I'm obsessed with cooling loops.

[Half a fork]

Hey guys! 

I like to spread that cooling power around. The link is a demonstration of my power brick cooling loop. It's unnecessarily complex, but is oh so satisfying. What fun and interesting methods do you use to cool stuff down?
 https://www.youtube.com/watch?v=-1lq7MkRtQ8&feature=youtu.be

 

If you really wanna know what's going on... (TL/DR: it's a fancy cooling loop is all, but I'm proud of it).
It has three parts; the boiler, the ice box, and the double cooling loop.

Boiler:
The boiler consists of two aquatuners and a steam turbine. One aquatuner services the main ice box (half full of petroleum), while the other services a smaller back up ice box below it. 

Ice box:
If the coolant is above -30c in either the main or the back up ice box it is pumped out into the aquatuner. The main tank's coolant is then checked for temperature; if it is under -30 then it is piped back into the ice box; if it is still above -30 after going through the aquatuner once it is piped into a liquid reservoir. Coolant from the reservoir will be directed back to the aquatuner, but only once the pump stops drawing coolant out of the main ice box. This means if the ice box is too hot it will first drain out 5000kg, then cool. This has the advantage of reducing the thermal capacity of the ice box so that it can cool a smaller volume of petroleum, faster, for the higher priority industry at the bottom (i.e. the power brick). 
Once the lower portion is cooled to -30 and the pump is deactivated the coolant in the liquid reservoir is looped through the aquatuner until it is ready to be dumped back into the main icebox. 

Double cooling loop:
Coolant is piped around the power brick and brought back to be temperature checked. If below 20c it continues looping with top priority; if above 20c it is directed to a liquid reservoir (on the left) ready for cooling. 
It exits the cooling reservoir at secondary priority into the main ice box and loops through before returning to be temperature checked. If the coolant is still above 10c it continues to loop at top priority, otherwise it enters the secondary liquid reservoir (on the right). Here it acts as a cooling charge, ready to be dumped into the power brick cooling loop at secondary priority. 

__________________________________________________________________________________________________________
At some point I'm going to crack a way to attach multiple cooling loops to a single line... I'm almost there with distinct graded loops!

[EleriusX]

Nice. I would point out that passing petroleum through an aquatuner isn't the most efficient cooling/DTU. It is more than twice as efficient to use polluted water instead, early game, and then more than 4 times more efficient to use super coolant, in the late game. The reason is that the aquatuner cools the input liquid by a fixed amount (14C). Because of this is is best to use a liquid that has the highest thermal capacity available. Water and polluted water both have 4.179, but polluted water has a lower freezing point, so is better for this use. Petroleum only has 1.76 capacity, so the effective "heat deletion" is quite a bit less when cooled by an aquatuner.

[Nightinggale]

Summary for readability: The cooling of the aquatuner is:

heat capacity*14*(mass of liquid)

Since it will always use 1200 W regardless of the cooling, you should prefer to maximize heat capacity of the liquid going through it. Also you want to maximize the mass as in you want a full pipe.

Another thing you want is to have a steady flow through the aquatuner. It doesn't start or stop instantly. It can handle one dot every 200 ms (that's the standard pipe speed). However it takes 200 ms to start and 200 ms to stop. Stopping is just animation, but starting actually consumes power. This means starting an aquatuner takes the same amount of energy as it takes to cool one dot, but you don't get any cooling for it. It's waste. This means sending one dot of liquid through once in a while means the power usage is the same as for cooling 20 kg in a full pipe.

[natanstarke]

Buuuut polluted water boils easily its far safer to use petroleum to really hot areas.

 

[EleriusX]

59 minutes ago, natanstarke said:

Buuuut polluted water boils easily its far safer to use petroleum to really hot areas.

Oh, absolutely, but that hot petroleum should exchange heat with polluted water which can flow through an aquatuner. This can be done with a closed loop and a reservoir, so it doesn't even need power for the petroleum loop.

[Craigjw]

I use slush geysers.  I send the liquid to a large heat exchange made of gold tiles and gold pipes, which has a water input also.  The water comes out at 15c to supply the living quarters.  The warmed PW, is at around 30c when it exits, this I then use to cool coal and petrol generators and a large battery bank, it's then sent to heat up a pincha pepper farm, then it's piped to cool a couple of natural gas vents and finally condenses a steam geysers.

When my gold volcano is active, one of the slush geysers is diverted to feed the cooling of this, which I receive 20c gold from.

[badgamer123]

actually i looking for non-cooling-loop cooling...as i find that filling every part of the base with AT is a bit annoying(for god sake thermos regulator is too bad.) it also reduce all those pipe...

back to topic..petroleum is really not that effective,but the easy to make and large temp range is much better then super coolant.when you get to the point of space....energy is no longer a problem.
I would said it only lose to PH2O 

[KittenIsAGeek]

18 hours ago, Half a fork said:

At some point I'm going to crack a way to attach multiple cooling loops to a single line... I'm almost there with distinct graded loops!

