Water plumbing heat experiment

[Sevio]

6 minutes ago, Saturnus said:

Maybe I should mention that one of the things I noticed in the original heat exchanger design was that they have equal direction of flow, ie. the hot and cold comes the same side so that means the best you can achieve is equalization to the median.

Yes, one of my earlier boilers did use a cocurrent exchanger in the cooling tank which I realized was not optimal. While Final Boiler Fantasy V does use a countercurrent heat exchanger in the cooling tank, I ended up removing the granite gas bridges and wire bridges I initially put there because it was too effective at cooling down the output water, causing the freezing protection switch to come on too often. In retrospect, I could have left them there and simply cranked up the tepidizer to counteract that, also resulting in increased throughput.

[Saturnus]

3 hours ago, Sevio said:

After letting it run for a while with the airlock doors on 10x super speed I ended up with the following results:

1. Hot cooled down to 53.9 C, Cold heated up to 55.9 C. t(delta)=53.5C+49.0C=102.5 C - t(avg) = 54.9 C (96% heat retention)
2. Hot cooled down to 49.4 C, Cold heated up to 49.3 C. t(delta)=57.5C+42.4C=99.9 C - t(avg) = 49.35 C (86% heat retention)
3. Hot cooled down to 59.5 C, Cold heated up to 54.8 C. t(delta)=47.4C+47.9C=102.7C - t(avg) = 57.15 C (100% heat retention)

 

The airlock actually give a worse overall result in my view. But may I ask how you get 47.4+47.9 to be 102.7? Shouldn't it be 95.3?

Anyway, I ran another test where I made both channels zig zag alternatively. And I increased the pressure to full 800kg/tile and let it run for 100 cycles.

1. Hot cooled down to 46.5 C, Cold heated up to 45.4 C. t(delta)=60.4C+38.3C=98.7C
2. Hot cooled down to 50.2 C, Cold heated up to 49.4 C. t(delta)=56.7C+42.5C=99.2C
3. Hot cooled down to 56.8 C, Cold heated up to 52.6 C. t(delta)=50.1C+45.7C=95.8C

The tests are pretty conclusive. Zigzag channels are far superior to straight channels, and even overcomes the heat destruction bug to be the most efficient design. Double zigzag channels does not increase efficiency. Interestingly, the straight channel was unchanged in any way but lost a tiny bit more performance over time compared to the zigzag channels by the increased water pressure. If you use straight channel, less water pressure is better. If you use zig zag channels, more water pressure is better.

ZZ_squiggly.sav

[Le0n1des]

When you maximize the pressure - you kill the heat destruction bug, which is basically - the same bug that enables to instantly cool a column of fluid by cooling only the thin top layer... With maximum pressure there is no thin top layer

[Sevio]

7 hours ago, Saturnus said:

The airlock actually give a worse overall result in my view. But may I ask how you get 47.4+47.9 to be 102.7? Shouldn't it be 95.3?

I copy-pasted that list from your post and filled in the new values of my experiment, but it looks like I forgot to fill in that one value, my bad. I fixed it in the original post as well.

So how do you come to the conclusion that the airlocks give a worse overall result? Experiment 1 of my airlock test reached a temperature delta of 102.5 C and heated the cold fluid up to 55.9 C. Your version of the one-way zig zag channel reached 54.4 C for the cold fluid and while it did have a higher total temperature delta, that was entirely due to the hot fluid cooling down more than it should have due to heat destruction,

There are two reasons why I'm focusing on the heat destruction bug and the cold fluid's output temperature rather than the total temperature delta in this test setup: I'm not willing to rely on a heat exchanger that takes advantage of that heat destruction bug (it could get fixed in my perhaps optimistic expectation), and secondly the throughput of a boiler is limited by the highest possible temperature that the cold fluid can reach when it comes out.

[eggsvbacon]

Any tutorials on making a simple system for laymens who just want to have fun in clean wawa?

I came here because I started experimenting with water tepidizers, hoping someone had figured the correct way to leverage water pressure sensors, length of dig etc to vaporize every drop in the short time they power on and off. 

