Looking for a better salt purifying/cooling situation

[SquarelyCircle]

So, I've got a salt water geyser. The obvious thing to do is to put it through a desalinator, making fresh water and salt. This requires 960 watts per 10kg packet of salt water. Furthermore, the salt water is 95 degrees. I could use the water for hot applications, but honestly I don't have that many. I want it coldish.

So, I came up with the following design in hopes that it would solve my problem... but by golly it's energetically expensive.
I heat the salt water up so that it becomes steam. The steam is absorbed by the steam generators, deleting about 5 degrees (?) worth of heat. I then cool the resulting water down to 5 degrees and send it to my base. I'm using 4 aquatuners because I want the water to be available immediately, as opposed to going through the cycle many times. Theoretically, I think the energy usage should be the same as if I had just one aquatuner and cycled the water.

Either way, 4 aquatuners is 4.8kW. The auto sweeper settup takes 480W (I don't know if it draw that continuously or only when in use). The steam generators are producing a pretty consistent 1kW. So, I believe this setup nets a cost of 4280W. If we subtract the cost of running the two desalinators, we could say it's a comparative cost of 3320W with the benefit of cooling 95 degree water down to 5 degrees rather rapidly.

That seems like a lot to me. So, I'm looking for either a design or pointers/tips that will help me to make a more efficient system.
If you have ideas about how to calculate electrical costs to make better decisions before building something, that would be helpful, too. I feel like I don't understand how the factors all compare, other than just calculating out the costs of buildings. Even though the buildings aren't necessarily on all the time, and the amount of cooling acquired may be different between the two setups, etc.

https://steamcommunity.com/sharedfiles/filedetails/?id=2081178197

[SquarelyCircle]

Of note, I don't have the middle sweeper hooked up to power anymore.

[gimmethegepgun]

Something to take note of is that if the water's being used in an electrolyzer there's no real reason to cool it down. It will output at at least 75C regardless, and it will almost always delete heat in doing so.

[Dracuras]

hm.. depending on why you need it that cold you could remove one aquatuner and it should come out around 20C. due to thermal mass mechanics even relatively cool water, say 22C+ would delete some heat when turned to gases by an electrolyzer. so 95C going in will destroy a lot of heat.

[Xilo The Odd]

its also possible to create a loop that will feed fluid back into the same aquatuner if its not cold enough if your trying to take what the steam gens output, it can also fit within a 2x5 space complete with input duct and everything using only a single tuner in this way helps with power drain. helps keep power drain down. if you want to bring it past boiling quickly there are other methods as well like dribbling it on a volcano's magma.

but yeah putting the water into the electrolizer hot is overall better for deleting heat.

[Exlax]

one of the things I like to do is use one aquatuner with the steam turbine setup like you have but instead of running the salt water through it run PWater through it in a loop with thermal liquid pipes into the reservoir that i am dropping the desalinated water into and then piping it into my base. that lets me store large amounts of cold water in my base and use it as a temperature control by putting it near my industrial areas so it radiates cold to the area around my hot machines.

[Strygald]

Most maps have volcanoes, why aren't you using the thermal energy from a volcano to turn the salt water into steam and then power steam turbine with it?

[XceptOne]

Your design ideas are actually quite good. It's just heavily overengineered for what you want to achieve (I think).

First, for most geysers it's not necessary to process all material as it is spawned. Instead you might want to use its average spawn rate as basis for following processes.
So, while your saltwater-geyser might spawn something like 12kg water per second, averaging that over its whole activity/dormancy-cycle will yield something between 2.8 and 3.2 kg water per second continuously.

Planning with that in mind, instead of the full 10kg/s pipe-flowrate, will lead to much smaller and also much less powerhungrys builds.
It also gives you an idea about the energy flows involved in a process.


My example will assume a constant flow of 3 kg saltwater per second.
For example, pump -> liquid reservoir (as a buffer, optional) -> liquid valve (set to 3 kg/s) -> ...rest of process...
There should be enough space to store all the liquid needed during the geysers dormancy phase.

Gauging average energy cost:
Liquid pump: can pump 10kg/s @ 240W, will pump only 3kg/s on average, meaning the pump is only running 3 out of ten seconds -> will use 72W on average
(doing thge same with a desalinator would mean pumping and cleaning 3kg/s saltwater costs 360W on average (reservoir and valve put behind the desalinator))

Aquatuners:
An aquatuner running water will move max. ~585 kDTU heatenergy per second.
Applying that to 3kg water results in a temperature change of ~46.7°C.
So using two(! instead of four) aquatuners will cool 3kg of water by ~93.3°C every second. That will get your water to near freezing (or below with smaller flowrates) and will cost 2.4kW.

We are at ~2.5kW powerconsumption now.

With an optimal steam temperature of 200°C a steam turbine, using 2kg of steam per second, will remove ~877kDTU/s of heat. So two aquatuners can run ~1.3 steamturbines resulting in a maximum of ~1.6kW power productcion (with power control station).

So, if build well you can desalinate and almost freeze those 3kg saltwater per second for around 1kW of power.
Proper-sized buffers behind that process will ensure you can meet consumption-spikes, if needed (most likely not, as you plan follow-up processes around those very same average flow rates).


