Water plumbing heat experiment

[Grimgaw]

46 minutes ago, Sevio said:

There is a good reason why there are no liquid pipe bridges in my heat exchanger above though: They would break quite easily inside a heat exchanger with steam because they have a base overheat temperature of 75 C, which is increased to 90 C for some materials like granite and obsidian. Wire bridges are the only bridge that don't have this problem.

Out of curiosity, what happens with liquid pipe bridge when it overheats? Does it stop conducting heat?

[Sevio]

22 minutes ago, Grimgaw said:

Out of curiosity, what happens with liquid pipe bridge when it overheats? Does it stop conducting heat?

It takes damage until it gets broken. It will look annoying but I'm not sure if it will actually stop conducting heat... Could be worth sacrificing the materials if it still conducts heat.

[Cilya]

Gas bridges are better (conductor) than electric bridges but worse than pipe bridges. It's consistent with the assumption that the mass is a key factor.

 

Since I discovered that Liquid Tepidizers are 8 times more efficient than aquatuners, I tried a design wherre the water is preheated to 85°C with a tepidizer and then heated to 120°C with an aquatuner. Since the water only needs to get heated by 35°C, I expected an output of 3kg/s. I actually got half. I don't know where half the heat of my aquatuner (working 100% of the time) have gone.

I know why your design has a better power efficiency (it heats the input water with the condensed water) but I should have a better throughput... yet I don't.

[Sevio]

@Cilya Thanks for confirming, that's what I thought

I also did a brief test with all bridge types conducting heat from a hot tile of polluted water to a cold tile. In order of performance:

1. Granite Liquid Bridge
2. Granite Gas Bridge
3. Iron Wire Bridge

This follows the expected order based on their masses and specific heat capacities, where total heat capacity of the building seems to be the determining factor in heat conductivity with water.

The differences between the bridge types are quite  pronounced, especially that between Granite Gas bridge over Iron Wire bridge, due to having twice the mass and just over 175% of the heat capacity. Also useful to note is that a broken liquid bridge will continue to conduct heat at the same rate as a healthy one! You'll have to put up with the empty health bar and a broken building message though.

And last but not least a peculiarity for use in heat exchangers: Sculpting blocks are 3 tiles tall, can be made of 400 kg of granite and performed even better than liquid bridges at vertical heat transfer! Blank Canvas are also made of 400 kg granite and don't need a surface, so could perhaps also be useful in some situations. This allowed for an upgrade of my heat exchanger like so:

And it did help tremendously, I can now run stable throughput of 3500 g/s without a steam bottleneck building up. 3750 g/s and even 4000 g/s worked for a little while but after one or two evaporation cycles caused dangerous buildup of polluted water in the Aquatuner basin that would have led to a spill and breakdown.

35 minutes ago, Cilya said:

I know why your design has a better power efficiency (it heats the input water with the condensed water) but I should have a better throughput... yet I don't.

You are getting 3000 g/s with Tepidizer preheating, that's still a better throughput than my build in cold mode. (not using tepidizer) In hot mode, the tepidizer and heat exchanger work together to push polluted water temperature to 87.1 C in the pipe as it goes to the Aquatuner.

I'm not sure what end temperature you're getting when you pump the polluted water to the Aquatuner, but I imagine it's slightly lower which would explain the difference in throughput vs my build in hot mode.

Edit:

36 minutes ago, Cilya said:

I expected an output of 3kg/s. I actually got half.

Sorry misread your post earlier, I have no idea why you're getting so little throughput, maybe a screenshot could help?

[Cilya]

I don't have the 3kg/s. That should be the production of steam since if I do the calculations, but I don't know where the Heat goes.

The aquatuner heat exchange is

Heat exchange = 10 x PW Specific Heat x 14

and the amount of water it can heat from 85°C to 120°C is

Steam  flow = Heat exchange / (120 - 85) / PW Specific Heat = 10 x 14 / (120 - 85) = 4kg

So when working at 100%, the aquatuner should heat 4kg / s from 85°C to 120°C. If it doesn't then heat is lost somewhere. I still didn't find where.

I chose not to cool the output water so the output is still hot.