@SevioThis is some wizard stuff that goes over my capacity for understanding, really neat boiler thing, with the The First Law Cleaned.sav but, what's that thing in the top left do? Is that an oxygen liquifier? Seems to be broken when I run it, the liquid cooler thing starts taking heat damage when I set it to input water from the pressure thing, never pumps clean water down, water pressure over there doesn't go up, the gas intake thing above says it's over pressurized too, only works if I break a wall on it. Pretty sure that's not what you intended.

What's the thing above the fancy new boiler? What's that gas pipe for at the top? Why is it turned off? Why do you have air locks separating oxygen? Why is the contaminated pool for the dupe off gasing into the clean o2? I assume the dupe is just there, not really currently to do anything. Not criticizing, just curious as to what everything is. I looked at all the posts, but I couldn't really tell what was what and what connected where or how a laymen might apply this to their game.

Also, how power intensive is this? I saw the big hydrogen bank thing, also I was fascinated, it's reusing hydrogen? I never used a hydrogen generator, but it looks like no new gas is going in? Are all these battery banks to run the newer water cleaner?

tl;dr I'm looking for a simple way for a laymen to make clean water that will not be broken by updates that I can replicate, especially if it can be done without debug and or a trainer.

[Sevio]

@eggsvbacon I've gone through a lot of iterations to get to the final boiler design, so in any of my posted savegames there will probably be an old, inactive boiler design that I discarded.

If you're looking to exploit only the tepidizer for boiling water, you basically want to drip feed it water and activate the tepidizer with a hydro switch when it gets above a few kg of water level. How much water to activate at and how fast you can drip feed it need fine tuning depending on the input water temperature. But I wouldn't say that is safe to rely on for future updates.

1 hour ago, eggsvbacon said:

What's the thing above the fancy new boiler?

Probably some leftover thermal experiments I did at the start of this thread.

1 hour ago, eggsvbacon said:

What's that gas pipe for at the top? Why is it turned off?

It's for pulling polluted oxygen out of the boiler, which evaporates slowly but steadily wherever polluted water is exposed to air (or steam)

1 hour ago, eggsvbacon said:

Why do you have air locks separating oxygen?

They're not necessary in debug mode but they represent access airlocks that dupes could use to service the boiler, and for getting access to build the upper parts.

1 hour ago, eggsvbacon said:

Why is the contaminated pool for the dupe off gasing into the clean o2? I assume the dupe is just there, not really currently to do anything.

I like to keep a dupe around in my debug saves in case I need to plant something in a pot, or chisel some sculptures as it turned out with some of my boiler designs.. So I made him a little habitat with all he needs, every once in a while I clean out his polluted water pool and refresh his oxygen, but there's quite a lot in the room so the CO2 and polluted oxygen don't become a problem until the oxygen gets low.

1 hour ago, eggsvbacon said:

Also, how power intensive is this? I saw the big hydrogen bank thing, also I was fascinated, it's reusing hydrogen? I never used a hydrogen generator, but it looks like no new gas is going in? Are all these battery banks to run the newer water cleaner?

There's no magical hydrogen reuse going on. It's just a large room with a silly amount of hydrogen in it. One pump is constantly pulling hydrogen out of the room to feed the generators, but there is so much in the room it takes forever to empty it out. I built it so power wouldn't be a problem while I focused on the boiler design.

The huge battery bank is what I use to measure the power usage of the boiler. I can separate the generators from the batteries with a switch at the start of the cycle and let it run until the next one starts. Then look at how much the batteries have left to figure out how much power was used during the cycle.

Final Boiler Fantasy V uses about 1.43 kW of power steadily so you would need 2 natural gas generators running constantly to power it and have a bit of surplus. Slightly more if you're running the tepidizer at maximum rate as well.

1 hour ago, eggsvbacon said:

I'm looking for a simple way for a laymen to make clean water that will not be broken by updates

TBH, boilers are a fun challenge to build but as it turns the Water Purifier is a far more convenient way to make pure water. There's apparently enough sand on the map to run one for thousands of cycles and just one of them will run as fast as the fastest boilers for a fraction of the power.

