Super-Compact Metal Volcano Tamer

[Zarquan]

I've been having a lot of fun with aerogel recently.  The creation of perfectly insulating natural tiles that can be made where normal tiles can't opens up a whole new world of possibilities in compact volcano tamer designs.  I also recently noticed that certain buildings and almost all sensors work while entombed, so I built myself a metal volcano tamer.

This tamer works using the freezing-in-mesh-tile teleportation mechanic often seen in magma volcano tamers.  The aluminum flows right when the volcano erupts, causing the liquid to come in contact with the steam room, which quickly makes it condense and merge with the metal on the weight plate.  The weight plate exists to ensure that no micro-quantities of metal end up on the line.

The line cycles through the steam room until it is below 140 C.  When this happens, the buried solid shutoff valve activates and runs the aluminum through the water cooling the steam turbine and dumps it onto a metal tile.

This design as-is does not produce usable metal as the volcano is erupting, as it produces too much heat during the active period.  In the dormancy, however, the steam room will cool and eject the material.  While the system is definitely net power positive (generating large amounts of power during the active phase), it should be connected to the grid for the dormancy phase, as the hot aluminum will heat up the steam turbine's water and it will need to be cooled by the aquatuner.

[blakemw]

That's pretty nice. I can't say I'm a fan of loops though, I always prefer to use a Conveyor Meter to just emit small packets of metal that don't loop, because they are small (like 0.5 kg) they reach thermal equilibrium instantly, and it's easy to counterflow them past the exhaust water Valved down to 1 kg/s.

For example in this case, 2 Aluminium Tiles of Counter-current heat exchanger (Aluminium Tiles and Aluminium Pipes) drops the temperature of the Aluminium to 102 C, while the 1 kg/s of exhaust water enters at about 150 C.

This is using the exhaust water to remove more of the heat from the aluminium instead of paying for the Aquatuner to do it. Using 2 tiles of heat exchange produces a significantly better result than only 1 tile of heat exchange, but even a single tile of heat exchange with the exhaust water will likely produce a better result than exchanging heat with the steam, as in the aluminium temperature will be reduced to something like 120 C instead of something like 160 C before paying the Aquatuner for further cooling.

I often use this technique and it's typically easy enough to integrate the 1 or 2 heat exchange tiles into the access lock so it doesn't even really increase width. And the Liquid Valve could be integrated into floors or walls by exploiting aerogel.

[Zarquan]

8 hours ago, blakemw said:

That's pretty nice. I can't say I'm a fan of loops though, I always prefer to use a Conveyor Meter to just emit small packets of metal that don't loop, because they are small (like 0.5 kg) they reach thermal equilibrium instantly, and it's easy to counterflow them past the exhaust water Valved down to 1 kg/s.

For example in this case, 2 Aluminium Tiles of Counter-current heat exchanger (Aluminium Tiles and Aluminium Pipes) drops the temperature of the Aluminium to 102 C, while the 1 kg/s of exhaust water enters at about 150 C.

This is using the exhaust water to remove more of the heat from the aluminium instead of paying for the Aquatuner to do it. Using 2 tiles of heat exchange produces a significantly better result than only 1 tile of heat exchange, but even a single tile of heat exchange with the exhaust water will likely produce a better result than exchanging heat with the steam, as in the aluminium temperature will be reduced to something like 120 C instead of something like 160 C before paying the Aquatuner for further cooling.

I often use this technique and it's typically easy enough to integrate the 1 or 2 heat exchange tiles into the access lock so it doesn't even really increase width. And the Liquid Valve could be integrated into floors or walls by exploiting aerogel.

Nice, I don't usually think about using the 1/10 pipe rule for cooling, but it makes a lot of sense here.  I would point out that yours is wider than mine by 1 tile, 2 if you count the chute, which I don't, as my output is "temporary" until I feel comfortable running it directly into my base.  But your design probably doesn't need a connection to the grid, as you clear out the material before the steam turbine runs out of power, so you probably save on energy.

One thing that you and I could both be doing is wiring this directly into our main grids though, as it is trivial to get the heavi-watt wire through the natural tiles and it is annoying to have a battery, transformer, and aquatuner on the same circuit.  I wish the adjustable transformers mod was part of the base game. 

