mathmanican

  • Content Count

    1416
  • Joined

  • Last visited

 Content Type 

Profiles

Forums

Downloads

Klei Bug Tracker

Game Updates

Hot Lava Bug Reporter

Everything posted by mathmanican

  1. The NG leaves the contraption at a higher temp than the input oil. So the NG preheats the oil some (not a ton, but adding 50C to the incoming crude is not negligible).
  2. The save game is already in the original post. Just open the spoiler under the first picture to see all the overlays and the save.
  3. So, with one port of a steam turbine open to 600+C temps, and a second port open on another side sucking in mg of 100C steam, You can actually use the inside of a steam turbine as your cooking plate. The turbine can run an aquatuner full time and still generate heat. The resulting supercoolant can not only freeze the cooling plate, but also leverage your cooling along the SG loop. I will probably have to use two different loops as I don't want to freeze things along the SG loop, but with this in tow, I'm going to try to 1/4th the size of this build. I'll probably fail, but might as well try.
  4. I got the inspiration from your various builds. Thanks for sharing them. BTW, I decided to reenable the door on the right side of the steam chamber. Basically, the SG at the contact plate is in the 170s, which means any heat that bleeds through to the steam turbine will get deleted and only 1/10th will be exhausted through the turbine itself. However, the turbine is cooled by the same supercoolant that freezes the methane, so this nets me 90% savings in cooling potential. The liquid tepidizer has to fire more often, but the turbine generates way more power than the tepidizer uses, so it's a win in pretty much every way (except perhaps the incoming crude now has a slightly lower input temp as the NG leaving is slightly colder. I want to see if I can get this thing to boil 20kg/s +, and need more cooling potential without another aquatuner, and the steam turbine amplifies the cooling potential. More boiling basically just requires a little longer shaft. A new problem emerged. The cooling is now too good. The methane does not want to return to gas form. I was worried about this with the original build, but never had the issue. Now it just builds. Nothing a tepidizer can't fix though. The last thing you want is the SG above this room to freeze while in transfer (had that happen yesterday). So you want methane gas transfer on the conveyors (the left most conveyor just interacts with tiny beads of methane liquid, not an issue). I'm pretty sure now that I have enough cooling potential to crank up the volume by quite a bit. Fun times.
  5. The door approach has a nice side advantage. In the event of a crude oil shortage, the boiler region will not be sucked to vacuum. I swapped to this, and will probably never go back. There is one other possible issue that may need planning around. In the unlikely event that petro forms over the vent, it can get trapped. In that case, all the crude instantly moves sideways. It hasn't happen in over 100 cycles with this current version, but something like it happened with a past version. Adding one tile above the vent, so that any petro that forms can move up, should stop this. With the extra space I have in the middle of my redesigned version, I'm adding this fail safe in.
  6. Do you mean like this. I just built it, got on to post it, and saw your reply. The sensor is set to 0kg. If the bead ever stops being there, the door instantly closes, and then reopens. No starter vent needed. The air tile below the vent can (and should) swap back to insulation.
  7. Have you tried a solid door and a pressure sensor below it to detect this? The mesh tile wont be needed anymore.
  8. I'm back with some more toys. A single-liquid bead pump that will never over pressurize. A bypass pump that will never overpressurize (and works faster) A 10kg/s, single aquatuner, crude to natural gas boiler that could handle even more (yep @suxkar you can do this with one aquatuner, and even have extra heat and cooling). All three are in this one picture. The counters kept track of how many times certain things occurred over 50 cycles. (643) Each time the supercoolant that freezes the SG returns, if it could not shed enough cold then it has to be heated up. The liquid tepidizer fired 643 times, every time the pool got colder than -167C. This means there is actually more cooling power than is being used, and the tepidizer has to inject heat to make sure the aquatuner runs at 100% capacity to boil oil. This could be replaced with a regular water buffer zone, and a liquid storage tank, but it was easy to test as is. (76) If the aquatuner gets too hot (a thermosensor is set at 800C), then a loop of hydrogen on a 60 sec buffer takes the extra heat from the boiler room to the steam turbine. An alternate approach could be to inject more oil, as the boiler plate could handle it. (1) If the aquatuner gets too cold, then the oil is shut off. This never happened. Here is a single-liquid bead pump. The flow rate can be increased to any amount you want (up to max flow which is 250kg/s for crude), by adding more vents to the right. The spiral boiler can't handle 250kg/s, but the bead pump will allow it. The pump must drop liquid off the left side (see here for more details). The airflow tiles on the left are needed to force the air that bubbles up to the left. Without a place for the air to go, the pump still beads, but about half the liquid also drips to the bottom (ruining the entire bead pump boiler). You can see another option by @TripleM999 here. I chose this configuration as it allowed me even more thermal contact with the exiting NG. Here is an example of a bypass pump that will never over pressurize. Rather than releasing liquid directly above the crude, release the liquid to the side. This lets you keep a small amount of liquid over the vent, while the tile above the crude still does the bypass swapping with huge amounts. The current pressure in this room is almost 7000kg, and rising. The pump also works a tad faster, as you get to swap elements every other tick, instead of once every 5 ticks. Or, you can set the valve really low and get swapping to happen once every 10 seconds. You can essentially choose how fast you want the flow to happen, provided you pay attention to single-layer liquid flow rules. Enjoy. I'm currently wanting to improve the heat exchange between NG and incoming crude oil, and figure bubbling the NG through crude would be a fun way to do it. I'll share a screenshot when it's done. I also think a redesign of the entire middle region is in order, to improve aesthetics an remove unused pieces.
