Petroleum Generator, heat, and math

[WanderingKid]

So I'm having a confusing discussion with myself about the math for the heat of a petroleum generator, and I'm hoping someone can help me figure out where I've gone wrong.  Thanks in advance.

So, a Petroleum generator generates 20 kDTU (kW) of heat.  Additionally it produces equal temperature CO2 (500g/s) and PH2O (750g/s).  So, that 20kDTU of heat should end up sinking into the water (assuming you've let it run into a nice small lake before wandering off to the pump) and the CO2 it's producing.

To be able to maintain the cost of the heat quickly, you need to be able to transmit 20 kDTU heat into the unit each second.  Hydrogen in a Thermo Regulator produces 33.6 kDTU.  That's 2.4SHC * 1000g * -14C.  This should be more than enough to maintain the petroleum generator.  However, time is our enemy in this case, as the petroleum generator is generating heat faster then we can diffuse it for a single cell in a single second.

The thermal conductivity of Hydrogen is 0.168.  This means our Hydrogen attempting to maintain the coolness of the petroleum generator should start it's -14C run as close to the PH2O output of the generator as possible, to get the most cooling in the most important position.  Even working with the worst case of minimal transfer, a 14C delta of 1000g of H we should get 2,352 BTU of cooling.  That's not quite a single degree of cooling lost from the H2 to cool the generator.  We have spent 33.6 kDTU of our price to be down to ~31.3 kDTU.  Spread across the 12 squares of a generator with radiant pipes and tempshifts, there's probably a log function there I don't know but we'll rough it out to 12 squares x 1800 BTU, or 21 kBTU, more than the cost of our heat from the Petroleum Generator.

Now, here's my confusion.  In theory, almost another 60% of the heat gained by the generator is theoretically covered by the thermo regulator, in 33.2kBTU - 20 kBTU.  The leftover price of 12 kBTU (roughly) for our Thermo Generator should either be being distributed to additional area around the petro generator, or be recycled into a colder start point on its next loop.   However, this is not happening.  I've heat positive, and I can't figure out what I'm missing.

If it's insulated it's insulated abyssal.  The heat from the small PH2O tank and excess CO2 is being used as the coolant for the H Loop for the Thermo Regulator, so after an initial build up it regulates the amount of material in the tank.  Tempshift is not against the insulated tiles.

What am I missing?

[BlueLance]

I look at this and understand virtually none of it. Out of curiosity, as a pleb what are you trying to do XD

[Djoums]

Same, not sure I get everything. If your generator heats up by 1°C, doesn't that mean the water and CO2 coming out of it will also have 1°C more ? In that case you'd need more than 20kW of cooling to keep the temperature constant overall.

For 1°C that would be 750 * 1 * 4.179 = 3134W for the water, and 500 * 1 * 0.846 = 423W for the CO2.

[Christophlette]

Could we have a picture of the setup ? Because a lot of things can deteriorate your cooling power. Walls, doors, how you cool the generator and such.

[Carnis]

There is a x 200 hidden modifier.

I do not understand The new heat units, so not sure. But it seems to me that 100w x 200 is 20kW, so your hydrogen should work, or cool down.

Is your hydro piping ok? If The thermo regulator is backed up, you might have a 50% utilisation rate.

How hot are the tempshift? Delete them. Radiant pipes & buildings do not interact with tempshifts at all.

The heat transfer with buildings is:

Hydrogen - Pipe - Oxygen/CO - generator.

If you add tempshift, they only interact on The third layer between tiles (solid, liquid, vacuum or gas)

[simonchvz]

Without duplicating your math, are you considering the mass of the generator itself as a sink (and by extension floor tiles, background wiring, piping, vents, etc)?  Some of that missing heat may be present, but being absorbed by those easily dismissed systems.

6 minutes ago, simonchvz said:

Without duplicating your math, are you considering the mass of the generator itself as a sink (and by extension floor tiles, background wiring, piping, vents, etc)?  Some of that missing heat may be present, but being absorbed by those easily dismissed systems.

Edit: In this case, the mass from those other systems would need to be supplying heat, temporarily making you heat positive until everything stabilizes in a closed system)

 

[Carnis]

Another option is, The pW pool is radiating The heat Back to The generator. I think this is actually the most likely explanation.

[KittenIsAGeek]

What is the temperature of the petroleum going into the generator? The generator outputs at input temperature PLUS the heat generated by functioning.

[Carnis]

Just now, KittenIsAGeek said:

What is the temperature of the petroleum going into the generator? The generator outputs at input temperature PLUS the heat generated by functioning.

Are you sure?

