Thermo Aquatuner power efficiency

[Cilya]

I was surprised I did not find any information on this, sorry if it was already discussed.

What is the specific Heat

In terms of heating, the power efficiency is how much you can heat with how much power. The energy needed to heat something by 1 degree depends on the thing and is called the heat capacity. The heat capacity is directly related to the mass of the matter considered as if you take twice the amount of this matter, it will need twice the energy to heat by the same amount. Thus it's convenient to define how much energy is needed to heat by one degree one kilogram of something: this is the specific Heat.

How did I think it worked in Oxygen not Included

Physic in Oxygen not Included is often not realistic, and this is the case for the Thermo Regulator and the Thermo Aquatuner. These consume a fixed amount of power and transfer heat from their gas/liquid contents to the environment. More precisely, the heat extracted from the gas/liquid is add to the building itself which can then heat the surroundings. They reduce the temperature of the gas/liquid going through by 14°C, whatever the content of the pipe is. In terms of energy, as defined above, we can calculate the heat removed from the contents as the product of the temperature change and the heat capacity of the contents.

Heat removed = Temperature difference x Heat capacity = Temperature difference x Specific Heat x Mass

for instance, water has a heat capacity of 4.179 J/g/K and if the pipe contents packets of 10kg we can calculate the heat removed on each packet of water cooled by 14° as

Heat removed = 14 x 4.179 x 10,000 = 585.06 kJ

The heat is transferred from the liquid/gas to the building each time something goes through the thermo regulator or thermo aquatuner. Let's define the heat transfer as the heat transferred divided by the time needed to transfer it, here one second. For our example, we thus have 585.06 kW of Heat transfer.

The liquid tepidizer does not transfer heat but produce it from electric power instead. It's description states that the heat production is 20.32 kW.

In game simulation

I wanted to check the above prediction in game. I build the following small setup to compare how fast both the tepidizer and the thermo aquatuner would heat water. If put 12x1000kg of water at 300K in each bottom cell and let the animation run for a complete cycle (600s).

After a complete cycle

• the water inside the tepidizer cell have heated up from 300°K to 348.2°K and
• the water inside the aquatuner cell heated up from 300°K to 307.00°K

Let's do the calculation.

Theoretical tepidizer Heat added = Heat produced x Time = 20.320 x 600 = 12.192 MJ

Actual tepidizer Heat added = Temp diff x Heat Capacity x Mass = (348.2 - 300) x 4.179 x 12 = 2.417 GJ

Theoretical aquatuner Heat added = Packet cooling x Heat Capacity x Packet Mass x Time = 14 x 4.179 x 10 x 600 = 351.036 MJ

Actual aquatuner Heat added = Temp diff x Heat Capacity x Mass = (307 - 300) x 4.179 x 12 = 351.036 MJ

Conclusion

So the model seems to work of the aquatuner, but not for the tepidizer. The tepidizer actually heated up 200 times more than I had expected and much more (almost 7 times) than the aquatuner. This means that to heat up to 90°C, the tepidizer is a huge power efficiency (8.6 times more efficient) gain over the aquatuner. If you did know, why you didn't tell me ?

 

[Reaniel]

I believe it was already noted somewhere back in the days of battery water boiler, but yes, the actual heat output of machines is 200 times than the listed number.

 

 

[Cilya]

Oh yes, I've already read that thread. I don't know what specifically in this thread you are pointing out, but the Thermo Regulator have been patched since then. Before the patch, there was indeed a net cooling, but not anymore.

[Kasuha]

20 minutes ago, Cilya said:

Oh yes, I've already read that thread. I don't know what specifically in this thread you are pointing out, but the Thermo Regulator have been patched since then. Before the patch, there was indeed a net cooling, but not anymore.

I believe there was never net cooling on thermo regulator. Those claims came from the observation that the regulator can cool itself and more, but at the same time there was the bug that material moving in pipes was not changing temperature - so as soon as it got cold enough, there was no problem for the regulator to cool itself. Only when that bug was fixed and we made more accurate measurements we realized that there's no heat lost on it.

 

[Cilya]

Ok. I remember reading this guy on reddit who did the same calculation as me a long time ago. The thermo regulator was described as releasing a constant amount of Heat instead of moving Heat which it actually does now. His experiment does not relies on the heat exchange between the pipe and the exterior, but you can't do this experiment without a pipe. (could have done with an abysallite pipe though) I don't know if the difference he saw in temperature can be explained by the four pipes needed to set things up.

[Kasuha]

There are two ways to compare power efficiency of the thermo aquatuner. First, per processed mass. Aquatuner can process ten times the mass (10 kg vs 1 kg) for five times the power (1.2 kW vs 240 W) and that would make it two times more efficient. More useful is in my opinion second metric, displaced heat per unit of power in optimum regime. That would be processing hydrogen with the regulator (1 kg packets with 2.4 J/g/K and 240 W), and polluted water with the aquatuner (10 kg packets with 6 J/g/K and 1.2 kW) and then it is 12.5 times more efficient.