[eggsvbacon]

@Sevio I meant the big line of manual airlocks on the right hand side from the base to the ceiling, not a part of any structures, what's that about? Unless I misunderstood your answer.

That's interesting about the water purifier, but, it doesn't kill germs, which is rubbish, which requires making use of hot areas, tepidizers etc, which, by the way, I'm completely incompetent when it comes to power management, things start breaking, but looking at your build gives me a bit better understanding of transformers.

I find myself quite amused by the liquid and gas physics in the game, I keep remaking my games however, nothing I ever make is elegant, I've never been good at planning much, I'm playing checkers while everyone is playing chess.

To be fair though, my neurological sleep disorder can get in the way of higher modes of thinking at times, which is quite often, which is why I came here looking for a tidy, fairly easy to build system to purify water in a small loop, because my sub-concious demands it.

I thought it was a tepidizer feature, but I did find it a bit odd that they turned off. I tried placing mesh and standard tiles with the pressure meters on them, water would drip around, unfortunately, more water dripped down than evaporated. I wasn't sure how far I would need to expand deeper and how many I'd need to add to do it. Now that I know it isn't meant to be a feature, not sure I want to bother.

I'm not looking to purify the map, just a bit of waste water from the showers/toilets and from various equipment. After viewing the forum I learned that liquid oxygen making is a thing, so I'm curious about that. Hopefully haven't derailed your topic too much.

Anyway, thanks for the thoughtful replies.

[Sevio]

8 hours ago, eggsvbacon said:

I meant the big line of manual airlocks on the right hand side from the base to the ceiling, not a part of any structures, what's that about? Unless I misunderstood your answer.

I noticed the top half of the room was getting pretty hot because of all the transformer heat, while the bottom half of the room got very cold from the cooling tank. So that big line of airlocks from bottom to top was an experiment at using them as heat pipes. Because the underlying material (Steel Door <DO NOT TRANSLATE>) has a heat conductivity of 54, I wanted to see how well they would conduct heat from the cold half of the room to the hot half. It does work pretty well considering the distance, but it's still very slow. It's probably not very practical in a real base because they have to be closed to work.

8 hours ago, eggsvbacon said:

That's interesting about the water purifier, but, it doesn't kill germs, which is rubbish, which requires making use of hot areas, tepidizers etc

Do you use the cooking station to make more advanced foods? If you're relying on uncooked meal lice which is completely feasible to use forever in the current update, there's actually no downside to having germs in your clean water as it doesn't transfer to dupes in any way other than via cooking recipes. But conceptually I can see how that would feel wrong, and I'm sure it won't stay like this forever. You can also let your purified water sit in a tank for a while to let the germs die off.

8 hours ago, eggsvbacon said:

I came here looking for a tidy, fairly easy to build system to purify water in a small loop

Unfortunately there's not really a very simple way to purify water via boiling because you can't do it at a reasonable speed and power efficiency unless you reclaim most of the heat out of the steam and clean water. That requires a fairly large heat exchanger area. Even the highest throughput boiler configuration from Saturnus in this thread (about 5500 g/s if I recall correctly) only just barely tops a single Water Purifier. But since that uses the tepidizer constantly and outputs hot water, it can do with a smaller heat exchanger than a colder boiler.

[0xygen]

So, the Liquid only allowed to falling down directly to the water tank from the Liquid Vent to avoid the bug, and even the naturally flooding and falling down water causes the bug. My Thermo Regulator cooling need to be fix too

[Saturnus]

@Sevio It'll interest you to know that I have done a few experiments where I seem to be getting over 9kg/s output with a single aquatuner. The version in the save file puts out 9445g/s stable (5667kg/cycle). Might go up a bit more too as it seems to climb in throughput very very slowly. The goal is now 10kg/s with a single aquatuner.

There's no heat exchanger or tepidizer here as I'm only interested in the boiler/condenser optimization. But it does have separated polluted and clean water so no chance of germ migration. There's a cooling reservoir for the aquatuner and polluted water is pumped in at the temperature it would have been at if pre-heated by a tepidizer.