IMO generator deletes input temperature and outputs @generator temp.

[KittenIsAGeek]

Hmmm. I may have to test again, then.  Its been a while since I last ran tests.  I'm not home today, so it'll have to be this weekend before I can test.

[jfc]

just get some PW coolant to sit on the generator, problem solved, if you have a slush geyser ofc

[WanderingKid]

2 hours ago, Carnis said:

There is a x 200 hidden modifier.

I do not understand The new heat units, so not sure. But it seems to me that 100w x 200 is 20kW, so your hydrogen should work, or cool down.

I thought this modifier was changed, similar to the AETN now outputting -80 kDTU, showing the correct cooling values for the AETN.

3 hours ago, Carnis said:

How hot are the tempshift? Delete them. Radiant pipes & buildings do not interact with tempshifts at all.

The heat transfer with buildings is:

Hydrogen - Pipe - Oxygen/CO - generator.

If you add tempshift, they only interact on The third layer between tiles (solid, liquid, vacuum or gas)

So I'm adding an extra variable, thanks.  I'll kick that out.

20 minutes ago, KittenIsAGeek said:

What is the temperature of the petroleum going into the generator? The generator outputs at input temperature PLUS the heat generated by functioning.

Right, it should be 16 kDTU normal + 4kDTU Radiated (Tooltip).  So the 16 ends up in the machine itself, 4 ends up in the environment. … Which means I need to spike the coolant to the machine and not control for the radiated.  That may show me what I need with finally getting some sleep and thinking it over.  I think I need to force the refund instead of using too much radiated heat control so I spike the cooling through PWater on the Generator to control the rest of the environment.

This has been my testing setup until now, but I'm going to redo it:

Ground up redesign for the testing.  If I'm right from what you've been able to tell me, I need to completely rework the cooling system to be much, much tighter on where the radiant pipes need to be in play.  Otherwise while I am getting enough cooling, the petro generator is able to get ahead of that cooling in the time it takes to radiate the cooling back into the generator.

… I'll be back when I can adjust the design and test if that's the confusion.

[trukogre]

Wouldn't it be a lot simpler just to drip some -20 C polluted water on the generator?  That would keep it below 0 C, which would mean that the petroleum generator would be operating with net negative heat generation (potentially, depending what you do with the generated pwater).  I don't see the point of a complicated cooling system unless you're trying to keep the pet gen at -200 C to generate super cold dirty ice--or if it's just fun to design, of course.

[Carnis]

17 minutes ago, WanderingKid said:

I thought this modifier was changed, similar to the AETN now outputting -80 kDTU, showing the correct cooling values for the AETN.

So I'm adding an extra variable, thanks.  I'll kick that out.

Right, it should be 16 kDTU normal + 4kDTU Radiated (Tooltip).  So the 16 ends up in the machine itself, 4 ends up in the environment. … Which means I need to spike the coolant to the machine and not control for the radiated.  That may show me what I need with finally getting some sleep and thinking it over.  I think I need to force the refund instead of using too much radiated heat control so I spike the cooling through PWater on the Generator to control the rest of the environment.

This has been my testing setup until now, but I'm going to redo it:

Ground up redesign for the testing.  If I'm right from what you've been able to tell me, I need to completely rework the cooling system to be much, much tighter on where the radiant pipes need to be in play.  Otherwise while I am getting enough cooling, the petro generator is able to get ahead of that cooling in the time it takes to radiate the cooling back into the generator.

… I'll be back when I can adjust the design and test if that's the confusion.

So.

The tempshifts move all The heat Back into The system. The radiator does nothing here, since There is 2 heat transfer systems:

From generator Via hydrogen radiator into pW

From pW to generator Via tempshift.

Cooling The petroleum is a dead end, dont try.

[KittenIsAGeek]

2 minutes ago, Carnis said:

Cooling The petroleum is a dead end, dont try.

I cool petroleum all the time.  It makes a decent extremely-wide temperature range coolant.

[psusi]

48 minutes ago, KittenIsAGeek said:

What is the temperature of the petroleum going into the generator? The generator outputs at input temperature PLUS the heat generated by functioning.

Yea, no.. the input temperature for petrol and natgas generators is ignored.  That's why the prevailing advice on the forums and wiki are that you can generate massive cooling by keeping them at low temperature and getting out lots of cold, high specific heat PH2O.

3 hours ago, Carnis said:

How hot are the tempshift? Delete them. Radiant pipes & buildings do not interact with tempshifts at all.

I'm pretty sure that is not true.  I have used tempshift plates to increase the heat transfer between thermo regulators and radiant pipes, and between radiant pipes and ice in a cold biome.

