average. Had fewer melted tiles after I installed this heat spreader-cooling system. Yeah, that sounds like a much better solution. Mine was just the biggest, dumbest, loudest one that fell easily to hand. I was thinking about it after I went to bed (yes I'm normal shut up) and a heat recovery system sounds ideal. Self powering space equipment. and such. Also yes, diamond and obsidian is a must with regular launches.
|
#
?
Aug 15, 2022 12:38
|
|
|
|
| # ? May 30, 2024 14:22 |
|
|
Or just let stuff get hot. If you don't have a lot of steel for whatever reason you can use tungsten for all your wires & loaders/unloaders, since you can just get it all back later if you need to make thermium with it.
|
|
#
?
Aug 15, 2022 14:53
|
|
|
Does anybody know what the range on drecko feeding is? Do they have to be on the same tile as the plant or could they be on the ceiling and take a bite out of a mealwood growing below them?
|
|
#
?
Aug 15, 2022 15:33
|
|
|
Dreckos need access to the tile directly I think I've always done my petroleum boilers with either a metal refinery or aquatuner. The thing I always read about the magma powered ones is that unless everything works perfectly always, you end up with with the magma solidifying inside the mesh tile and backing up the system. And sometimes it just happens anyway invisibly.
|
|
#
?
Aug 15, 2022 16:47
|
|
|
The only boiler I have managed to make work consistently is the diamond spike in the core. It's also relatively simple to build. I use Nilaus' video for this one: https://www.youtube.com/watch?v=OS14NAgT_RU
|
|
#
?
Aug 15, 2022 16:58
|
|
|
If you have a lava source above your system (actually this works when pumping too) use the mesh 'condenser' like shown here: https://www.youtube.com/watch?v=srzPi5El8Qs If you also control the inlet magma (instead of directly from a volcano like he does) you can regulate the temperature of the buffer easily, so it can sit at like 600-700C, then you don't have to spike the temp in your boiler chamber so much and you'll have less trouble with cracked pipes. If you have a large pool of lava already you should use a single-tile-high dropper driven by a thermo sensor, so that you don't flood it too fast, which will cause solid tiles to form inside the mesh. I've never had a problem with one of these since switching over. If you want to get fancier you can put a thermium loader setup inside the buffer chamber to recover the hot debris into a regular steam chamber. Also I suspect that using gaseous supercoolant will equalize the heat more rapidly than the steam w/ lead pool.
|
|
#
?
Aug 15, 2022 17:25
|
|
|
Rescue Toaster posted:Or just let stuff get hot. If you don't have a lot of steel for whatever reason you can use tungsten for all your wires & loaders/unloaders, since you can just get it all back later if you need to make thermium with it. that's what i had been doing. continuous missions with two rockets (both with boosters) means poo poo starts melting
|
|
#
?
Aug 15, 2022 17:34
|
|
|
I use a thermium aquatuner for my petroleum boiler. I will use magma and pitcher pumps to pre-heat it though. Otherwise it takes ages for the boiling chamber to come up to temp.
|
|
#
?
Aug 15, 2022 17:35
|
|
|
I broke into the oil biome for the first time recently. I shaved away some of the abyssalite between the oil and magma biomes for aesthetic reasons, and I noticed that the 'hot' abyssalite was producing some petroleum and sour gas. I didn't want to deal with anything odd while I figured oil out so I covered the heat with some tiles, but how does this work? I thought abyssalite wouldn't transfer any heat.
|
|
#
?
Aug 15, 2022 17:50
|
|
|
Sokani posted:I broke into the oil biome for the first time recently. I shaved away some of the abyssalite between the oil and magma biomes for aesthetic reasons, and I noticed that the 'hot' abyssalite was producing some petroleum and sour gas. I didn't want to deal with anything odd while I figured oil out so I covered the heat with some tiles, but how does this work? I thought abyssalite wouldn't transfer any heat. It's a mechanic called "flaking". https://oxygennotincluded.fandom.com/wiki/Flaking
|
|
#
?
Aug 15, 2022 18:00
|
|
|
it is always advisable to build a little insulated vestibule and vacuum it out when getting into the core or any magma pool
|
|
#
?
Aug 15, 2022 18:14
|
|
|
abyssalite was also changed at some point to have higher thermal conductivity iirc
|
|
#
?
Aug 15, 2022 19:04
|
|
|
I was always a huge fan of the “counter flow” style boiler since the end product petroleum is the same temp as the starting oil. I assume that still works?
|
|
#
?
Aug 15, 2022 19:44
|
|
|
pokchu posted:I was always a huge fan of the “counter flow” style boiler since the end product petroleum is the same temp as the starting oil. I assume that still works? Works fine, and once up to temp it requires just the tiniest dribble of input heat. Petroleum has a slightly higher SHC than oil does, so the steady-state output temp ends up like 20C hotter than the oil input temp.
|
|
#
?