Just for reference:
saltwater -> pump -> reservoir -> valve(3kg/s) -> steamchamber (saltwater turned to steam) -> steamturbines (steam turned to water, 4kg/s) -> reservoir -> valve(s) (3kg/s out, 1kg/s back to steamchamber) -> heatexchanger -> usable, clean, near freezing water
+
separate cooling loop with aquatuners in steamchamber and pipes connected to heatexchanger


Some random thoughts:

Desalinating in your steamchamber requires some sort of pressure control, in that 2 turbines will draw 4kg/s steam from the chanmber but the geyser only produces less than that. A part of the cleaned and cooled water needs to be reintroduced into the steamchamber (basically another reservoir + valve). I think you are doing that, but it's hard to tell.

The power draw of a single auto-sweeper can mostly be neglected. If you want to make sure, only allow it to work during a short time each day.

If you upgrade the cooling loop from water to using supercoolant (lategame), you will only need a single aquatuner and the whole thing turns power positive (again, with power control station).

You should still ask yourself if you really need that much cold water.
While 1kW power seems doable, thats still a lot of work. Usually, I would look for a cool water source (cool slush / polluted water) first, and use the hot stuff for oxygen generation for example).
All in all, building something involving two aquatuners or more should always trigger the question if you really need to do this (sometimes the answer remains yes, othertimes...)

[Angpaur]

Postat inițial de XceptOne:

An aquatuner running water will move max. ~585 kDTU heatenergy per second.
Applying that to 3kg water results in a temperature change of ~46.7°C.
So using two(! instead of four) aquatuners will cool 3kg of water by ~93.3°C every second. That will get your water to near freezing (or below with smaller flowrates) and will cost 2.4kW.

You need to think again about that part. A reminder - aquatuner always decreases temperature of any amount of liquid by 14C. So 2 aquatuners can reduce by 28C, no matter if there is 100g, 3kg or 10kg running in it per second. To move that 585kDTU per second you need to run 10kg/s of water/pwater.

But overall nice post.

[chaney]

Would running 10 kg/s through the Aquatuner, passing 3 kg/s out of the system and recirculating the other 7 kg/s eventually converge to (14 * 3 1/3) or about a change of 46.7 C in the output?

[SquarelyCircle]

Okay, without completely restating what you've said, the big takeaway that would improve my design is considering the average rate of resource production. If I calculate how much water I need to be cooling per second, I might be able to reduce this design to 2 or even 1 aquatuner, but produce the same net end result.

One of my reasons for not doing so originally is that I thought getting the steam to a higher temperature would increase the effficiency of the steam turbines. So, if I were able to get it to 200 degrees, even if it would eventually have a quiescent period, it would net more energy gain. I also assumed that the overall cost of cooling the water, whether it was by 1 or 4 aquatuners, would be the same. So, overall, I imagined that I was getting a savings by rushing the heating of steam to as high a temperature as possible.

While Xceptone's post is what I needed to hear, I think in reality, I'm going to take the earlier advice and just desalinate and pump this into my oxygen production. It'll take some repiping, but the O2 is currently using intermediate temp water, so it would be cheaper to pump this in and cool that stuff. Or find a better cool geyser?

Thanks everyone.

[XceptOne]

Postat inițial de Angpaur:

...
You need to think again about that part. A reminder - aquatuner always decreases temperature of any amount of liquid by 14C. So 2 aquatuners can reduce by 28C, no matter if there is 100g, 3kg or 10kg running in it per second. To move that 585kDTU per second you need to run 10kg/s of water/pwater.

But overall nice post.

Thanks,
you are right, and I should have explaind that one better.

So to clear things up and trying to answer this

Postat inițial de chaney:

Would running 10 kg/s through the Aquatuner, passing 3 kg/s out of the system and recirculating the other 7 kg/s eventually converge to (14 * 3 1/3) or about a change of 46.7 C in the output?

Err, I don't know, but I would guess it doesn't work well.

As Angpaur said, to get the max out of your tuner it needs to run 100% of the time with full 10kg packetsize.

Plus, personally I think looping a liquid/gas packet multiple times through the same pipe region is just bad design (except for closed loops, obviously). Mainly because you often end up with a system that is hard to control and unreliable. But also because you are heavily limiting throughput (if every packet loops twice, you are effectively at 5kg/s instead of 10kg/s behind that loop).
As a personal rule: loops shall always be closed -> no input/output after setup.

That means, for me, the easiest way to use an aquatuner in a controled fashion is to always have it in a separate, closed cooling loop, using water/pwater or supercoolant with full packet size.
Example:
reservoir -> heatexchanger (likely a bunch of metal tiles with radiant pipes running through) -> aquatuner(s)/bypass -> reservoir

The target material can then be cooled in the heatexchanger (in a single pass, hopefully; think another line of metal tiles with radiant pipes directly touching the first line, for example).

Placing a temp sensor (or two) in a good spot (behind the reservoir, or measuring the target material behind the heatexchanger, for example) and linking it to the aquatuner/bypass-construction will let you control the actual cooling power.