[Kasuha]

6 minutes ago, Cilya said:

Ok. I remember reading this guy on reddit who did the same calculation as me a long time ago. The thermo regulator was described as releasing a constant amount of Heat instead of moving Heat which it actually does now.

Okay, I may be wrong about it. I can remember there was something like "producing heat 170 W" in the regulator's tooltip at times which does not show up now, it just says "cooling". So things were changing around it. Although the "170 W" I saw with 1 kg packets of hydrogen closely match what it should actually report (considering the x200 multiplier).

[Cilya]

10 minutes ago, Kasuha said:

There are two ways to compare power efficiency of the thermo aquatuner. First, per processed mass. Aquatuner can process ten times the mass (10 kg vs 1 kg) for five times the power (1.2 kW vs 240 W) and that would make it two times more efficient. More useful is in my opinion second metric, displaced heat per unit of power in optimum regime. That would be processing hydrogen with the regulator (1 kg packets with 2.4 J/g/K and 240 W), and polluted water with the aquatuner (10 kg packets with 6 J/g/K and 1.2 kW) and then it is 12.5 times more efficient.

 

Yes, I had the same conclusion. I believe the proper way to describe the efficiency of the thermo aquatuner compared to the one of the thermo regulator is to associate at each temperature what is the most efficient conductor to exchange heat. To cool things down to -20°C the polluted water is the best conductor. To cool things down to -180°C the liquid methane is the best conductor. But to go beyond that and cool things even further, hydrogen gas remains the best (better than liquid oxygen), thus keeping the usefulness of the Thermo Regulator.

But that was not really the subject of this thread. Rather than comparing the aquatuner to thermo regulator, I wanted to compare them both to the liquid tepidizer.

 

[Sevio]

If you want to compare them purely on power efficiency, I'll just put the numbers all in one spot, assuming optimum regime for Regulator and Aquatuner: (I'm using J/s for the transfer/heatup rates only to differentiate it from electric power consumption)

• Regulator Hydrogen transfer rate: 2.4 * 1000 * 14 = 33600 J/s (heat capacity * mass per packet * temperature delta)
• Regulator power consumption: 240 W
• Aquatuner Polluted Water transfer rate: 6 * 10000 * 14 = 840000 J/s
• Aquatuner power consumption: 1200 W
• Tepidizer heatup rate: 20320 * 200 = 4064000 J/s
• Tepidizer power consumption: 960 W

Now we can determine the Coefficient of Performance (COP), aka how much heat is produced per unit of electrical power spent:

• Regulator: 33600 / 240 = 140
• Aquatuner: 840000 / 1200 = 700
• Tepidizer: 4064000 / 960 = 4233.3333...

From the numbers above, we can see the Tepidizer is 6 times as efficient as the Aquatuner, which in turn is 5 times as efficient as the Regulator.

Spoiler

COP is a statistic also used for real-world heating and cooling devices, however the numbers ingame are vastly inflated as real-world fridges and airconditioners can reach a COP of about 4 and electric heating devices like water boilers, microwaves and electric stoves all have a COP of less than 1.

As noted above, the Aquatuner is not efficient when cooling liquid oxygen compared to a Regulator cooling gaseous hydrogen:

• Aquatuner Oxygen cooling rate: 1.01 * 10000 * 14 = 141400 J/s
• Aquatuner Oxygen COP: 141400 / 1200 = 117.833....

But it is a very compact way of liquefying a lot of oxygen, as no large radiator is required.

[Cilya]

1 hour ago, Sevio said:

• Tepidizer heatup rate: 20320 * 200 = 4064000 J/s

You wrote 200 because that's what I measured or do you have more insights about that ? I'm really interested about the reason it is so efficient. Is it a bug or a feature ?

[Reaniel]

4 hours ago, Cilya said:

You wrote 200 because that's what I measured or do you have more insights about that ? I'm really interested about the reason it is so efficient. Is it a bug or a feature ?

The 200 number was actually inferred inside that old post by Kasuha and others I've linked.

The short version:  

Machine heat output is calculated as kW instead of W, or mW instead of kW, for whatever reason (maybe to make heat management a bigger part of the game?)  So that's a factor of 1000.

But then machines actually only weigh 1/5 of the listing mass in the heat engine (don't ask me why).  To "hide/compensate" for this fact, the machine heat output also goes down to 1/5.

So the end result: 200x

Again, only the devs can answer the question as to why they designed the heat engine like this.  I can only tell you that it is 200, and how it gets to 200.