Aug 15, 2022 20:06
|
|
|
Panty Saluter posted:that's what i had been doing. continuous missions with two rockets (both with boosters) means poo poo starts melting But tungsten doesn't melt until 3400C? Unless the wikis are massively out of date even the hydrogen engine won't exceed 3000C in the direct heated path (as opposed to the steam cloud, which is much cooler near 2000C). Steel can definitely melt especially wires running directly under the engine. But tungsten shouldn't... unless something has changed. For the tiles, Francis John showed a trick in his last playthrough where you drop like 1kg of coal on each tile spaced under the rocket platform, and the steam heats it into solid tiles of refined carbon, which also will not melt from a hydrogen engine.
|
|
#
?
Aug 15, 2022 20:20
|
|
|
I've definitely been caught in the trap of converting all my tungsten to thermium because it feels like it should be a strict upgrade. But now I know to save a bunch for space construction.
|
|
#
?
Aug 15, 2022 20:42
|
|
|
oh jay posted:I've definitely been caught in the trap of converting all my tungsten to thermium because it feels like it should be a strict upgrade. But now I know to save a bunch for space construction. Convert a ton of it back to niobium too, and use that for wherever you had copper or iron ore in your core base. It has a huge decor bonus.
|
|
#
?
Aug 15, 2022 21:03
|
|
|
Rescue Toaster posted:But tungsten doesn't melt until 3400C? Unless the wikis are massively out of date even the hydrogen engine won't exceed 3000C in the direct heated path (as opposed to the steam cloud, which is much cooler near 2000C). Steel can definitely melt especially wires running directly under the engine. But tungsten shouldn't... unless something has changed. No, I meant i had been ignoring the heat until it started melting things. I have no tungsten right now because there is basically no wolframite on this asteroid
|
|
#
?
Aug 15, 2022 21:05
|
|
|
You can probably cool it all by adding some backing plates (can't remember the name) and then release some gas there to soak up the heat before disappearing into space. Basically create a flow of co2 or oxygen past the wires and stuff that is melting
|
|
#
?
Aug 15, 2022 21:36
|
|
|
OzyMandrill posted:Part I Interlude: Theoretical Physics I do very little calculating in this game. Most of it can be done with a few basic ratios (dupes consume 1k calories/day, 100g/s = 60kg O2), and the principles of wing it and see. Most of the time you have overpowered tools and physics breaking machines so it doesn't matter, but when building devices that rely on the ONI physics engine, there are some real world principles that work, because the underlying heat transfer mechanisms do roughly approximate the real world. So, why do we counterflow heat? What's the point? For avoiding unnecessary complexity, why the insistence on direction being so important. So, consider a simple case of 2 things that flow, that we put through a pair of pipes next to each other, to the materials get to exchange energy. The amount of energy changed in one second depends on the thermal conductivity, and the temperature difference. Lets assume thermal conductivity is the best we can do, so we dont care what the actual value it, it will be the same at all points in our machine. Case A: Flowing together, at the start, the temperature difference is maximum, so the rate of change is maximum. By the end, the hot is just above medium, and the cold is just below medium, the difference is quite small, so even adding another few sections of exchanger won't change the outcome much: Hot -> Medium Hot -> Warm -> Slightly cooler Cold -> Medium Cold -> Cool -> Slightly warmer The best we can do is somewhere close to the middle, but never quite meeting Case B: Counterflow streams, so the hot source will meet the warmed output of the cold stream, but still a decent delta. At the other end the hot is cooled down, but the cold input is at its coldest, giving a good delta for heat transfer. Hot > Medium > Warm > Cool Warm < Cool < Medium cold < Cold Here we can cross over the midpoint, and output the hot stream at a colder temperature than the middle, and the same for the hot, all due to the counterflow. Applied Physics Putting this together, we can develop better ONI machines. And by better, I mean less materials, more reliable, more efficient. But, especially in stuff like petrol boilers, magma is such a powerful source that it easily outpowers the heat needed to boil the 10kg/s oil. But, if it is not efficient, then the outgoing petrol will not exchange much heat with the incoming oil, and instead of getting petrol at 100 degrees, you get petrol at 150 degrees, and that difference in heat is literally magma heat that you are wasting and exporting around your base where it will need to be dealt with. I was going to save that bit as a surprise for anyone following along - the downside of these boilers is you introduce a BIG heat source into your base that will need extra cooling if you don't have a decent amount already. But, with a boiler you have the kW to power it now. Part II : Petrol boilers The basic idea is the oil goes in one end, gets heated to ~390 degrees, and drops into a hot pool at about 405C. As long as you maintain this pool at 405-410 then the oil will change to petrol at a 1:1 ratio. This has to happen outside a pipe though, but then we can just let liquid physics move the petrol past our pipes to cool them . That's the theory and that gives the designs as seen above. Simple flat corridors, which works ok, but has a significant issue with efficiency (outputs hot petrol) and the problem is that two neighbouring tiles of liquid will share thermal energy at 625x the standard rate