Precision aquatuner cooling - Am I missing anything?

[Nebbie]

So I came across this and replicated it for my base's cooling loop, but found that it had a problem with non-full pipes in that the liquid would keep circulating and dip below the desired temperature because it always does one run through the aquatuner (and I'd rather not have my polluted water freeze). With the modifications, the piping's a bit of a mess, but I don't see how I could improve it, and I'm not completely sure that I haven't introduced a new problem; is there anything obvious I'm missing to improve the design?

Spoiler

 

[abud]

Yours is a different concept than the one that you refer to. For example, in original design aquatuner will always get full packet continuously (output from the reservoir), but yours not (it may goes up instead). And aquatuner without constant input probably not the best idea. We can disable aquatuner from time to time, but input coolant should be always available full packet.

 

If you don't mind me asking. What do you want too cool which makes you need finer control than default aquatuner 14C range? I'm just curious.

[Nebbie]

38 minutes ago, abud said:

Yours is a different concept than the one that you refer to. For example, in original design aquatuner will always get full packet continuously (output from the reservoir), but yours not (it may goes up instead). And aquatuner without constant input probably not the best idea. We can disable aquatuner from time to time, but input coolant should be always available full packet.

 

If you don't mind me asking. What do you want too cool which makes you need finer control than default aquatuner 14C range? I'm just curious.

You mean less than a full packet could get in the aquatuner? That'd require the packet get into the loop non-full to begin with.

I'm really just pursuing perfection on exactly how my base is cooled. It might be useful later, for things like sleet wheat.

[abud]

I mean this pipe will never empty in the original design

You can take it out from loop, but put replacement after you took it out, probably return from the heated coolant. Something like this for example

This way, the one that exit the loop (probably because cold enough) will get replacement from top-right (assume it contains heated coolant return from far away).

What will happens if that bottom right output is back-up though? In both original or yours, there is no protection against liquid that too cold and may icing.

 

Another concept is what I usually use, where coolant doing little loop until reach desired temp then released.

Only if you interested to see alternatives to do things.

Spoiler

 

1. Temp sensor to prevent steam chamber getting too hot. set to above 200

2. Pipe temp sensor to protect aquatuner. For PW coolant, set to above 0

3, Pipe element sensor to detect back-up. set to PW

When they backup or steam getting too hot, aquatuner will standby until cleared + extra 10 second (set by filter gate).

 

 

[Saturnus]

On discord I sometimes stipulate that you don't need to cool a sleat wheat farm to freezing temperatures since they're fine with up to 5C, and you can do that by cooling the water to 2C. Often I get people saying it's impossible to precisely keep water at that temperature with aquatuners. But it's actually simple, and similar to the approach by @abud above.

What you need is to fill up a reservoir. Doesn't need to be completely filled but 2/3rds to start with at least. This is your thermal buffer that makes sure that changes in water temperature is slowed down.

Then it passes by a temp sensor. If above 16C the 1st aquatuner is turned on the water is cooled and returned to the reservoir cooling the water in it down slightly. If 16C or below the 1st aqautuner is turned off and the water bypasses it to the 2nd aquatuner which is just on a NOT gate so only one of the aqautuners is on at any one time. The 2nd aquatuner cools the water to 2C (actually between 1.7C and 2C depending on how full the buffer reservoir is).

The maximum flow rate of the system is equal to: 140 / (Tin-Tout) where Tin is input temperature and Tout is desired output temperature. The 140 is from the maximum flow rate (10) multiplied by the temperature drop per pass (14). For example 95C geyser water cooled to 2C for sleet wheat gives a maximum flow rate of 1.5kg/s output. Enough for 45 sleet wheat plants.

 

[Qubit]

@Saturnus,

Let’s posit the following requirements:

1.  Want to recover some of the AT’s energy consumption, so using steel construction to permit its placement In a steam chamber below a turbine.

2.  Want to minimize steel consumption, so limiting ourselves to a single AT. 

3.  Still want very narrowly temp controlled water for sleetwheat.  

Does the formula for calculating water throughput change with the loss of an AT?  (Does the 140 drop to 70?)

 

[bobucles]

20 minutes ago, Qubit said:

Does the formula for calculating water throughput change with the loss of an AT?  (Does the 140 drop to 70?)

The aquatuners are split by a NOT gate, so exactly 1 is active at any time. The reservoir tuner will run until cold, while the farm tuner will stall once the farm pipe is full.

Once the aquatuner is in a tiny wet box, Setting up a steam system is pretty simple. There are countless threads about it.