You could squeeze in a 3rd heat exchange tile to the right of the conveyor.  It could hold the material waiting to go in to the meter from the loader if you moved the outgoing bridge up by 1.  Perhaps then you could run one half of the steam turbine water output through the upper tile, and the other half through the now-middle tile and through the new bottom tile, which would likely improve the performance of the system.

[blakemw]

1 hour ago, Zarquan said:

You could squeeze in a 3rd heat exchange tile to the right of the conveyor.

It would be overkill. 2 tiles is already overkill, it's just that 1 tile is underkill. Players have a strong tendency to overdo heat exchangers with metal volcano tamers.

[Zarquan]

8 hours ago, blakemw said:

It would be overkill. 2 tiles is already overkill, it's just that 1 tile is underkill. Players have a strong tendency to overdo heat exchangers with metal volcano tamers.

Overkill?  Perhaps.  Free in your build? Yes.  It would take no more space and could improve the efficiency of the system.  I personally try to keep my materials below 50-60 C, as I have a lot of ethanol liquid structures and I don't want it to boil.  I prefer for the temps to be in the 10-30 C range, as I often load it up in to rockets and don't want to stifle my decorative plants.

I will say that if 1 is underkill, then 2 is not overkill, as 2 is the next smallest number to something ineffective.

[Prince Mandor]

Side question: Sometimes I see my aerogel tiles (100mg of glass ) having temperature of surrounding environment. This is especially funny at line of them, where some is cold and some is warm, so it catches my eye on heat map.

Zarquan and blakemw, what is yours experience in this, why they may change temperature? As far as I can remember, objects with so little mass must not participate in heat exchange, but I see they are, and aerogel from ice doesn't work

[blakemw]

16 hours ago, Zarquan said:

I will say that if 1 is underkill, then 2 is not overkill, as 2 is the next smallest number to something ineffective.

But we are dealing with heat exchangers here, and a heat exchanger has theoretical perfection beyond which no improvement is possible. For a normal ("parallel flow") heat exchanger perfection is achieving the equilibrium temperature of the two flows, while for a counter-current heat exchanger perfection is swapping the temperatures of the two flows (at least at the end the more thermally massive stream enters), like water enters at 95 C, aluminium leaves at 95 C. Hence we can speak objectively of how perfectly a system is performing.

When a Conveyor Meter is used to slice packets, I often use 0.5 kg because Volcanoes tend to erupt at 0.3 kg, the small packets exchange heat very quickly. For a normal heat exchanger (e.g. with Steam room) then such a small packet of aluminum actually fully equalizes in a single tick if it goes into or on top of a metal tile (like aluminium has more than 400x the TC of Steam). So assuming you can find room for a single metal tile, even a single tile of heat exchange in the steam room is overkill, it's just 1 tile is the smallest you can have.

The very nature of countercurrent heat exchange means that heat exchange has to be done in multiple steps. Let's say for argument's sake the 0.5 kg/s Aluminium stream is 600 C, and the 1 kg/s Water stream is 95 C, in game this is how well a countercurrent heat exchanger performs in terms of final aluminium temperature.

• 1 Tile: 146 C : 90.3% of ideal (23 kDTU/s leftover)
• 2 Tile: 100.5 C : 98.9% of ideal (2.5 kDTU/s leftover)
• 3 Tile: 95.5 C : 99.9% of ideal (0.22 kDTU/s leftover)

So strictly speaking, there are gains to be had from a 3rd tile, but just 2 tiles are already bringing you to within 1% of perfection. And it takes about 2 watts to transport 1 kDTU/s of heat using an Aquatuner, and of those 2 watts, about 1 watt is recovered in the Steam Turbine. So if you are going to subsequently put the Aluminium behind the Steam Turbine to do a second stage of heat removal with the Aquatuner (or hypothetically might have to do some extra cooling at some point down the line), the 2nd tile saves like 20 watts, while the third tile only saves like 2 watts - this "2 watts" is a relatively objective way of valuing the third tile. It's possible for me to care about 20 watts even though I know I probably shouldn't, but caring about 2 watts is approaching the ridiculous. In fact I'd be quite happy with 95% of perfection (which is generally my "threshold of caring" when I decide a build is sufficiently optimized) which is why I say 2 tiles is overkill.

