Taming Volcano, Steam Turbine, Heat Problem (Exploity?)

[goatt]

I am always scared of volcano, this is my first try. Here is the demo design. My original write-up of this topic was not good enough, so I edit it now to add more elaboration.

Demo in motion in spoiler

Spoiler

Since steam turbine and volcano combination along with stone hatch are straight forward, the main problem I'm trying to solve is auto-sweeper / robo-miner / shipping loader's overheaing problem.

Conventional cooling is to use a background wall plus a liquid/gas medium and coolant piping to cool those machinery.

Here is an experiment in spoiler, simulating magma and outside world

Spoiler

from left to right, in the rooms, are

||1645 C magma 2000 kg ||700 C H2 60 mcg  ||20 C H2 60 mcg with tempshift   ||20 C O2 2 kg |

||1645 C magma 2000 kg ||700 C H2 600 mcg ||20 C H2 600 mcg with tempshift ||20 C O2 2 kg |

As can be seen, middle room which filled with low pressure H2's temperature gets affected by 2 kg O2 on the right, instead of 2000 kg magma on the left.

The O2 room stands for normal environment outside, and magma room simulates the magma room. We can use this model for taming volcano.

@mathmanican 's explanation on this:

4 hours ago, mathmanican said:

If you keep the amount of gas small enough, then heat transfer calculations get rounded to zero (perfect insulation).  It's the same idea as a chlorine clamped steam turbine.  A tempshift plate won't interact with the robominer if you're in a perfect vacuum, but it will if there is any gas.  The gas will take on the temperature of the robominer and the walls (with the walls having more mass and controlling the temperature more). With tempshift plates outside as well, you keep the walls nice and cool, which keeps the robominer cool. The gas enables the wall and robominer to share temps via the tempshift plate (the hydrogen doesn't really do anything other than turn on temp sharing of the two objects).  The heat from the volcano never makes it up to the robominer, as no heat actually gets transferred through the hydrogen.  I'm guessing you could use about 30mg of chlorine to achieve the same thing as 30 mcg of hydrogen (haven't tested this exact build). 

 

The machinery in volcano room will have the same temperature of metal walls, but receiving nearly 0 influence from thin gas in volcano room, which is nearly equivalent to perfect insulation, as seen in gif above.

 

More detail in spoiler (mistake: 80 mcg in the picture is not low enough)

Spoiler

- volcano area's gas pressure must be near vacuum, but can NOT be vacuum. The example I showed in video gif clip was about 80 mcg less than 50 mcg, but the lower the better. This will allow auto-sweeper, robo-miner and loader's temperatures to be affected largely by metal walls' temperature instead of H2's.

- there is a mistake / improvement in the above picture. I should remove 3 template shifts outside volcano room, which will decrease volcano room's influence on outside temperature.

- If initially you have a really really low gas pressure, such as 30 mcg, the result will be even better. My volcano room looks hot cuz it had 100 mg H2 in the beginning.

As @Xyer pointed out, the hatch room shouldn't be vacuum. Yes thank you for pointing out the mistake. The auto-sweeper should be installed against an ouside wall with thin gas in the room.

As @mathmanican pointed out, this is happening because of calculation rounding. So is it exploity?

In my opinion, it's not. Since the temperature increase is so small, as small as 0.0000001 C/s for estimation, if calculation counts it, the temperature will rise 1 degree every 1000 cycles, this is really trivial. That's why I think it's not really exploity exploit, it's close to fair calculation.

[goatt]

4 hours ago, nakomaru said:

I recommend a perfect vacuum, 100g of oil touching your equipment (feeder sweeper too), and a tiny loop of any liquid pipe that touches the oil and the outside.

I'm not sure what you mean. how to make oil touch sweeper in the air? You mean build a small block below sweeper as liquid holder? In my setup, it is actually "perfectly insulated". It doesn't look like perfect cuz I made a mistake and was too lazy to correct it since it's just a demo. I'll probbaly update one when I feel like it. But in other words, it's insulated and requires no extra cooling.

[goatt]

@nakomaru I realize I didn't give enough detail. I'll rewrite this topic later. But my way works very well. I'll try to prove it later.

[Xyer]

Wouldn't the sweeper in your ranch eventually overheat due to being in a vacuum?

[mathmanican]

9 hours ago, nakomaru said:

I would call magma→hydrogen→tempshift→metal quite the opposite of perfect insulation. 

 

7 hours ago, Xyer said:

Wouldn't the sweeper in your ranch eventually overheat due to being in a vacuum?

 

8 hours ago, goatt said:

I realize I didn't give enough detail.

Let me see if I can explain what @goatt was missing.  If you keep the amount of gas small enough, then heat transfer calculations get rounded to zero (perfect insulation).  It's the same idea as a chlorine clamped steam turbine.  A tempshift plate won't interact with the robominer if you're in a perfect vacuum, but it will if there is any gas.  The gas will take on the temperature of the robominer and the walls (with the walls having more mass and controlling the temperature more). With tempshift plates outside as well, you keep the walls nice and cool, which keeps the robominer cool. The gas enables the wall and robominer to share temps via the tempshift plate (the hydrogen doesn't really do anything other than turn on temp sharing of the two objects).  The heat from the volcano never makes it up to the robominer, as no heat actually gets transferred through the hydrogen.  I'm guessing you could use about 30mg of chlorine to achieve the same thing as 30 mcg of hydrogen (haven't tested this exact build). 

Perfect insulation occurs when there is zero heat transfer. This occurs if you make sure the gas content is low enough that heat transfer computations are rounded to zero. Fun little byproduct of our numerical approximation to heat transfer, and digital rounding errors from floating point number computations.  This exploit is not really fixable without watching your fps drop from 40 to .00001 (had to do a rounding error).   

Here's a link to the original chlorine clamped turbine build, if you want to read more about this exploit. 

Spoiler

 

 

By the way @goatt, this is an awesome build. It's a simple way to deal with robominer cooling.  Have you tried chlorine, as it is much easier to obtain perfect insulation with chlorine than with hydrogen. 

[goatt]

@mathmanican you explained well. I’m sick and having flu from my roommate, feeling so bad can’t do anything. Thank you for the help.

I haven’t tried chlorine, I thought all the gas can work, and thin any residue gas out in volcano room. But it  will definitely try it next time.

[goatt]

9 minutes ago, nakomaru said:

Thanks for catching on to the approach he was going for. This is indeed a great use of the rounding error to let only tempshifted tiles interact.

Thank you for making me realize I needed to improve the topic. I've added some example. Hope it helps.

[goatt]

@nakomaru I totally agree it's "exploit" of real world physics. But I was talking about game decimal rounding. I think we can agree that game physics is just game physics. But game-wise, you also calculated the entropy of heat transfer medium whose result was near 0, and it's a game physics exploit, not because of a game player's choice, but because of game's physics implementation. But we both can agree upon some game mechanisms are breaking real world science.