of transfer. This will dwarf the flow rate, and means that along a horizontal corridor, there is not much drop in temperature. You can beat this with a taller stack but this comes down to a fundamental design issue of heat exchangers: Heat exchangers work best with isolated thermal regions. It should settle down to (total temperature delta) / (nRegions) as the approximate spread, and the closer the regions are to the neighbour, the better/more efficient the system. So, the more separate thermal areas that the flows cross through, the better. One of our flows is liquid as cells. This will then exchange heat with any pipes we encounter, and the surrounding solids. And any other buildings that cross the cells the liquid flows through. The more mass we can throw in the way, the better, as that facilitates the thermal transfer. And there are very useful 3x1 buildings that you can use for this - bridges! Conductive power bridges are even made out of refined metal so work really really well to spread heat in a 3x1 area, but even stone bridges help add mass which all adds up in helping to exchange heat between the flows. Someone on reddit did some tests and made a nice waterfall one that was optimal, but I don't have the vertical space, and requires getting the petrol back into a pipe, so I combined that with the simple Francis John style to get a stepped waterfall:  I warned you it was going to get weird. So, the breakdown is this:  The petrol will be generated in the little cup on the top. I doubt it needs to be much bigger than this, tho it can be deeper if you want. I have 2 igneous rock tempshift plates, so 1.6 tons of slow changing mass in there, so it should be fairly slow to change temperature. The heat exchanger is then made of alternating 'plates'. The green blocks are where the liquid falls down a tile and lands on a nice aluminium block. There's an igneous block giving some slow thermal mass behind it, and all the bridges I can fit going up/down in this little 4x1 block. This should directly link the 4x1 area into a single thermal unit that will be one temperature. Next to it is a blue 'spacer' column, The air tile is to isolate the step below thermally from this step and stop the backflow of heat that can reduce your performance. The plumbing diagram shows how the oil will run up, with conductive just in the areas where there is solid to transfer heat with. The bridges follow this pattern:  and you can squeeze in power, ventilation, and transport along with the respective pipes (to stop those annoying red warnings!), and to add almost 100kg more mass per tile for heat exchanging, as well as working like a vertical shift plate and sharing heat from the flow path with the thermal mass of the regular tiles that are included. The blue columns particularly are just thermally isolated tiles using the bridges to 'dip' into the petrol stream as it passes below. The idea is that each step should have equivalent thermal performance to an entire row of a FJ style boiler, and I hope to pack about 7 or 8 of these pairs into this one. My previous build of this design snaked back and forth a bit, but here I have the space for a neat version. I also like that it is 2 high by design, so naturally accessible for cleanup after the inevitable 'accident'. Automation-wise will be 2 temperature sensors in the little well. One will control the door to the magma crucible to maintain temp in our little diamond scoop of > 407ish) and the other will control the oil flow (only if temp > 406ish). The only other control will be a limiter valve on the oil input to match what we can realistically supply. While you can run this design in bursts, they work much more efficiently when running constantly, with a nice steady flow, and a thermal equilibrium established. This only works if you limit the oil at the very start - before it enters the boiler. If you stop oil in the boiler, it can 'cook' to petrol in the pipes near the end, which causes element damage. If you run the oil when the pot is not hot enough, you get it into the heat exchanged (bad!) and it doesn't mix with the petrol, causes a 2-high blockage and can cause the system to back up. All this is avoidable is you just turn off the oil before it even enters the very start of the system. Empty pipes won't break. Anyway, that's enough for now, and I haven't finished the build, so hopefully update 3 will be finishing, and starting up, and we can see how close my rough guesses match up! OzyMandrill fucked around with this message at 23:01 on Aug 15, 2022 |
|
#
?
Aug 15, 2022 22:07
|
|
|
I think you can get the same effect with just a staircase. Each square is isolated from the neighbors, they can't exchange heat diagonally. In the past I tried making weird isolated metal tiles to exchange two counterflows, and it ended up just eating up boatloads of refined metal for no real gain over the staircase. Here's the staircase I built for my water distiller/power plant. The important part is it drops to the left. If you have stairs to the right you will delete heat by accident. I think it's the opposite if you did a staircase with the cold crude flowing down and the hot petroleum being pumped up, but that would only work with a thermium pump, but should be doable? I'm curious if something could be done with having the crude and petroleum slide over the top of each other in direct contact, since heat exchange of two liquids is incredibly fast without the hot liquid being in the pipe. I might have to do some experimenting with that, although it would probably require thermium pumps too. And by the time you have thermium you can just make a thermium aquatuner & supercoolant for your petroleum boiler, and the absolute efficiency barely matters.
|
|
#
?