I think they might have tweaked the aquatuner efficiency in a silent update as it seems like a dramatic step up. I did include some new ideas though so maybe not? 

 

Test Bed Boiler.sav

[Sevio]

Very interesting results @Saturnus, that's an amazing throughput. The addition of oil allows us to put a small amount of mass in the bottom tiles of the aquatuner instead of forcing us to have a large basin full of polluted water. The save file clearly shows it works, even though the math says it shouldn't...

The polluted water is being sent to the Aquatuner in the pipe at 98.8 C. To evaporate it, it needs to go to 122.3 C so that's a 23.5 K difference.

An Aquatuner can output 6 * 10000 * 14 = 840000 J/s of heat with polluted water as the medium, which gives us 840000 J/s / 6 J/g/K / / 23.5K = 5957.45 g/s of theoretical throughput... I'm very confused as to what's going on here unless I have my math wrong.

[Saturnus]

52 minutes ago, Sevio said:

Very interesting results @Saturnus, that's an amazing throughput. The addition of oil allows us to put a small amount of mass in the bottom tiles of the aquatuner instead of forcing us to have a large basin full of polluted water. The save file clearly shows it works, even though the math says it shouldn't...

The polluted water is being sent to the Aquatuner in the pipe at 98.8 C. To evaporate it, it needs to go to 122.3 C so that's a 23.5 K difference.

An Aquatuner can output 6 * 10000 * 14 = 840000 J/s of heat with polluted water as the medium, which gives us 840000 J/s / 6 J/g/K / / 23.5K = 5957.45 g/s of theoretical throughput... I'm very confused as to what's going on here unless I have my math wrong.

That's why I wonder if the aquatuner stats may have been changed in the code because given the stats on the info sheet it should really be impossible. And just in case it's now at 9600g/s output just by letting it run longer. The fact is that I measured the feed pump to run 100% of the time, 100% full all the time. So it's fed 10kg/s to the boiler, it's just that the condenser haven't caught up yet.

Another idea on what's going on is that the aquatuner doesn't see the polluted water at all. It only sees the crude oil.

EDIT: I have to moderate that. Every 4-5 cycles the feed pump drops out for a second meaning a loss of 4-5 packets so the feed rate is around 9950g/s on average. After running on about 40 cycles so is the output.

I also have to mention the other clever innovation. Without the mechanical airlock and the connected wire bridge the oil temperatures can fluctuate wildly, from under 110C to over 160C. The whole boiler section is optimized to keep temperatures as stable as possible.

[Sevio]

51 minutes ago, Saturnus said:

I also have to mention the other clever innovation. Without the mechanical airlock and the connected wire bridge the oil temperatures can fluctuate wildly, from under 110C to over 160C. The whole boiler section is optimized to keep temperatures as stable as possible.

I see, do normal tiles not do the job well enough? In any case it helps preheat the polluted water as it flows overtop the Aquatuner.

Maybe, just maybe this design of yours is actually capable of triggering the heat destruction bug in reverse. The amount of polluted water on top of the crude oil is always very tiny and is constantly switching between steam and polluted water. I wonder if the crude oil's high temp is causing the polluted water to insta-vaporate before it has a chance to properly stabilize and draw the heat it should.

[Saturnus]

12 minutes ago, Sevio said:

I see, do normal tiles not do the job well enough? In any case it helps preheat the polluted water as it flows overtop the Aquatuner.

Maybe, just maybe this design of yours is actually capable of triggering the heat destruction bug in reverse. The amount of polluted water on top of the crude oil is always very tiny and is constantly switching between steam and polluted water. I wonder if the crude oil's high temp is causing the polluted water to insta-vaporate before it has a chance to properly stabilize and draw the heat it should.

I tried many different set ups but the mechanical airlock method works by far the best.

That may very well be the case that there is a reverse heat bug taking place. I do notice one strange thing. Most of the time no dirt is created even though it's creating stream as fast as you can pump water to it.

The heat in the mechanical airlock is technically energy taken from the aquatuner. That's why, I think, it works so well. The wire bridge transfers heat from the crude oil to the mechanical airlock which in turn is cooled by the incoming water so the temperature is kept in check while at the same time heating that water before it meets the hot crude oil.