I hope that does clear that up a bit.

Postat inițial de SquarelyCircle:

Okay, without completely restating what you've said, the big takeaway that would improve my design is considering the average rate of resource production. If I calculate how much water I need to be cooling per second, I might be able to reduce this design to 2 or even 1 aquatuner, but produce the same net end result.

Yes

...and 2 tuners if using water/pwater as coolant

...and not so much improving your design directy, as much more getting your calculations closer to reality (and hopefully having explained why)

Postat inițial de SquarelyCircle:

One of my reasons for not doing so originally is that I thought getting the steam to a higher temperature would increase the effficiency of the steam turbines. So, if I were able to get it to 200 degrees, even if it would eventually have a quiescent period, it would net more energy gain.

Thats actually right, and the math I did assumes you will keep the temperature of your steamchamber as close to 200C as you can (turning of turbines if steam gets to cold ...that whole setup wouldn't be able to fully power two turbines anyway)

Postat inițial de SquarelyCircle:

I also assumed that the overall cost of cooling the water, whether it was by 1 or 4 aquatuners, would be the same. So, overall, I imagined that I was getting a savings by rushing the heating of steam to as high a temperature as possible.

Well, yes and no.
Cooling something by a constant amount will always cost a constant amount. Using an aquatuner that cost can be increased by either running smaller packets through it or using a coolant with lower shc (just don't do that). Otherwise there are no changes in power cost as using more tuners means they will run less.
But then, as tuners only move heat from the coolant to themselfes/their surrounding, using more of them doesn't create more heat either.
Both ways you are looking at the same amount of energy that a steamturbine can effectively recycle (again only run it when the steam is hot enough ...if it gets too hot, build more turbines)

Postat inițial de SquarelyCircle:

... I think in reality, I'm going to take the earlier advice and just desalinate and pump this into my oxygen production...

And in conclusion, thats exactly what i would have done, too.

But then, it's the small ideas that make me smile. Like using your saltwater as a heat-transfer-medium in an already planned cooling device and getting it cleaned up for free in the process. As I said, I liked that.
Actually I'm doing something quite similar: using oil to transfer geothermal heat (magma biome) to a couple of steam turbines, getting out free, cooled down petrol in the process. (Now, having written this, I guess most petrol refineries will work quite similar ...just a different viewpoint)

[chaney]

Postat inițial de chaney:

Would running 10 kg/s through the Aquatuner, passing 3 kg/s out of the system and recirculating the other 7 kg/s eventually converge to (14 * 3 1/3) or about a change of 46.7 C in the output?

I tested in sandbox mode, and the answer is "yes."
https://steamcommunity.com/sharedfiles/filedetails/?id=2084241166

There is an S=100 C water tank on the right. That is pumped into a Liquid Reservoir which feeds the Aquatuner at a constant 10 kg/s. The Aquatuner output has 3 kg/s let out through a Valve to the left at a final discharge temperature F. The other 7 kg/s is recirculated to the Reservoir. The steady state temperature in the Reservoir is then R = (7*F + 3*S)/10. With the Aquatuner, F=R-14.

Solving the equations:
R=67 1/3 C
F=53 1/3 C

So the total drop from source to final output is 46 2/3 C, which checks with the number of passes a typical particle makes through the Aquatuner (3 1/3) times the 14 C drop per pass.

I wanted to do the exercise to see if the game works as I expected. For the problem under discussion, throughput over 3 kg/s is irrelevant as that is the nominal source rate, so the recirculation was set to fit that net throughput and see what maximum temperature drop was achieved with a single Aquatuner.

The pump is obviously not operating efficiently (easily fixed) but the rest of the setup seems to be fairly robust to me as long as power is available. If you don't need so much temperature drop, intermittent operation with a single pass would save electrical energy, but this shows what you can get with the single cooling unit, and that XceptOne's thermal calculations can be implemented with the right architecture.

I look forward to the experienced voices pointing out the flaws in the partial recirculation loop, I'm eager to learn!

[Angpaur]

Postat inițial de chaney:

XceptOne's thermal calculations can be implemented with the right architecture.

Yes, this is true. Original post from XceptOne didn't mention recirculating water and without this included the calculations were wrong.

What you did here, people achieve by making a loop controlled by pipe termosensor connected to aquatuner. A 10kg packes are looped via aquatuner until desired temperature is achieved. However since aquatuner with each loop pass always lowers packets temperature by fixed 14C, then final temperature drop will always be a multiplication of 14.

What I like about your solution is that it doesn't need any automation - just a pure math to get exactly desired temperature. However it will work as expected only if incoming water temperature is constant. But this is easy to achieve if you use geyser water.

[chaney]

Yup. The minimum temperature drop will be 14 C with 10 kg/s throughput, but by reducing the throughput (by adjusting the Valve down,) the temperature drop will be increased arbitrarily and continuously (within the limits of the calculation precision and material freezing point.)

Unsteady input temperatures or interruptions in inflow or power will get different results, the calculation is for the stationary case, and there is a period of transition before the system reaches that steady state. Those may be problems, but with buffering it should be fairly robust and stable.