The second aquatuner can be removed, but it looks pretty tricky. Ideally I think you want to chill the main reservoir down to 16C, then give it one final pass for the farm. The tricky part is getting that final pass without disabling the aquatuner because once the water is down to 16C, you want the tuner to stop chilling the storage. So work until it's chilled, or work to pass the frozen water out. Half the fun is figuring out the best setup, good luck!

[Saturnus]

1 hour ago, Qubit said:

1.  Want to recover some of the AT’s energy consumption, so using steel construction to permit its placement In a steam chamber below a turbine.

Easily done. Keep in mind the one posted above is a concept set up. You need to add to it.

1 hour ago, Qubit said:

2.  Want to minimize steel consumption, so limiting ourselves to a single AT. 

No idea why you think aqautuners must be steel. They can be gold amalgam even in a steam turbine set up.

1 hour ago, Qubit said:

3.  Still want very narrowly temp controlled water for sleetwheat.  

No problem. Nothing needs to change at all.

See set up and save below. Note the steam generates about 300W constantly. To avoid overheating the steam turbine on the self-cooling set up power is occasionally cut to the aquatuners resulting in a total through-put that is about 98%-99% of the one calculated. Only gold amalgam and gold is used everywhere.

Note the system below also has a fail safe if there's no input water or if output is blocked, as well as a manual reservoir filler.

 

Timeline.sav

[Qubit]

11 minutes ago, Saturnus said:

See set up and save below.

Cool!   Will study.  Slowly I learn.  But I learn.   Thank you!

*sigh*   I've been playing ONI since late November of last year and somehow never noticed that aquatuners have a base overheat temp of 125, not 75.

[abud]

This is my first time understanding double aquatuner for precise control too. I was about to say, If you insist on single aquatuner, give up on very precise temperature control. Good thing that @Saturnus can convince @Qubit.

Single aquatuner without direct cooling, but way worse than setup above. Set to 3C

Spoiler

 

[Qubit]

Okay, for those willing to further coach this noob, a couple of followup questions:

1) Am I understanding from this that 143 degree steam is a magic number below which the 95 degree output water from the turbine alone will suffice to self cool the turbine? 

2) What purpose does blocking one of the input ports on the turbine, if any, serve in @Saturnus's example, given the relatively low temperature steam with which the turbine is dealing?  Are we actually trying to slow heat deletion?

Sorry-- I'm sure these are very elementary questions.  Despite playing for awhile I've only been closely following the forums for a few weeks.  It was only about a month ago that I finally grokked that a pipe run into the output port of a bridge allowed for flow prioritization.   Before that, omgosh ... it was the dark ages.

[Saturnus]

34 minutes ago, Qubit said:

Okay, for those willing to further coach this noob, a couple of followup questions:

1) Am I understanding from this that 143 degree steam is a magic number below which the 95 degree output water from the turbine alone will suffice to self cool the turbine? 

2) What purpose does blocking one of the input ports on the turbine, if any, serve in @Saturnus's example, given the relatively low temperature steam with which the turbine is dealing?  Are we actually trying to slow heat deletion?

1) Yes, for 4 ports open. See 2) below.
2) You got the save. See what happens if you remove the tile blocking the 5th port and let it run for a while.

One mistake in the save file set up though. Move this element sensor one step right like this.

And if you didn't get to it. With all 5 ports open the steam turbine produces 80W more power but also after a while heats up above 100C shutting it down for an extended period as a self-cooling set up takes a very long time to cool down just 0.1C when it's overheating. The better option is to block the port, sacrifice 80W power and have it run forever.

[Qubit]

Roger! 

[KittenIsAGeek]

4 hours ago, Qubit said:

@Saturnus,

Let’s posit the following requirements:

1.  Want to recover some of the AT’s energy consumption, so using steel construction to permit its placement In a steam chamber below a turbine.

2.  Want to minimize steel consumption, so limiting ourselves to a single AT. 

3.  Still want very narrowly temp controlled water for sleetwheat.  

Does the formula for calculating water throughput change with the loss of an AT?  (Does the 140 drop to 70?)

 

You can do the same thing with a temperature sensor and a valve after a single aquatuner. If, after output, the water is warmer than 2c, it feeds back into the reservior.  This means that when the reservoir water is 16c, it will pass through the aquatuner, drop to 2c, and the valve will turn on and divert the packet out to your cold storage.

Note that you'll need to make sure your output to cold storage never gets full, or if it does that you can shut off the aquatuner.  Otherwise you'll start breaking pipes.

[Saturnus]

21 minutes ago, KittenIsAGeek said:

You can do the same thing with a temperature sensor and a valve after a single aquatuner. If, after output, the water is warmer than 2c, it feeds back into the reservior.  This means that when the reservoir water is 16c, it will pass through the aquatuner, drop to 2c, and the valve will turn on and divert the packet out to your cold storage.