[trukogre]

4 minutes ago, KittenIsAGeek said:

I cool petroleum all the time.  It makes a decent extremely-wide temperature range coolant.

He didn't say 'Don't cool petroleum".   He said don't cool petroleum that you're about to delete from the world (paraphrasing), which was fairly solid advice.  Petroleum which has been deleted from the world makes an extremely poor coolant, because it's gone.

[KittenIsAGeek]

@trukogre Thanks for clarifying.  I've been working 12-hour shifts all week. My brain is jelly and I didn't catch that inference.

[WanderingKid]

I am finally getting minimal shift, which is about what I expected with the mass involved for the rest of the system.  In about a cycle the Granite block the generator is sitting on has dropped 0.4C, which is enough to start stabilizing the generator.  All items starting at 20C during the rebuild, it spiked up to 33.5C (or so) before stabilizing down to around 30C once the cooling system was fed enough Hydrogen to get it up and running.  I need to shove a few worts in there so I can ignore it for now, but it would appear the biggest problem that I was having was the petroleum generator didn't have enough immediate cooling to get past the generation of the system.

I'm going to run it for a bit and make sure these expectations continue.  On a side note though.... the 6 cells of radiant vent pipe pretty much push the 14C change from the thermos generator back to start position.

[trukogre]

@KittenIsAGeek : Ouch that will be me next week sadly.

@WanderingKid:  The petroleum generator is outputting all that pwater at 30 C there, whereas if you cooled it down to 0 C it would output at 0 C, right?  That difference is 3x the cooling you're getting from your thermoregulator.  Have you tried turning on the cooling system for 30 seconds before turning on the petroleum generator, to see if that reaches a colder steady state equilibrium?

[WanderingKid]

I can confirm negative temperature interaction, albeit slow.  I'm okay with slow, that makes the math make sense.

40 minutes ago, trukogre said:

Wouldn't it be a lot simpler just to drip some -20 C polluted water on the generator?  That would keep it below 0 C, which would mean that the petroleum generator would be operating with net negative heat generation.  I don't see the point of a complicated cooling system unless you're trying to keep the pet gen at -200 C to generate super cold dirty ice.

Where would I get the -20C water from?  Either I cool the drip, or I cool the generator.  One way or another, I need to get cooler.

37 minutes ago, Carnis said:

So.

The tempshifts move all The heat Back into The system. The radiator does nothing here, since There is 2 heat transfer systems:

From generator Via hydrogen radiator into pW

From pW to generator Via tempshift.

Yeah, that was part of the problem. It'll be more efficient once I start pushing the excess CO2 back out of the room too, I believe.  The other part was I was cooling the PWater basin *first*, which meant I'd moved to much of the cooling spike for the -14C away from the generator itself.

22 minutes ago, KittenIsAGeek said:

I've been working 12-hour shifts all week. My brain is jelly and I didn't catch that inference.

Thanks for the help.  I know that feeling.  Rest well.

9 minutes ago, trukogre said:

@WanderingKid:  The petroleum generator is outputting all that pwater at 30 C there, whereas if you cooled it down to 0 C it would output at 0 C, right?  That difference is 3x the cooling you're getting from your thermoregulator.  Have you tried turning on the cooling system for 30 seconds before turning on the petroleum generator, to see if that reaches a colder steady state equilibrium?

You're correct.  When I set these systems up with more specific expectations (for example, I've got 26kg of CO2 and climbing) I'll probably pre-set the Cooling system and its associated mass temperatures items to the target temp (-5 C, to be exact) so that I don't have to let it run for too long before I get the system to calibrate.

[trukogre]

30 minutes ago, WanderingKid said:

Where would I get the -20C water from?  Either I cool the drip, or I cool the generator.  One way or another, I need to get cooler.

 

It doesn't matter where you get an initial amount of -20 C water from, melt some dirty ice from an ice biome, cool some with an aquatuner, whatever.  You don't need very much.  Using that initial amount of -20 C dirty water to cool the petroluem generator, the petroleum generator now generates 750 g/s of pretty dang cold polluted water, from nowhere.  It won't be exactly -20 C, but it will be close enough so that cooling the pwater output back to -20 C with an aquatuner will only require you to run the aquatuner like 10-20% of the time.  If you divert like 250 g/s of that newly generated dirty water into the aquatuner tank, and heat the pwater up to 90 C or so there, that will easily be more than enough heat sink to dump all the heat into.  

So, tl'dr, where do you get the -20 C dirty water from?  The petroleum generator makes it for you, plus extra.  The more you run the petroleum generator, the more cold dirty water you'll have. 