(Also, if the Aluminium first exchanges heat with the Steam, and then with a single tile with the Exhaust Water, it's also going to be 95% of the way to perfection).
 

7 hours ago, Prince Mandor said:

Zarquan and blakemw, what is yours experience in this, why they may change temperature? As far as I can remember, objects with so little mass must not participate in heat exchange, but I see they are, and aerogel from ice doesn't work

Generally the minimums only apply to heat exchange between cells, not for heat exchange with buildings or debris - and plants count as debris for heat exchange purposes.

The minimums which do apply to heat exchange with buildings and debris are related to floating point imprecision and it's difficult to give it "rules" because it's not developer defined behaviour, basically there has to be enough DTU in a single tick to cause the temperature to jump to the next representable floating point number, for both objects, otherwise neither object changes temperature. A small "donor cell" may not be able to transfer enough DTU in a single tick to change the temperature of a massive debris or building. Hence Aerogel cells may be observed failing to exchange heat with thermally massive debris though it'll exchange heat with smaller debris. This is less observed with buildings because buildings just aren't anywhere near as thermally massive as some debris can be (for a start buildings have their thermal mass divided by 5, and they are usually made of materials with low SHC. Like a Wheezewort is 400 kg of Genetic Ooze and has as much thermal mass as would a 7000 kg building made of Igneous Rock).

I admit I haven't examined the pipe:content heat transfer in detail but the way I recall it I think it's like debris in that sometimes heat exchange happens and sometimes it doesn't probably based on floating point imprecision.

[Zarquan]

On 5/17/2024 at 7:02 AM, Prince Mandor said:

Side question: Sometimes I see my aerogel tiles (100mg of glass ) having temperature of surrounding environment. This is especially funny at line of them, where some is cold and some is warm, so it catches my eye on heat map.

Zarquan and blakemw, what is yours experience in this, why they may change temperature? As far as I can remember, objects with so little mass must not participate in heat exchange, but I see they are, and aerogel from ice doesn't work

On 5/17/2024 at 2:21 PM, blakemw said:

I admit I haven't examined the pipe:content heat transfer in detail but the way I recall it I think it's like debris in that sometimes heat exchange happens and sometimes it doesn't probably based on floating point imprecision.

I recently had an experience with this, where I tried running pipes of volcano-sourced magma through aerogel, which immediately evaporated.  Things like pipes and buildings exchange heat, but they don't exchange heat with things like other tiles.  Now I am going to try to make my aerogel out of steel or tungsten.

@blakemw Based on your suggestions, I have upgraded my design.  Because I didn't want to burn my dupes, I still have the loop (which doesn't hurt), but turned up the eject temperature to a higher temperature (200 C).  After the loop I have a conveyor meter only sending 1 kg/s through a couple of metal tiles.  However, I believe mine has a minor improvement.  You are only using 1 of the 2 kg/s steam turbine water to cool the aluminum.  I am using both, which means that the hot aluminum coming out is interacting with metal tiles cooled by 95 C water.  This should result more heat being drawn out of the metal and a higher water temperature going in to the steam room, but I will not be able to provide any meaningful benchmarks until the volcano is active again.

Ignore the damaged pipes, I broke an automation wire and overcooled the water, but it is fixed now and hasn't received any new damage since.

Also, conveyor meters work while buried, so that's cool.

Spoiler

 

[blakemw]

10 hours ago, Zarquan said:

Because I didn't want to burn my dupes, I still have the loop (which doesn't hurt)

It doesn't make any difference whether you have a loop or not.
Since it's sometimes hard to understand very compact layouts, I've added annotations to the build I posted earlier:

There are 3 tiles of heat exchange.

1. With the Exhaust Water, dropping the temperature to something like 140 C
2. With the Exhaust Water, dropping the temperature to something like 102 C
3. With the Steam Turbine petroleum layer, dropping the temperature to like 60 C (or whatever you've set the cooling loop to) - only 1 tile is needed because petroleum is plenty conductive enough.

So even with only the very short heat exchanger, the aluminium is coming out as cool as you please.