Aug 16, 2022 13:34
|
|
|
Yeah, it's basically stairs but packing 1 ton+ of thermal mass into each step so it can cope with full 10kg/s rate in a short run. Didnt know about deleting heat, eh, as long as I get cool petrol I'll take it (tho they have fixed a lot of the funny bugs like that), I vaguely remember seeing a weird one on reddit that did both oil and petrol as liquids in a corridor design, oil flowed down and petrol up I think. Something about bead pumps, and it worked really quite well, but a bit fiddly. Possibly bugs on load/random lockups, but again - maybe those have been fixed. It was a while ago but looked awesome. I hope they do more with the nuclear stuff. The research reactors cool, but I want to melt the uranium ore in molten salt and have it react when over a certain weight in tile, so I can build my own reaction chambers and cooling systems.
|
|
#
?
Aug 16, 2022 14:08
|
|
|
It's been years since I've played this so clearly things have changed. What do I need to do to get the polluted water into the water sieve? 
|
|
#
?
Aug 17, 2022 16:53
|
|
|
Mayveena posted:It's been years since I've played this so clearly things have changed. You got the input and output switched. Also, a toilet/sieve loop generates a surplus of water, so you'll want something to take it up (people usually pipe it to a farm tile with a thimble reed or something
|
|
#
?
Aug 17, 2022 16:56
|
|
|
HolHorsejob posted:You got the input and output switched. Also, a toilet/sieve loop generates a surplus of water, so you'll want something to take it up (people usually pipe it to a farm tile with a thimble reed or something Thanks! drat it HAS been a long time 
|
|
#
?
Aug 17, 2022 17:01
|
|
|
I was getting inputs and outputs wrong for at least the first thousand hours 
|
|
#
?
Aug 17, 2022 18:19
|
|
|
I feel like a whole ux class could be taught just on ONI source/sink presentation.
|
|
#
?
Aug 17, 2022 19:13
|
|
|
white and green are not the colors I would have chosen. I get them mixed up hundreds of hours later.
|
|
#
?
Aug 17, 2022 19:34
|
|
|
it does have the sidebar thing telling you what the input/outputs are whenever you place something at least, but it's the fact they flipped the colors a while back that gets me
|
|
#
?
Aug 17, 2022 20:02
|
|
|
Bhodi posted:white and green are not the colors I would have chosen. I get them mixed up hundreds of hours later. The up and down arrows are also not visually helpful. Though I tried searching for icons for source/sink producer/consumer, etc... and nothing really obvious jumped out.
|
|
#
?
Aug 17, 2022 20:16
|
|
|
Rescue Toaster posted:The up and down arrows are also not visually helpful. Though I tried searching for icons for source/sink producer/consumer, etc... and nothing really obvious jumped out.
|
|
#
?
Aug 17, 2022 21:04
|
|
|
Khorne posted:One is pointing in & the other is pointing out. -> ] in, <- ] out Yeah but on some machines the in arrow points away from the machine and the out arrow point towards it. When I finally internalized it, it was from remembering the colors and not looking at the arrows.
|
|
#
?
Aug 17, 2022 21:32
|
|
|
Sokani posted:Yeah but on some machines the in arrow points away from the machine and the out arrow point towards it. When I finally internalized it, it was from remembering the colors and not looking at the arrows.
|
|
#
?
Aug 17, 2022 22:19
|
|
|
Khorne posted:It's not based on orientation relative to the machine. The "]" portion of it signifies the machine. If an arrow points toward the "]" in the picture it's an input. I understand that, it just makes it unclear especially at a glance.
|
|
#
?
Aug 17, 2022 22:53
|
|
|
Rescue Toaster posted:The up and down arrows are also not visually helpful. Though I tried searching for icons for source/sink producer/consumer, etc... and nothing really obvious jumped out. Yeah agreed, the colors are kind immaterial to me. A better icon would have been like ➡0 or 0➡. imo
|
|
#
?
Aug 17, 2022 23:06
|
|
|
I got over 4k hours and I still get it wrong. I have to look at a bridge cos they have arrows on them to remind me which end goes where.
|
|
#
?
Aug 17, 2022 23:11
|
|
|
Even with the arrows on the bridge I install them backwards 9 times out of 10.
|
|
#
?
Aug 18, 2022 00:16
|
|
|
Wish there was a mod to label them more clearly. I understand why they chose that way I just don't agree with it
|
|
#
?
Aug 18, 2022 00:54
|
|
|
Green. Go. Out.
|
|
#
?
Aug 18, 2022 01:18
|
|
|
|
| # ? May 30, 2024 14:22 |
|
|
 Wiki says mealwood doesn't need fertilizer? So what's going on? Thanks for any help.
|
|
#
?
Aug 19, 2022 16:34
|
|