The reason you cool the drip instead of 'cooling the generator' is that you can't stick an aquatuner on a generator to cool it.  I mean, you do realize that your design 'cools the pipes' instead of 'cooling the generator', right?  For the same reason, you can't stick a thermoregulator on a generator either.  The drip method is simpler and works pretty well, because A. aquatuners > thermoregulators and B. Klei's thermodynamic simulation treats a machine sitting in 2 inches of dirty water like it's completely submerged in that water, when it comes to conductivity.  If that seems exploity and you choose to eschew it, I could totally understand that; I personally avoid radiator based systems whenever I can because they work unrealistically poorly (without radiant pipes that is, which are a weird magical hack that I also dislike) and seem to negatively impact fps, but that's just my personal choice.

 

[jfc]

16 minutes ago, trukogre said:

It doesn't matter where you get an initial amount of -20 C water from, melt some dirty ice from an ice biome, cool some with an aquatuner, whatever.  You don't need very much.  Using that initial amount of -20 C dirty water to cool the petroluem generator, the petroleum generator now generates 750 g/s of pretty dang cold polluted water, from nowhere.  It won't be exactly -20 C, but it will be close enough so that cooling the pwater output back to -20 C with an aquatuner will only require you to run the aquatuner like 10-20% of the time.  If you divert like 250 g/s of that newly generated dirty water into the aquatuner tank, and heat the pwater up to 90 C or so there, that will easily be more than enough heat sink to dump all the heat into.  

So, tl'dr, where do you get the -20 C dirty water from?  The petroleum generator makes it for you, plus extra.  The more you run the petroleum generator, the more cold dirty water you'll have. 

The reason you cool the drip instead of 'cooling the generator' is that you can't stick an aquatuner on a generator to cool it.  I mean, you do realize that your design 'cools the pipes' instead of 'cooling the generator', right?  For the same reason, you can't stick a thermoregulator on a generator either.  The drip method is simpler and works pretty well, because A. aquatuners > thermoregulators and B. Klei's thermodynamic simulation treats a machine sitting in 2 inches of dirty water like it's completely submerged in that water, when it comes to conductivity.  If that seems exploity and you choose to eschew it, I could totally understand that; I personally avoid radiator based systems whenever I can because they work unrealistically poorly and seem to negatively impact fps.

 

Petroleoum generator generates cold water? didn't know about that

[trukogre]

1 minute ago, jfc said:

Petroleoum generator generates cold water? didn't know about that

petroleum generator generates 750 g/s of polluted water, at whatever temperature the petroleum generator is currently.  That could be cold, or hot.  If you keep the pet gen at -200 C, it will drop -200 C dirty ice.

[WanderingKid]

29 minutes ago, trukogre said:

The reason you cool the drip instead of 'cooling the generator' is that you can't stick an aquatuner on a generator to cool it.  I mean, you do realize that your design 'cools the pipes' instead of 'cooling the generator', right?  For the same reason, you can't stick a thermoregulator on a generator either.  The drip method is simpler and works pretty well, because A. aquatuners > thermoregulators and B. Klei's thermodynamic simulation treats a machine sitting in 2 inches of dirty water like it's completely submerged in that water, when it comes to conductivity.  If that seems exploity and you choose to eschew it, I could totally understand that; I personally avoid radiator based systems whenever I can because they work unrealistically poorly (without radiant pipes that is, which are a weird magical hack that I also dislike) and seem to negatively impact fps, but that's just my personal choice.

Radiant pipes, as you mentioned, are a weird magical hack.  They bypass the normal container->pipe component and directly affect Container-> Cell.  I had a giant radiant pipe radiator setup to keep the entire local area cool instead of trying to target the generator at first.  After this thread I realize that was a mistake.  Why didn't I do drip cooling?  Because I was trying to figure out why the radiator didn't work the way I expected for the temperature differentials instead of adding more variables into the system.  This first post in this thread was not about building the better mousetrap, but understanding why the cooling system (existing mousetrap) wasn't getting me where I needed it to go and what components I was missing.

I have few problems with my radiators, but as you mentioned, that's a personal thing.  Half the game is magical hacks and systems that are simplified.  It's a game.  I enjoy playing with the engine and what I can come up with, but occasionally I run into a 'game mechanic' wall that doesn't make sense to me.  If Klei decides to get more specific simulation out of their thermodynamic system that could be fun too, but right now, the entire thing is 'exploity', so I don't mind doing drip cooling (I do it for a number of different systems including my NG stuff to cool down slime to kill slimelung) but most of the time I can use a regulator to do an aquatuner's job if I set it up right.