[Sevio]

@Cilya Looks like you missed my revision to my post after I realized I misread yours, could you post a screenshot?

[Saturnus]

@Sevio you might be able to make heat exchanger even more efficient with gas or liquid loops. I use them in the liquefier design I'm working on and because the valve is metal, ie wolframite, it's incredibly good a equalizing temperature from a liquid to a surrounding gas or vice versa. In your set up you can even use liquid pipes with PW loops.

 

[Cilya]

It a very simple design of boiler. The polluted water is preheated by a tepidizer to 85°C and heated by the aquatuner from 85-87°C to 120-121°C. The aquatuner work with a feedback loop, exchanging a fraction of its content at each round with the input polluted oxygen. This loop is also used to condensate steam.

The aquatuner is kept submerged at levels similar to your design: when the level goes below one tile, it is refiled to 1 tile + 1/4.

19 minutes ago, Saturnus said:

@Sevio you might be able to make heat exchanger even more efficient with gas or liquid loops.

I've tried some of them. Liquid going through pipes helps to transfer heat but nowhere near the heat transfer provided by bridges. Usually wolframite is not better than iron because the conductivity from the pipe to its content is the minimum of the two. I've tried with mercury (the two left experiments) which can benefit from the wolframite pipes, but it's still has a very low heat transfer though better than water. Hydrogen gas is worse than water both for heat capacity and conductivity.

[Saturnus]

@Cilya I think you missed the point of my post. The point was to have the valve in the liquid because it's high mass compared to a pipe. In earlier updates there didn't seem to be any heat transfer from valves to or from whatever is in the pipes. This seems to have changed. It also means that you should be careful where you place valves because they (apparently) can transfer heat, and since valve can only be made of metal this can lead to unwanted heat loss.

The other point is that since we want to preheat the water as much as possible after the tepidizer and before going into the aquatuner boiler you can take advantage of the steam always being above 100C. This will make cooling more efficient and heat up the water a couple of degrees making the work the aquatuner needs to do less.

[Cilya]

I didn't thought the valve had a particular role. Are you sure of that ? I'm testing it right now and I can't see a difference. The blue valves are at 9°C despite water at 61°C continuously flowing through hit. The exchange experiment to the right (with 4 valves) doesn't equalize the temperature faster than the one without valves.

In what conditions do you observe a difference induced by a valve ?

[Sevio]

18 minutes ago, Saturnus said:

@Sevio you might be able to make heat exchanger even more efficient with gas or liquid loops. I use them in the liquefier design I'm working on and because the valve is metal, ie wolframite, it's incredibly good a equalizing temperature from a liquid to a surrounding gas or vice versa. In your set up you can even use liquid pipes with PW loops.

I never thought that valves would exchange heat with their pipe's contents. Are you sure this is the case? Even if they do, the sculpting blocks are definitely a lot simpler to set up and it has made my heat exchanger so much more effective that there is very little steam in the exchanger at all now in cold mode.

8 minutes ago, Cilya said:

The aquatuner work with a feedback loop, exchanging a fraction of its content at each round with the input polluted oxygen. This loop is also used to condensate steam.

This is interesting to see, I had not considered this to be viable within reasonable space because heat exchange between pipes and contents is so low.  But evidently the large temperature difference between the cooling medium and the steam is enough to make it work. You're definitely saving a lot of power on pumping that way.

14 minutes ago, Cilya said:

The aquatuner is kept submerged at levels similar to your design: when the level goes below one tile, it is refiled to 1 tile + 1/4.

Is your hydro switch controlling the pump or the Aquatuner? If you have the Aquatuner running continuously and the pump is activated to refill when it drops below the needed level, there will be some runtime where the top tile is filled with steam instead of polluted water. I've spent a lot of time monitoring the Aquatuner closely and it seems to prefer outputting its heat in the top right tile, which is always the hottest in my design. While there is steam in that tile, most of the Aquatuner's output will be wasted on heating the steam.

