Help me understand heat transfer?

[uraharakisuke]

I've been browsing past threads on heat transfer and so far what I understand is that the heat transfer between 2 objects is the log mean of their conductivity. If using insulated however, the game just takes the lower conductivity instead, right?

 

With this in mind I've been repairing my ice biomes using double insulated igneous tiles. But always without fail, I notice something weird. The insulated tiles in contact with the cold air always changes temperatures much faster than the ones in contact with ice or cold granite (see pic). The igneous used was the same temp for the whole wall yet the part in contact with air (pol oxygen) is changing in temp much faster. But conductivity of pol oxygen is much lower than that of ice, so shouldn't this be happening the other way around?

 

One last question, is double insulated igneous the best way to repair ice biomes or should i be using insulated ceramic instead (single only, can't afford double insulated ceramic)?

[natanstarke]

Double layer insulate more yes. As for how the heat works, gases and liquids exchange temperature with each other at the spd of the faster one conducting the heat, the lower the number in the heat conductivity faster it will move heat to the other element, heat capacity is how much heat a element can store " inside " of itself, hope you understood it.

 

[Angpaur]

4 hours ago, natanstarke said:

As for how the heat works, gases and liquids exchange temperature with each other at the spd of the faster one conducting the heat, the lower the number in the heat conductivity faster it will move heat to the other element

Wait... did you just insinuated that lower conductivity means faster heat transfers?!

Please refrain from posting about heat again. Thank you in advance!

 

5 hours ago, uraharakisuke said:

I've been browsing past threads on heat transfer and so far what I understand is that the heat transfer between 2 objects is the log mean of their conductivity. If using insulated however, the game just takes the lower conductivity instead, right?

Average value is used only in one case. Generally in most heat transfer cases a lower conductivity of 2 materials is used. There are 2 exceptions: normal and radiant pipes transfers with its contents and building to tiles.

There is so much confusion about how heat transfers work in ONI. Please read below post and it should clarify things:

Spoiler

 

Regarding that matter with gases - there is additional multiplier of 25 for heat transfer of gases and other tiles, so that is why you observe faster temperature change. Also both low mass and heat capacity of gases have heavy impact on this.

 

Specific heat capacity and mass is taken into equation to calculate temperature changes - high mass and relatively high SHC of 2.05 of ice tiles will prevent them from heating up fast.

 

EDIT: while what I wrote in the spoiler is still true it doesn't concern the problem you were asking about.

So to understand why gases can change insulated tile temperature faster than ice you need to know that game simulation sometimes just skips calculations if temperature change would be too small. That is why you won't observe any temperature change of any insulated tile if you put next to it for example 700kg of hot petroleum. But if you place there 10 kg of petroleum there will be temperature change of insulated tile.

Same happens with gases - low mass and additional multiplier of 25 makes them able to change temperature of insulated tiles, because at least one of side of the equation will have temperature change big enough to calculate it.

[uraharakisuke]

6 hours ago, Angpaur said:

......that game simulation sometimes just skips calculations if temperature change would be too small.

That's really weird but i guess it makes sense from a performance/fps point of view and it also perfectly explains my problem.

 

Thank you for linking that thread. things are much clearer now!

[Yunru]

6 hours ago, Angpaur said:

Generally in most heat transfer cases a lower conductivity of 2 materials is used.

That only happens when using insulated tiles.

[Angpaur]

14 minutes ago, Yunru said:

That only happens when using insulated tiles.

No. Please read the post I linked.

[Yunru]

1 minute ago, Angpaur said:

No. Please read the post I linked.

I refuse!

Reading is something I have no time for!

Ever.

What am I responding to?

[Angpaur]

TL;DR - all tiles like liquids, gases and solid are exchanging heat using lower conductivity of 2 materials.

[Sasza22]

Iirc gasses and liquids have a special modifier when exchanging heat (i on`t remember if it`s x5 or x25) but your insulated walls will transfer heat much faster with gas than with solid (which is what you observed in the OP). This is also the reason double insulated walls are so effective.

[uraharakisuke]

1 hour ago, Sasza22 said:

Iirc gasses and liquids have a special modifier when exchanging heat (i on`t remember if it`s x5 or x25) but your insulated walls will transfer heat much faster with gas than with solid (which is what you observed in the OP). This is also the reason double insulated walls are so effective.

yeah i always just assumed that people preferred double insulated due to their log mean being low. But I wasn't aware of the x25 modifier or the fact that extremely small temp increases are just ignored.

 

Things make much more sense now. Always thought it was weird how single insulated igneous tiles didn't become hotter when in contact with molten iron.

[Nightinggale]

11 hours ago, Angpaur said:

Please refrain from posting about heat again.

Why? Do you fear it will result in a heated debate?

[Gurgel]

On 6/5/2019 at 7:43 AM, Angpaur said:

So to understand why gases can change insulated tile temperature faster than ice you need to know that game simulation sometimes just skips calculations if temperature change would be too small.

I very much doubt it does that. It would not make sense, it would actually increase effort and make the code more complex and potentially a lot slower.

But as discussed numerous times, the game uses IEEE 754 "single precision"  floats with only 23 bit of resolution. If you add, for example, 1/(2^24) to 1 with that, the result will be unchanged. 

[mathmanican]

6 hours ago, Gurgel said:

I very much doubt it does that.

It has for a really long time.  From @Yothiel's amazing post, below, we learn this important tidbit, namely

• if |ΔQ'| < 1.0E-4, no transfer is done

This was the key fact that lead me to make a chlorine clamp for the steam turbine, and get infinite energy from nothihng.

Here's his post - most of it is still applicable. 

 

[Coolthulhu]

On 6/5/2019 at 4:58 AM, uraharakisuke said:

With this in mind I've been repairing my ice biomes using double insulated igneous tiles. But always without fail, I notice something weird. The insulated tiles in contact with the cold air always changes temperatures much faster than the ones in contact with ice or cold granite (see pic). The igneous used was the same temp for the whole wall yet the part in contact with air (pol oxygen) is changing in temp much faster. But conductivity of pol oxygen is much lower than that of ice, so shouldn't this be happening the other way around?

There is a hidden modifier on transfer rates dependent on phases in contact. I don't remember the exact multiplier, but I think it was 25 times for gas on solid.