 

[trukogre]

On 9/3/2017 at 10:59 AM, Cilya said:

 

Yes, I had the same conclusion. I believe the proper way to describe the efficiency of the thermo aquatuner compared to the one of the thermo regulator is to associate at each temperature what is the most efficient conductor to exchange heat. To cool things down to -20°C the polluted water is the best conductor. To cool things down to -180°C the liquid methane is the best conductor. But to go beyond that and cool things even further, hydrogen gas remains the best (better than liquid oxygen), thus keeping the usefulness of the Thermo Regulator.

But that was not really the subject of this thread. Rather than comparing the aquatuner to thermo regulator, I wanted to compare them both to the liquid tepidizer.

 

The subject of your thread is flawed.  The best uses of the thermoregulator and aquatuner are

1. Cooling things down.

2. Bringing things from 80C to boiling.  

 

The tepidizer can't do either of those.  You're comparing apples and oranges.  The apple is better at being an apple, and oranges are better at being oranges.  You're comparing the tepidizer doing what it's designed to do, to the aquatuner's byproducts.  It's like you're comparing an electrolyzer to a coal generator, and concluding that the coal generator sucks because co2 isn't as useful as either hydrogen or oxygen.   Yes, but what is the reason people build coal generators?

[Grimgaw]

1 hour ago, trukogre said:

 You're comparing apples and oranges

Are you implying that fruit cannot be compared?

[trukogre]

8 minutes ago, Grimgaw said:

Are you implying that fruit cannot be compared?

No. E.g., Gala apples are better than Red Delicious at all things, except being red.  Gala Apples are more Delicious, more appley, more deadly, better at PUBG, faster at a dead sprint in a pure nitrogen atmosphere, and look better in a string bikini. How bout dem apples?

[Cilya]

2 hours ago, trukogre said:

The tepidizer can't do either of those.

You are right, I can't compare the Tepidizer and the Aquatuner. But like apple and oranges have comparable characteristics, the Tepidizer and Aquatuner have a comparable one. They both have a power efficiency when it comes to heating things. That is their power efficiency I compare. This is the characteristic people are interested in when building boilers.

[trukogre]

12 minutes ago, Cilya said:

You are right, I can't compare the Tepidizer and the Aquatuner. But like apple and oranges have comparable characteristics, the Tepidizer and Aquatuner have a comparable one. They both have a power efficiency when it comes to heating things. That is their power efficiency I compare. This is the characteristic people are interested in when building boilers.

No, people are interested in the ability to operate above 80 Celsius without using an exploit that also breaks the tepidizer when it comes to building boilers, and secondarily they're interested in the ability to run the boiler without heating up the whole asteroid, and only thirdly in power efficiency of heating.  This is why pretty much every boiler design you see centers around aquatuners, and not tepidizers.

[The Plum Gate]

On 9/3/2017 at 9:54 PM, Reaniel said:

But then machines actually only weigh 1/5 of the listing mass in the heat engine (don't ask me why).  To "hide/compensate" for this fact, the machine heat output also goes down to 1/5.

...

Again, only the devs can answer the question as to why they designed the heat engine like this.  I can only tell you that it is 200, and how it gets to 200.

 

About this, I've recently learned from other threads that this heat conduction is occurring across the tiles that it occupies, so a tepidizer would be taking up 4 tiles, and so would an aquatuner. Is this being factored in people's tests - I'm almost certain it is.

As for the discrepancy in 'effective mass' vs the displayed mass of the machine - this is the one number that can be tweaked for balancing purposes without having to rebalance the cost to build it or any associated variables - including power consumption. So they settled on 1.2Kw for the aquatuner - because power system is balanced, and then turn to an essentially hidden variable for tuning how it works in the heat engine.

I liken these machines which exchange or output heat to toasters. If you couldn't change the amount of energy it used, and couldn't change it's size - in order to change how well toasted your bread gets then you would have to change how much of the energy is transferred to the toast - the only way to do this would be to change it's effective mass in the simulation as it works today.

Another analogy might be as simple as a light bulb - only the filament gets hot, yeah, the bulb gets warm, but not the whole lamp. What we see in game is the whole machine getting hot - I would hate to have to look at some small part of the machine to see that it's overheating.

-

So it would be nice to throw testers a bone in the info screen regarding it effective heating and cooling mass vs that of the machine mass. Most of them just indicate that they move heat from one place to another at a set number - but we all like to cram as much capacity as possible into the designs and knowing these would be helpful.

 

[Cilya]

1 hour ago, trukogre said:

No, people are interested in the ability to operate above 80 Celsius without using an exploit that also breaks the tepidizer when it comes to building boilers, and secondarily they're interested in the ability to run the boiler without heating up the whole asteroid, and only thirdly in power efficiency of heating.

