Does anyone have any idea of how exactly the thermal dynamics work?

[ithilelda]

Recently I have been fiddling with the tepidizers and aquatuners. To be honest, I am seriously confused by the thermal dynamics in this game.

first scenario: I surrounded a copper ore aquatuner with abyssalite tiles and painted the enclosed area with vaccuum. P-water is transfered through abyssalite pipes too. Thus all heat would be retained in the aquatuner itself in theory. The aquatuner was at 20 degree C initially, and after each cycle of cooling down 10kg of p-water, it raises exactly 9.1 degree C.

From the calculation of reason:

14 degree K difference of p-water * 10kg * 6J/g/K (the specific heat of p-water) = 840 KJ (C and K have the same scale, so when talking delta, delta C and delta K are equal.)

Heat from aquatuner: 58.51 * 20 * 1 sec = 1.17 KJ (machines produce 20x their tagged heat)

Total: 841.17 KJ

Thus the aquatuner will raise: 841.17 KJ / 1200 Kg (the mass of the aquatuner) / 0.386 J/g/K (the specific heat of copper ore) = 1.816 degree K

However, in reality it raises 9.1 degree K, around 5 times the anticipated amount of delta K. Anyone able to explain?

 

second scenario:

I submerged the tepidizer in the following setup:

the clock was set to turn on for only 1% of the cycle, which equals 6 seconds. The tepidizer was made of copper ore and was at 20 degree C. There are two tiles of naphtha, each with 700Kg mass and 20 degree C temperature. The 2 tiles on the right are vaccuum tiles. After the run and I waited for the system to settle down, the tepidizer and naphtha are both around 24.2 Degree C (avg. over 3 runs). The delta is 4.2 degree C.

The heat produced during these 6 seconds: [1400 Kg * 2.191 J/g/K + 400 Kg (mass of tepidizer) * 0.386 J/g/K ] * 4.2 degree K = 13.531 MJ

Thus the heat output power: 13.531 MJ / 6 sec = 2.255 MW

Due to control delay and roundup errors, we can see that the number is roughly half of the expected number (20.32 KW * 200 = 4.064 MW). It means that the tepidizer divides its output evenly among all four tiles, and outputs 1.016 MW of power to each tile it is on. If it is vaccuum, it will not heat it. This is pretty much conforming to what everyone else has known before. Unlike the aquatuner, the heat in the system is actually correct.

Here comes the interesting part:

Everything was the same, except that the two liquid tiles were moved left, and there were 3 tiles of vaccuum on the right. I tried my best to pause after the clock turned off. I took 5 runs because there may be human error involved. The average temperature of the naphtha tile on the right was 23.5 degree C, and it was 20.9 degree C on the left.

Thus during this 6 seconds, there were: 700 Kg * 2.191 J/g/K * 0.9 degree K = 1380.33 KJ of heat transfered to the left tile.

Then the heat transfer rate is: 1380.33 KJ / 6 secs = 230.055 KW.

 In theory, with 1 meter contact, the two tiles should transfer heat at: 0.2 W/m/K * 1 m = 0.2 W/K.

The highest temperature difference was 2.6 degree K, thus they should conduct heat at no more than 0.52 W. The reality, however, was half a million times higher. One can explain this by adding in convection in addition to the conduction rate I calculated above, but how much heat is transfered through convection? Is there a defined number to look up anywhere?

[Risu]

2 hours ago, ithilelda said:

However, in reality it raises 9.1 degree K, around 5 times the anticipated amount of delta K. Anyone able to explain?

1/5 of the mass of the building is used in temperature exchange. I assume this simulates the surface of the building.
 

[ithilelda]

1 hour ago, Risu said:

1/5 of the mass of the building is used in temperature exchange. I assume this simulates the surface of the building.
 

Hmmmm interesting. Does this ratio apply to every machinery? Moreover, do pipes/thermshift plates and other background solid objects obey this calculation also?

[Yothiel]

Absolutely all buildings have their mass divided by 5 for the purpose of heat calculation.

For the transfer between cells:

•  heat transfer between liquid is boosted by a 625 factor (to simulate convection I guess)
• all heat transfers between cells are boosted by a 1000 factor, because values which should dimensionally be watts are actually used as kilowatts.