Note that you can't vary the numbers very much before the balance is broken. The amount of crude oil, the pump trigger point etc. All of them can't be changed too much before you lose a lot of efficiency. 

[Saturnus]

@Sevio

What if I told you there was a way to have a 10kg/s boiler that doesn't even need to run a single aquatuner more than about 50% of the time? You probably wouldn't believe it, right? I mean, that'd require you to preheat water to over 108C...

I give you The Supercharger.

Test Bed Supercharger.sav

[Sevio]

Does this work off your previous discovery of having a pump create steam from nothing?

[Saturnus]

24 minutes ago, Sevio said:

Does this work off your previous discovery of having a pump create steam from nothing?

This is not an exploit. And not based on a random bug. It's just very clever engineering. What I really discovered here is a sweet spot where the tepidizer stay on 100% of the time because it's got just the right temperature water at just the right pressure rushing past it to keep it from turning off. And trust me, it took me hours to test all the possible variables.

What is happening here is, as you can see if you try to run it, carefully matching the pressure at the left end of the tepidizer to be between 380kg and 400kg per tile while pumping it away at the other end. Liquid, gas and wire bridges are plastered all over to try and keep the water in the left most tile of tepidizer below the 95C cut-off point. Again the airlock is what helps the most at regulating the temperature. In the 10kg/s distiller I can get it to output up to 113.1C  (briefly, 112.2C stable) when the input water is 46C. However, if the input water goes up to 48C the output temperature drops to 103C, and at 50C it's a measly 98C. Going below 40C feed drop the output temperature by about 2C for every 1C lower the input is until it reaches about 98C at which point it continues to drop on a 1C to 1C basis as you'd expect.

I actually don't know if the pumps creating steam from nothing works in the OI. Anyone tested it?

[Sevio]

When you first posted that design I did load up your save to check it out and I noticed the result would be far worse than the one demonstrated if I started tweaking the values like water level and pump speed with a valve, so well done on finding this local maximum. Since this is very sensitive to input water temperature and most polluted water generating structures output at 40 C, condensing the steam that you'd create after boiling sounds like a bit of a challenge though...

[Saturnus]

26 minutes ago, Sevio said:

When you first posted that design I did load up your save to check it out and I noticed the result would be far worse than the one demonstrated if I started tweaking the values like water level and pump speed with a valve, so well done on finding this local maximum. Since this is very sensitive to input water temperature and most polluted water generating structures output at 40 C, condensing the steam that you'd create after boiling sounds like a bit of a challenge though...

Yeah. The updated 10kgs distiller I posted in it's own thread solves most of this by aquatuner work rate being directly depended on the input water temperature and the heat exchanger/inter-cooler being purposely less efficient/smaller. That means the heat exchanger on it's own will heat up input water too much but the cooling from the aquatuner loop keeps it's in balance. That gives you a fairly large input water temperature range but it optimised for about 30C. This is chosen due to the main sources are probably shower and lavatories which will output at their building temperature, and that should be under 30C to be comfortable. Add in some water at 40C machines and it should be about 30C. However. Tweaking the heat exchanger/inter-cooler to be more or less effective to suit you particular build should really be straight forward.

EDIT: Oh, and try changing statues/canvas' for tombstones. 800kg of 2 granite goodness. The most massive building in the game.
 

 

[np5jx61]

不要在游戏中套用现实世界的知识。这个游戏散热是以水循环的方式，滴水过程中损失热量是因为这个水循环散热的代码造成的

[Le0n1des]

Getting back to the topic... So is using tombstones better for heat transfer, due to the ability to pack more mass in the same space or statues, due to temperature equalization of 3 tiles?

[Saturnus]

24 minutes ago, Le0n1des said:

Getting back to the topic... So is using tombstones better for heat transfer, due to the ability to pack more mass in the same space or statues, due to temperature equalization of 3 tiles?

Tombstones are only 2 tiles high so they can't substitute statues in all cases but in the cases it can, it's 3 times better than statues.