My build avoids that by having an outside valve set to a fixed rate of polluted water input and the hydro switch controls the Aquatuner in an OR fashion with the thermo switch. Aquatuner is allowed to run if temperature is less than 119.5 C OR the second tile goes above 100 kg. This does produce a small amount of downtime after each vaporization cycle, which on reflection probably helps to explain why I couldn't reach over 3500 g/s.

[Saturnus]

4 minutes ago, Sevio said:

I never thought that valves would exchange heat with their pipe's contents. Are you sure this is the case? Even if they do, the sculpting blocks are definitely a lot simpler to set up and it has made my heat exchanger so much more effective that there is very little steam in the exchanger at all now in cold mode.

Not entirely sure. However, I also found something better... even better than sculpting blocks and paintings. Lavatories and liquid pumps are 400kg of solid metal. Liquid pump don't require a foundation either, like paintings.

[Sevio]

1 hour ago, Saturnus said:

Not entirely sure. However, I also found something better... even better than sculpting blocks and paintings. Lavatories and liquid pumps are 400kg of solid metal. Liquid pump don't require a foundation either, like paintings.

They probably work well enough but the limiting factor as we've determined with wire vs liquid bridges is not the heat conductivity. It doesn't come into play since the lower conductivity (that of water) is used for the heat transfer (this was figured out by Coolthulu some time ago), however a higher heat capacity does matter as the difference between liquid bridge and gas bridge shows. Granite has 177% of the heat capacity of Iron, and the mass of paintings and sculpting blocks is also 400 kg.

[The Plum Gate]

I'm surprised no one has thrown in any ( metal ) showers, but I suppose compactness is a desirable design choice.

[Saturnus]

1 hour ago, The Plum Gate said:

I'm surprised no one has thrown in any ( metal ) showers, but I suppose compactness is a desirable design choice.

As Sevio noted above, Coolthulu have previously proven that heat conductivity is irrelevant for heat transfer as it is the substance with least heat conductivity (in this case the water) that is used in the calculation but instead relies on heat capacity and mass per tile which means that granite is the best material, and sculpting blocks (or sculptures) is the best building to use as it gives the highest heat capacity and the highest mass per tile.

I was unaware of this myself but I checked it and can confirm this is the case. So shower coolers are actually less efficient than sculpture coolers. 

[trukogre]

On 9/2/2017 at 11:37 PM, Sevio said:

My initial boiler design had a heat exchanger with a deeper pool of polluted water that the channel then flowed into, which triggered the heat loss bug as shown in the original post and prevented the exchanger from heating up to an average between the steam and polluted water. That was an issue I was able to find and fix with the experiments I did.

I've read that example you posted and it does indeed explain how boiling polluted water leads to a fundamental loss of heat. Well, almost. Boiling polluted water also produces dirt, which is presumably where the rest of the heat is going. Unfortunately that heat can't be practically reclaimed so we're left with an unavoidable heat deficit. Even when I teleported the generated dirt at 120 C into the cooling chamber, it gives off its heat far too slowly to keep the water from cooling down.

Two issues with that.  A.  The amount of dirt that is produced is theoretically 1%, which is insignificant, and sometimes it doesn't even show up, which is why I omitted to include it in the earlier calculations.

B.  If you were to include the created dirt into the equations, it would actually *increase* the calculated heat deletion, not decrease it.  The new specific heat differential would be 6 versus .99(4.18) + .01(1.48), which is a larger difference than that achieved in the earlier calculations.  This is using the figures I've seen, I believe from Risu?, that claim that 1 kg of pwater should boil into 990 g of clean water and 10 g of dirt.

This teaches us that presuming, or assuming, doesn't replace doing the math, and also shows why the fact that pwater has a higher specific heat than clean water is so weird, since dirt has a much lower specific heat than water, why would adding dirt to water in a very small amount raise the specific heat by a large amount, when it should realistically *lower* the specific heat by a tad.

Most of the conclusions of your post are better explained by this simple fact than by the means you are trying to explain them with. Although the vertical bug still definitely exists, and is indeed most likely affecting your models here, your presumption that there's no heat being lost from the specific heat differential seems to be leading you to incorrect conclusions about the causes underlying your observations.  It appears on first glance that the correct interpretation of your new final design is that your new finished design uses the dropping water exploit to create heat, although it isn't quite creating as much heat as the specific heat differential is destroying, which is why over a long period of time it begins to cool, unless you activate the tepidizer you also included.