I know there's an appeal to using a loop and measuring the temperature of the debris, but the reality is, a small piece of debris is going to be the temperature of the thing it passes through, no more, no less, no need to measure. Even before the Conveyor Meter was introduced I'd stopped using loops, I just used an exit Conveyor Shutoff connected to automation which limited the flow rate of the debris (e.g. Timer Sensor set to 1/39), holding a 20 kg piece of aluminium in place for 40 seconds, is as good as a 0.5 kg piece of aluminium spending 1 second in place, if I ensure the piece of aluminium is held in the steam room for 40 seconds, it's coming out at the steam temperature. Conveyor Meter is even more better than an exit Conveyor Shutoff, it's less bursty and more precise and less automation is required, like it's literally just plug a Thermo Sensor into the Conveyor Meter reset port and that's it, the single Thermo Sensor both limits the steam temperature and also controls the temperature the Aluminium leaves the steam chamber at, because it'll be leaving at steam temperature.

Loops work, but they don't work any better, and are much less minimal.

[Zarquan]

 

10 hours ago, blakemw said:

There are 3 tiles of heat exchange.

1. With the Exhaust Water, dropping the temperature to something like 140 C
2. With the Exhaust Water, dropping the temperature to something like 102 C
3. With the Steam Turbine petroleum layer, dropping the temperature to like 60 C (or whatever you've set the cooling loop to) - only 1 tile is needed because petroleum is plenty conductive enough.

I did understand your design, and I have a heat exchanger using the output of the steam turbine in my new design too.  Mine uses both legs of the 95 C water to cool the aluminum debris.  My aluminum is currently starting at 240 C and ends up with an output temperature of the aluminum being 99.2 C before entering the steam turbine cooling layer.  This has the water entering the volcano room at 108.4 C at the with the 240 C aluminum. 

The aluminum in my system leaves the conveyor loader, enters the now-redundant loop that I can't be bothered to remove, then goes up to the conveyor meter, then right and down through the two metal tiles each cooled with 95 C 1 kg/s  water (with aerogel natural tiles next to it), then through the two tiles, then through the cooling layer on the turbine (I don't have petroleum so I used water), then out to be used by the dupes.

I have the same cooling tiles as you, except I routed each of the two legs of the steam turbine valve through a different metal tile.  I could almost certainly do the same thing with fewer rails and without the loop, but my goal was physical size, not complexity.

I believe the loop in my design doesn't do anything other than to prevent aluminum from leaving the room above 250 C (new cap I set) to prevent my room from getting too hot.

[Saturnus]

Why so needlessly big? 

Obviously joking around a bit but if you are playing with aeroglass why not go all the way?

EDIT: I actually one-upped the build below to not require liquid on top of the build by leaving the 1840kg granite tile that is in the corner and making a 400kg steel tile instead of the metal. The granite tile serves as a massive thermal mass for the ST. And the 400kg steel is obviously better than 100kg at maintaining temperature when the 170c-ish metal rail block enters it. And added a cycle sensor to cut down of sweeper runtime. See image at end of post.

   

New image

[6Havok9]

On 6/4/2024 at 8:52 PM, Saturnus said:

   

New image

Wait.. the steam turbine foundation can overlap a volcano? Never thought of that and never tried it. The more you know!

[Saturnus]

On 6/6/2024 at 8:07 PM, 6Havok9 said:

Wait.. the steam turbine foundation can overlap a volcano? Never thought of that and never tried it. The more you know!

It's not something that would come up under normal circumstances but when making petrol well you can have the foundation of the ST overlap the top part of the oil well so I suspected it would work the same with volcano.

[Prince Mandor]

I find and lost this info numerous time. Now I lost it again, so can anybody help me? How you make vertical or horizontal aerogel walls over something? (it is easy to make it if you can build tiles to direct falling bottle, but if I cannot build tiles I fail)

I know there are some method to make vertical column of aerogel from below, I have several vents in my current gameplay covered in glass and dug in needed pattern, but I forgot how I did it. And my current experiments stack glass in one tile, so I obviously forgot some important point, and now I asking for some guidance

[blakemw]

3 hours ago, Prince Mandor said:

And my current experiments stack glass in one tile, so I obviously forgot some important point, and now I asking for some guidance

If you can't build tiles to direct the bottles you have to make aerogel tiles from the top down, as the bottles will get absorbed if they fall onto an existing arogel tile. Technically you could also use a different kind of aerogel like magma and glass.