Note that doing this will reduce maximum possible output by 75% as the with the tightest possible distance between reservoir and the first temp sensor without the reservoir itself being steel as illustrated allows for the output to fluctuate between 1.7C and 2.0C with about equal amounts of each 0.1C interval so returning everything below 2.0C to the reservoir effectively cuts the output to just 1/4th the calculated possible output.

All in all, a very bad idea.

[Nebbie]

7 hours ago, abud said:

,,,

What will happens if that bottom right output is back-up though? In both original or yours, there is no protection against liquid that too cold and may icing.

 

Another concept is what I usually use, where coolant doing little loop until reach desired temp then released.

Only if you interested to see alternatives to do things.

  Hide contents

 

1. Temp sensor to prevent steam chamber getting too hot. set to above 200

2. Pipe temp sensor to protect aquatuner. For PW coolant, set to above 0

3, Pipe element sensor to detect back-up. set to PW

When they backup or steam getting too hot, aquatuner will standby until cleared + extra 10 second (set by filter gate).

 

 

Yeah, back-up is the big issue here, and needs automation to detect. I continue to be surprised by how much automation is necessary to get basic things working perfectly in ONI. Looks like for basic base cooling my best option's just a liquid shutoff (hooked up to an above 14 C sensor, while it can go lower, I don't want to risk freezing in nearby pipes that contain water) into the aquatuner as a bypass of the loop. Interesting idea with the pipe to pipe temperature exchanges, as well.

3 hours ago, Saturnus said:

...

No idea why you think aqautuners must be steel. They can be gold amalgam even in a steam turbine set up.

...

Doesn't this require a layer of liquid on the bottom of the aquatuners in the steam chamber to exchange heat, though?

[Saturnus]

5 minutes ago, Nebbie said:

Doesn't this require a layer of liquid on the bottom of the aquatuners in the steam chamber to exchange heat, though?

Not necessarily. As long as you have a large enough thermal buffer. That can be done by just having more steam.

With 500kg steam per tile you don't need neither petroleum nor temp shift plates.

[KittenIsAGeek]

2 hours ago, Saturnus said:

Note that doing this will reduce maximum possible output by 75% as the with the tightest possible distance between reservoir and the first temp sensor without the reservoir itself being steel as illustrated allows for the output to fluctuate between 1.7C and 2.0C with about equal amounts of each 0.1C interval so returning everything below 2.0C to the reservoir effectively cuts the output to just 1/4th the calculated possible output.

All in all, a very bad idea.

If throughput is your goal, you are definitely correct.  However, I was replying to the question on if it could be done with a single aquatuner.    To be clear, in case I mistyped on my original post, it should return everything warmer than 2c to the reservoir.  Everything colder should go into cold storage.

This is the design I've used:

Spoiler

Hot water comes in at the upper left, cold water goes out at the lower right.  It is not fast, but it works.  If you have  a larger area for the aquatuner, so that you keep the temperature under boiling, then you can run without steel.  If you're using a steam turbine, then the valve needs to be steel.  You can redesign it like so to avoid using steel:

Spoiler

In this case, you can add a second thermo sensor so that the turbine only kicks on when the steam is over 160c. 

 

HOWEVER... I do agree with with you, @Saturnus, for greater throughput, you want a second aquatuner instead of a valve.  

 

 

[Nebbie]

1 hour ago, KittenIsAGeek said:

...

If throughput is your goal, you are definitely correct.  However, I was replying to the question on if it could be done with a single aquatuner.    To be clear, in case I mistyped on my original post, it should return everything warmer than 2c to the reservoir.  Everything colder should go into cold storage.

This is the design I've used:

Hot water comes in at the upper left, cold water goes out at the lower right.  It is not fast, but it works.  If you have  a larger area for the aquatuner, so that you keep the temperature under boiling, then you can run without steel.  If you're using a steam turbine, then the valve needs to be steel.  You can redesign it like so to avoid using steel:

  Hide contents

In this case, you can add a second thermo sensor so that the turbine only kicks on when the steam is over 160c. 

 

HOWEVER... I do agree with with you, @Saturnus, for greater throughput, you want a second aquatuner instead of a valve.  

 

 

One of my initial setups had the valve inside, made of lead, which didn't go very well, since I hadn't realized they had an overheat temperature unlike sensors. Ultimately I think it's best practice to only put things inside the steam chamber that absolutely must be, as unsealing/resealing it is not great if something needs to be changed.

Definitely will try a setup like this in my next base, thanks.