  9. Well @suxkar, I made a few modifications and got it up to 3500g/s crude with a single tuner. I'm pretty sure this could go up quite a bit more, if I made the boiling chute a bit taller so that I had to waste less heat on flashing the crude to SG. The diamond tiles in the SG/NG chamber brought my SG temp down almost 100 degrees before hitting the cooling plate, so definitely that's important. As for my original cooking area, I swapped it out for a 1000kg/tile steam room for the aquatuner, topped off by a diamond plate. I think this was the biggest improvement. Using molten lead and letting sour gas form in the steam room was less ideal. My guess is that this approach enables the heat deletion bug we see with turbines. As for the turbine I deleted, I'll probably put it back in. It only runs in the event of an oil shortage, and then protects the aquatuners. Everything else in the build gets sucked to vacuum if you turn off the oil input, without any breakage. Here is a last image. I'll probably start a new thread if I post any more.
  10. 10g per sec min on left vent. Try 1000g per sec on right. That will get it working fine. The left vent is a starter vent. It only needs to output minimal at start Flow numbers on crude are 10 times water. My hope is that this experiment makes you want to dig deeper and answer why. I'd love another liquid tamer on the forums.
  11. A dismal 2000g per sec right now. The NG leaves the loop at 340C so heat transfer is working. I use this to preheat 77c crude to about 160. I need to work on the boiler plate section. The single tuner can clearly cool more.
  12. With water the numbers I gave work fine. Crudes viscosity requires higher values. The numbers needed are in the waterfall thread. Let me know if you need more.
  13. The liquid vent and mesh should be deleted in that spoiler pic. Also my exchanger might be better with every other tile metal. My forte is liquid mechanics not heat exchangers. This is my plan for tomorrow.
  14. I've been playing with boilers the last few days. Figured you would like a new toy to play with. You can build a bead pump, with a single liquid, that will never overpressurize (skip the EZ-bead if you might overpressurize). Then you have no such limiting factor. You can go crazy, and then the only limiting factor is the viscosity of your liquid. I'll just post a picture for now. If you want more details, feel free to ask. The shuftoff valve is not used in this picture. The first valve is set at 200g, the second at 30. You can see way more details here. Just try to fail to make a waterfall and instead end up with a bead pump. You can add as many liquid vents as you want to this, to the right of the current vents, to increase the flow. So if you want to have a 200kg/s flow rate (max flow is 250 = 5* viscosity), just have 20 vents, followed by a 100 tile+ heat exchanger (yeah you would end up with massive amounts of gas in the exchanger - so 2*200*100 = 40000kg of sour gas in the biggest tile, but oh well). Or stack them up in domino fashion to the right and above the right most vent. All kinds of crazy configurations are allowed. Have fun. I'll probably make my own post soon with more details, like a bypass pump that never overpressurizes as well. This is useful for getting the NG out of the exchange loop (which you are doing), but your design will stop working if the top chamber ever exceeds 1000kg. Here's a pic for that one. I'm guessing you can figure out the mechanics. Supercoolant would have been a better choice than petro, but oh well.
  15. Now you're on to something. Last year I put a compilation post together, to make finding information easier in a year. It's about time to resurrect it.
  16. These two comments seem contradictory. I'm sure you two can figure out which is right. I would build my generator on mesh tiles, so the building cannot interact with the ground, and liquid can flow down. There will be a brief 0,2s tick where the outgoing water interacts with the building, insta freezes, and the building heats up more than it should, if the 40C min is right. The CO2 coming out as solid does not surprise me. There was a lot of debate the month or so before release about the changes they made, and I haven't seen any changes announced (though I have been much less active).
  17. You can insert a thermal buffer, kept around 130C or so, beneath this overheating pump. Just place it on the far left side of the vacuum space. Keeping 1000kg/tile of steam behind some steel walls on the top and and left side, will help rapidly cool the exhaust. What you are missing, next to each turbine, is something that you can actively cool for lots of cycles while waiting for the next launch. Something as small as 1 tile of high pressure steam, trapped behind walls, would do wonders at lowering the temperature of the new steam exhaust. Reinjecting 1/4 of the steam you are trying to capture will do a little to mitigate this, but that's only 500g/s per turbine, as opposed to a 1000kg buffer (so 2000 times more mass).