What you say is not in contradiction with what I have said. I'll try to explain simply. Since both are producing heat, they are both comparable to heat water to at least 85°C. Their efficiency in that matter is then interesting to compare. If the tepidizer is more efficient, and it is, then it is useful to use it to pre-heat water before getting it to the aquatuner to effectively boil it. To know whether you can improve power efficiency by pre-heating or not, you need to compare the power efficiency of heating of both buildings.

1 hour ago, trukogre said:

This is why pretty much every boiler design you see centers around aquatuners, and not tepidizers.

Including both boiler designs I have posted myself. This is why I am already well aware of this.

[Cilya]

33 minutes ago, The Plum Gate said:

As for the discrepancy in 'effective mass' vs the displayed mass of the machine - this is the one number that can be tweaked for balancing purposes without having to rebalance the cost to build it or any associated variables - including power consumption.

I don't understand. Having a displayed mass different from the effective mass (for heat exchanges) as a major influence on heat capacity. With the current state of the code, it also changes the heat produced by the tepidizer. But the heat produced / exchanged can be balanced by changing their numbers without unbalancing anything else. So this 1/5 only seems to balance the heat capacity without unbalancing the building cost like you said. Is that what you are saying ? What I don't understand is that I don't see a real urge to balance the heat capacity. It changes the thermal inertia of the system, but that's not a very important thing, is it ?

[Saturnus]

Since I just recently got time to actually play the game again, I'm surprised no one to my knowledge has used the fluid dynamics of water in the game that clean water floats on top of polluted water to take the output of a steam geyser which is a rather constant supply of 95C water plus additional 150C steam to vastly increase the efficiency of their boiler systems.

[Cilya]

Not directly both things together. But

• @Misha_SOS used a geyser to heat input polluted oxygen and
• @Sevioused water floating above polluted water to exchange heat both to cool output clean water and heat input polluted water.

They both used the fact that you should exchange heat between the input and the output to increase the boiler power efficiency. But the first design focuses on benefiting from the geyser while the second one focuses on compactness and thus heat transfer. (this is why the fluid dynamics where so important there)

I'm not really fond of superposing water and polluted water as it requires to be very careful that one doesn't block the other. Outside debug mode, you can have a shortage of polluted water or not enough space to store clean water anymore as the production/consumption is not constant through a game. Those could break the equilibrium of a boiler which relies too much on fine tuning. So i would better build very large exchangers than using superposition. In my games, space efficiency is not the key factor.

[Saturnus]

I don't see either of those caveats being any real reason not to use it in a play through. Just increase storage/buffer tanks for both.

The system @Sevio built doesn't seem to rely on a specific throughput of polluted water. It relies on minimal throughput, ie. you must have some polluted water running in the system but the minimum can be only a few grams per seconds. And the idea is not space efficiency. It is power efficiency. The more you can preheat the water before the aquatuner boiler the more energy efficient it is. You can then use that increased efficiency to increase the throughput, or you can just do it for the power saving.

[Cilya]

3 minutes ago, Saturnus said:

I don't see either of those caveats being any real reason not to use it in a play through. Just increase storage/buffer tanks for both.

There is always something going wrong. It really depends on how you play. If like me you constantly increase the size of your base, you regularly needs to increase the size of the system, free space, move things and what is more relevant here, move pipes. This creates interruptions in you supply. Sometimes, the supply is cut by mistake. A system which is resilient to these problem is also nice.

 

9 minutes ago, Saturnus said:

@SevioAnd the idea is not space efficiency. It is power efficiency.

The idea is both. I thought it was clear that we were already talking about power efficiency. This is always a concern in a boiler. But @Sevio clearly stated that he wanted to optimize the space used. This explains why he improved a lot the heat exchange rate in his designs. If you don't mind space, you can just build a larger heat exchanger.

[Saturnus]

1 minute ago, Cilya said:

There is always something going wrong. It really depends on how you play. If like me you constantly increase the size of your base, you regularly needs to increase the size of the system, free space, move things and what is more relevant here, move pipes. This creates interruptions in you supply. Sometimes, the supply is cut by mistake. A system which is resilient to these problem is also nice.

That's the reason I like the idea of combining it with a steam geyser.

You'll need a secondary PW input valve set to release a few grams per second (whatever is the minimum flow rate) with a short buffer pipe. That should make sure that part of system can survive a few hundred cycles interruption.

The clean water from the geyser automatically shuts down if the geyser overflows so if you use the boiler as the clean water tank it will have a built-in shutdown if you don't use enough clean water.

I need to devise a security against polluted water overflow but I think it might be as simple as a secondary hydroswitch controlled pump that dumps out excess polluted water if the polluted water input valve is set too high. 

[Cilya]

1 minute ago, Saturnus said:

I need to devise a security against polluted water overflow but I think it might be as simple as a secondary hydroswitch controlled pump that dumps out excess polluted water if the polluted water input valve is set too high. 

I really like the idea of backup systems. It feels much more realistic and add complexity to the design.