I'm currently doing a write up on what I know about heat transfers, but the text is getting longer and longer !

[eloy2030]

1 hour ago, Yothiel said:

Absolutely all buildings have their mass divided by 5 for the purpose of heat calculation.

For the transfer between cells:

•  heat transfer between liquid is boosted by a 625 factor (to simulate convection I guess)
• all heat transfers between cells are boosted by a 1000 factor, because values which should dimensionally be watts are actually used as kilowatts.

I'm currently doing a write up on what I know about heat transfers, but the text is getting longer and longer !

nice, can we get a preview?

[Tobruk]

2 hours ago, Yothiel said:

Absolutely all buildings have their mass divided by 5 for the purpose of heat calculation.

For the transfer between cells:

•  heat transfer between liquid is boosted by a 625 factor (to simulate convection I guess)
• all heat transfers between cells are boosted by a 1000 factor, because values which should dimensionally be watts are actually used as kilowatts.

I'm currently doing a write up on what I know about heat transfers, but the text is getting longer and longer !

That's wonderful. I cannot wait for that and finally know all the inn-and-outs.

[ithilelda]

2 hours ago, Yothiel said:

Absolutely all buildings have their mass divided by 5 for the purpose of heat calculation.

For the transfer between cells:

•  heat transfer between liquid is boosted by a 625 factor (to simulate convection I guess)
• all heat transfers between cells are boosted by a 1000 factor, because values which should dimensionally be watts are actually used as kilowatts.

I'm currently doing a write up on what I know about heat transfers, but the text is getting longer and longer !

This is absolutely amazing. Can't wait to learn more! One more question before your work came out. Does liquid-gas interactions get that 625 bonus too?

[Yothiel]

1 hour ago, eloy2030 said:

nice, can we get a preview?

I can at least give you some exercise as a teaser!

In debug mode, try painting in an isolated environment (e.g. inside neutronium):

• 1kg of gas hydrogen at 320K
• next to it, 800kg of solid gold amalgalm (not pure gold) at 6.4K (yup, cold!)

Try observing how heat transfers between these two conductive elements at very different temperatures. Be dumbfounded ^___^

6 minutes ago, ithilelda said:

This is absolutely amazing. Can't wait to learn more! One more question before your work came out. Does liquid-gas interactions get that 625 bonus too?

No. Basically, each element has a set of properties, among which are the specific heat and thermal conductivities. Some of these properties are "surface area multipliers", one multiplier per state (solid, liquid, gas). Note that this "state" does not reflect the state of this element (actually each element as only one state, and transition between states is treated as conversion to another element).

Now for example: if a cell of steam is exchanging heat with a cell of water, the calculated heat transfer will be multiplied by the "Liquid Surface Area Multiplier" of the steam (because water is a liquid) AND the "Gas Surface Area Multiplier" of the water (because steam is a gas).

All these values are stored in Unity packages and can be extracted. And so far, all liquids have a "Liquid Surface Area Multiplier" of 25, and all gases have a "Solid Surface Area Multiplier" of 25 as well. All other multipliers are 1.

As such, when liquids interacts together, the 25 multiplier is applied twice. And 25 * 25 = 625. Also, when gas interact with solids, you get a single 25. All other interactions have no modifier at the moment.

[eloy2030]

15 minutes ago, Yothiel said:

In debug mode, try painting in an isolated environment (e.g. inside neutronium):

• 1kg of gas hydrogen at 320K
• next to it, 800kg of solid gold amalgalm (not pure gold) at 6.4K (yup, cold!)

Try observing how heat transfers between these two conductive elements at very different temperatures. Be dumbfounded ^___^

I did it... Sorry, but what am I exactly looking for?

[Yothiel]

With these exact values, these two cells should not be transferring any heat between each other

[eloy2030]

Just now, Yothiel said:

With these exact values, these two cells should not be transferring any heat between each other

mmmh, I´m not following...

[Yothiel]

With pictures:

https://imgur.com/a/fZXsv

A strange situation where there is no visible reason for temperatures not equalizing between two cells.

[eloy2030]

42 minutes ago, Yothiel said:

With pictures:

https://imgur.com/a/fZXsv

A strange situation where there is no visible reason for temperatures not equalizing between two cells.

oh, now I get it, this is what i did...

That´s very interesting, any idea why?