  18. If that is true, then we reintroduce tons of ways to cool your base, that are way too OP. Seriously, if you can convert -180C NG to -180C PW (should be simple to test), then a google site search of the forums will show you all kinds of crazy ways to make massive amounts of cool stuff, at very little effort.... Last I read in the release notes was that outputs had a minimum temperature, and then output at building temp. If the minimum has been removed....
  19. This is my guess. I cannot confirm without running a test, but I'm sure you could do that as well. I believe the water comes out at a minimum of 40C, otherwise we return to -180C NG generators creating massive cold sinks.
  20. See @TripleM999's post on gas flow that is linked above. I wonder if 1 tile wide passageways would work up/down. In @TripleM999's post he has 1 tile wide left/right channels for the SG/NG exchange. I bet you could do something like it vertically. Always important.
  21. Unfortunately, @NurdRage deleted the post he made where he explained this, otherwise I'd link it. Liquid beads fall down the tube, and need something that holds lots of mass to transfer heat to them. Temp shift plates ruin the process. So instead, you use a liquid vent as a place holder to absorb heat. Cold beads falling are heated up by this 1 by 1 heat absorber. The rising hot gas is cooled down. Every 0.2 seconds this exchange takes place, and it's extremely efficient. Essentially, the liquid vents are there to hold heat in a specific tile, without transferring it to adjacent tiles. Perfect for a liquid bead boiler. Have you played with bridges (wire, automation, liquid, etc) connecting the sour gas and methane regions, in addition to your plates. You clearly want the two regions to transfer heat. I wonder if adding these bridges would help you shrink the overall height by quite a bit.
  22. Depends on whether or not you play using ONI's actual physics engine, or want to pretend and force yourself to play the game following whatever you believe actual physics is (you can't play the game with actual physics, as there is no ideal gas law for one). Your choice. Play the game the way you want, and keep whatever rules you want to abide by. For me, I'll exploit the crap out of ONI's physics engine as much as I can. (Isn't that what we do in real life, exploit the crap out of the laws of nature to get them to do as much for us as we can.) I assume you meant you would have to place it at the bottom (forgot "except" at the bottom). If you want it at the top, then I see plenty of places to put it along the sides of your silo. In all cases, you'll need a large cooling buffer (next section). You can use a liquid/solid buffer under the rockets (or along the sides of your silo) to absorb the huge amounts of heat. The rockets put out a constant DTU, and you can soak this up to keep temps low enough to enable the bypass pump. What I'm envisioning is a diamond/steel tile room full of >= 1000kg/tile steam to soak up the heat, directly under your rocket or along the side walls. The bypass pump can be placed next to, or under, the buffer room, so that hot gasses have to pass by this buffer (and cool down) to get to the bypass. You can easily handle the temps with petroleum. Steam turbines can keep the entire buffer room under 200C with ease, if you don't want to add extra power and have aquatuners actively cool this room. There are tons of options. If the bypass is too exploity for ya, then freeze everything to solid, steam and CO2. Power is not really an issue (if it is, then build a Crude->NG boiler). Then build a buffer at the top of each silo, lined with steel and filled with anything that you actively try to keep near 0K (or whatever keeps the buffer from solidifying or breaking walls - go with super cold 40000kg/tile hydrogen gas if you want). Freeze everything that tries to leave the silo using these buffers. If any gas tries to escape the top of the the room, it will insta-freeze. Use tempshift plates and doors to extend the reach of the cooling zone so NOTHING makes it out of your chamber. No bypass used here. If you don't like high pressure gasses, then don't use 40000kg/tile. Adjust these ideas to fit your gameplay style. @Zarquan ran an experiment once to see if he could capture ALL the gas coming from space, and he did it with ethanol (not super coolant). If you don't have a surplus of supercoolant, I'm sure you could use PW and keep the buffers full of hydrogen at around -10C (enough to liquefy any steam trying to leave). If you don't want the CO2, then just let it leave. With this approach, you just need a liquid pump at the bottom of your silo to slurp up cool water, or a sweeper arm to move away the ice. No bypass needed. Just keep your entire silo as cold as possible, at all times. I see that you have >=5 tiles of space between rockets, which means you can plop a steam turbine down between them, so you can enable even more aquatuner cooling of a super chilled buffer zone (between turbines). Using @Saturnus's trick for cooling steam turbines in a vacuum, you can put as many of these turbines as you need going up the walls of your silo. I'm sure there are many more ways to do this. I tend to enjoy using fluid mechanic solutions, rather than throwing power at it. There are lots of options. If/when you get one finished, please share. I'm sure others would love to see what you've done as well as me. If you try several options, please share them all.
  23. Bypass pumps might be the key. Build them a tad higher up so the temp is not as hot. They can work fine with 2 gasses. The cost is 10mg per sec of any liquid, basically nothing. Once the gasses are sucked from the chamber you can then manage them. You can build a natural filter with tile placement (such as a co2 lock and hydrogen/steam lock) and ignore thermium pumps. Dispose of the co2 if you dont want it or feed it to slicksters.