[Risu]

8 minutes ago, trukogre said:

The amount of dirt that is produced is theoretically 1%, which is insignificant, and sometimes it doesn't even show up, which is why I omitted to include it in the earlier calculations.

Technically any amount less than 0.1 kg wouldn't produce dirt as the game destroys items less than 0.001 kg in certain situations.
 

[trukogre]

Just now, Risu said:

Technically any amount less than 0.1 kg wouldn't produce dirt as the game destroys items less than 0.001 kg in certain situations.
 

Ah, that explains why the dirt shows up only in large scale boiling.  That mechanism seems obvious now that it's explained, as they so often do.

[Sevio]

7 hours ago, trukogre said:

If you were to include the created dirt into the equations, it would actually *increase* the calculated heat deletion, not decrease it.  The new specific heat differential would be 6 versus .99(4.18) + .01(1.48), which is a larger difference than that achieved in the earlier calculations.  This is using the figures I've seen, I believe from Risu?, that claim that 1 kg of pwater should boil into 990 g of clean water and 10 g of dirt.

I was not aware of the ratio of dirt produced, it seemed like quite a lot builds up up over time but of course on closer inspection it would be pretty small compared to the amount of water being processed here. So you're correct that there is a fundamental heat loss, which almost entirely explains the heat loss that I've been seeing in my design's cold mode.

7 hours ago, trukogre said:

It appears on first glance that the correct interpretation of your new final design is that your new finished design uses the dropping water exploit to create heat, although it isn't quite creating as much heat as the specific heat differential is destroying, which is why over a long period of time it begins to cool, unless you activate the tepidizer you also included.

I have to note my doubts here about creating heat via the dropping water exploit. Do you know of an experiment where this can be shown to be the case? I haven't been able to observe heat creation in any of the tests within my first two posts, if anything experiment 2 of my second post showed that dropping hot onto cold water in some situations destroys heat rather than creates it. I'm also not aware of any other tests where heat creation was shown by dropping liquids onto other things. Perhaps this needs a more well constructed test where a valve drops very small amounts of hot water onto a larger volume of cold water, but so far I've only ever seen or heard of heat being destroyed, not created by water drop quirks.

[trukogre]

Ah, I'd looked at your design closely but not at those tests as closely yet.  They seem to cover all the downwards cases, as you say.  I wonder if hot water from below might create heat, but that isn't happening in your design.

[Sevio]

Continuing my research into boiler efficiency, it occurred to me that having the tepidizer inside the heat exchanger isn't optimal for trying to maximize throughput. So I've been experimenting with a design where it has its own basin above the heat exchanger and the polluted water spills down into the cold side of the heat exchanger. The problem is that normal water has to also flow on top of that and creates a huge mess of water and polluted water tiles swapping position constantly, instead of the neat dripping effect that you often see with small water spills in a normal base. It still works for a while, but eventually it reaches a stall like so:

This prevents clean water in the heat exchanger from flowing out. Any ideas how to force the nice dripping effect to happen without water cells jumping all over the place?

[Cilya]

Couldn't you swap the tepidizer and the pump so flows doesn't cross ? (tepidizer left, pump right, fully submerged two tile below, or something like that)

[Saturnus]

Just extend the length so it flows past the tepidizer before the heat exchanger. Then the clean water will still flow and spill over.

[Sevio]

@Cilya I've deleted that experiment and decided to try this layout instead, this should push the cold PWater past the tepidizer and up into the heat exchanger, no crossed flows then. Only downside is that it takes up the space I originally used for the cooling basin, so this build will be less space efficient.

@Saturnus I've tried a couple different positions of dropping the polluted water but there was always the chance of the stall above happening so far. It might be solvable but the tepidizer basement method seems the most promising to me right now.

[Saturnus]

@Sevio I wasn't sure but I did seem to remember that vents don't become overpressurized in polluted water (due to the lower mass per tile than regular water), so I didn't propose that exact thing before I had a chance to test it.