[Prince Mandor]

6 minutes ago, blakemw said:

If you can't build tiles to direct the bottles you have to make aerogel tiles from the top down, as the bottles will get absorbed if they fall onto an existing arogel tile. Technically you could also use a different kind of aerogel like magma and glass.

I cannot even build two tiles horizontally, one bottle falls a tile from pipe, but next freeze at pipe height, always freezing them diagonally. I see here on last Saturnus design two tiles vertically and two tiles horizontally, all four without "cracks", this means there are no horizontal pipes was left inside. I obviously needs some step-by-step instruction, how horizontal part was made

[Occam Blazer]

3 hours ago, Prince Mandor said:

I find and lost this info numerous time. Now I lost it again, so can anybody help me? How you make vertical or horizontal aerogel walls over something? (it is easy to make it if you can build tiles to direct falling bottle, but if I cannot build tiles I fail)

 

[Prince Mandor]

Yes, thank you, I know how aerogel work. But you cannot build tiles over volcano, to define precise position of aerogel block. So I always get situation exactly as rightmost on this picture: two tiles diagonally from same pipe. It mysteriously never happens to be on same height

[Tigin]

2 hours ago, Prince Mandor said:

Yes, thank you, I know how aerogel work. But you cannot build tiles over volcano, to define precise position of aerogel block. So I always get situation exactly as rightmost on this picture: two tiles diagonally from same pipe. It mysteriously never happens to be on same height

At the very least, you can completely entomb the volcano first and then dig it out. It seems to work reliably for me on normal speed though

[Prince Mandor]

1 hour ago, Tigin said:

you can completely entomb the volcano first

Yes! How? How I can entomb volcano entirely? I done it several time a year ago, but cannot remember how to do it

[Tigin]

2 hours ago, Prince Mandor said:

Yes! How? How I can entomb volcano entirely? I done it several time a year ago, but cannot remember how to do it

I don't think there's anything special? Dump sand/regolith, use solid-to-solid transitions (temp shifts or rails), use bottles of anything and cool them down. This isn't really necessary since emptying the pipes where they need to go top-to-bottom works for me to get the right placement

[Prince Mandor]

8 hours ago, Tigin said:

emptying the pipes where they need to go top-to-bottom works for me

But don't work for me. Two tiles of aerogel stays diagonally if emptied from same horizontal pipe

[blakemw]

On 6/14/2024 at 9:41 AM, Prince Mandor said:

But don't work for me. Two tiles of aerogel stays diagonally if emptied from same horizontal pipe

I don't understand what is difficult, so I made a video:

 

It even went perfectly the first time I tried to take the video.

Perhaps it helps to run the simulation VERY slowly, like here I am pausing as the empty job completes. The ONI simulation is super jank and doesn't have a well defined "order of events" and some things happen "when the simulation gets around to it", hence there can be difference in behaviour based on game speed and simulation load.

[Prince Mandor]

9 minutes ago, blakemw said:

I don't understand what is difficult, so I made a video:

...

It even went perfectly the first time I tried to take the video.

Perhaps it helps to run the simulation VERY slowly, like here I am pausing as the empty job completes. The ONI simulation is super jank and doesn't have a well defined "order of events" and some things happen "when the simulation gets around to it", hence there can be difference in behaviour based on game speed and simulation load.

 

Thank you for your time. For me It is always first tile over pipe, second tile one tile below pipe and never in horizontal row for hours of experiments. Exactly like rightmost set on this picture

 

I will try it next time on slowest simulation speed with pauses and see what happens. Idea to slow down simulation to increase predictability of result just never come to my mind. So, my experiments was at x3 speed in real late-game base

 

Also, I again rediscovered a way to build glass towers. It needs 8 solid tiles of not-glass material in 3x3 square without one side tile. Aerogel pipe going through central solid tile of this square (I usually just build 5 tiles under neutronium geyser floor). Now pipe in center of this 3x3 square must be emptied. First bottle come to side, filling last tile of 3x3. But this don't interrupt emptying task already in progress, and next bottles spawns on top, forming vertical tower of any height.

This allow placement of tile at any needed position over anything (as long as we have 2 tiles height somewhere below, so not at bottom of world or rocket room)

 

[Saturnus]

Is there a problem leaving the pipes inside the tiles for the verticall aeroglass tiles? Just reuse them for the cooling loop, or ignore that they exist