If you use valves to restrict the flow and shutoffs to close them, you can set up restricted flow loops in series off the same line, provided you do not go past your 10kg/s max.  You can even use logic to prioritize a line based on temperature.  Lets say that you have a small loop for cooling your oxygen production and a larger loop for cooling sleetwheet.  If your sleetwheet hits, say, 0, you open up a shutoff that bypasses the valve onto that loop and force a shutoff to the oxygen to close, until the temp drops back down close to where you want it.

[Half a fork]

20 minutes ago, KittenIsAGeek said:

If you use valves to restrict the flow and shutoffs to close them, you can set up restricted flow loops in series off the same line, provided you do not go past your 10kg/s max.  You can even use logic to prioritize a line based on temperature.  Lets say that you have a small loop for cooling your oxygen production and a larger loop for cooling sleetwheet.  If your sleetwheet hits, say, 0, you open up a shutoff that bypasses the valve onto that loop and force a shutoff to the oxygen to close, until the temp drops back down close to where you want it.

Using valves is a good way to go, but I would lose out on cooling power due to less thermal capacity so I want to avoid valves - unless there's something clever possible to bring all cooling lines back up to 10kg/s. Otherwise each cooling loop on the line will cut into that 10kg/s max. 

Using logic to prioritize cooling is the way to go I feel, but maybe using it to deliver pre-cooled coolant to areas that require it, rather than shifting coolant from one area to another. I'm thinking; have a hot-side loop, when it gets too hot then kick it out, activate an element sensor and open a shut off on a cool-side loop where the coolant has been sitting in a chilled reservoir. Different shut-offs go to different areas, and we only open the one going to the area where the hot coolant is leaving. The coolant that is leaving joins onto a staging loop just before the cool-side loop; this staging loop has the same number of reservoirs as there are hot-side cooling loops -1. My problem is that not all the hot coolant will be able to join the staging loop at the same time - either they will have to be prioritised OR, like you suggest, we can use a valve (this way they won't be prioritised but all the hot-side coolant will exit slower which is not necessarily a bad thing).

[Nightinggale]

8 minutes ago, Half a fork said:

Using valves is a good way to go, but I would lose out on cooling power due to less thermal capacity so I want to avoid valves - unless there's something clever possible to bring all cooling lines back up to 10kg/s. Otherwise each cooling loop on the line will cut into that 10kg/s max. 

You can take one pipe and feed 5 valves, each set to 2 kg/s. Each valve feeds directly to a shutoff valve and the outputs are merged into a single pipe. You can then use 5 thermo sensors, one for each shutoff valve and you can control the cooling amount from 0 to 10 kg/s in steps of 2 kg/s. This might be a bit of an overkill to add that many valves, but I think you get the idea on how to do adjustable flow amount cooling. The benefit from doing this is your cooling might toggle between 4 and 6 kg/s, making the cooling capacity have less spikes than if it's 10 or 0 kg/s.

We can argue if restriction valve or shutoff valve should be first. The best option would be to mod a valve to be able to turn off using automation as there will always be either an issue of liquid stuck between the two valves or the first valve drains the input pipe when opening, preventing the rest of the valves from getting anything.

[Half a fork]

3 hours ago, Nightinggale said:

This might be a bit of an overkill to add that many valves, but I think you get the idea on how to do adjustable flow amount cooling. The benefit from doing this is your cooling might toggle between 4 and 6 kg/s, making the cooling capacity have less spikes than if it's 10 or 0 kg/s.

I'm not sure I fully understand this set up, could you provide a picture please?

If I have a constant 10kg/s spinning through the hot side and only kick out those balls that are over my target temperature then there shouldn't be any spikes in cooling because there's never going to be 0kg/s (except during start up, when there's heavy load on the system). I'm acheiving this using liquid reservoirs to store (cold) coolant in-line ready to deliver; as soon as there's space in the loop it spits out 10kg/s of chill. 

This is the solution I came up with (each room is temp controlled between 40c and 10c, but they could be set to all 10c if we wanted to). 

 

My main problem with this solution at the moment is uneven distribution of coolant in the holding tanks on the left. I think that once the temperature in all the rooms have balanced none of them will be spitting out any more coolant than any other; but if they do then that coolant may eventually migrate to other loops (because I'm splitting the cold coolant evenly on the storage side [I've changed the pipes since taking that shot], but unevenly on the hot side). I also don't know a great way of stacking <10kg/s liquid pipes up to the full 10kg/s so the volume of coolant piping through the ice box is either 2500, 5000, 7500 or 10000 - that's not too important though. 

[Nightinggale]

4 minutes ago, Half a fork said:

My main problem with this solution at the moment is uneven distribution of coolant in the holding tanks on the left.

High Flow Storage can solve that. With 3 inputs and 3 outputs you can set up the 4 tanks as one chain. As a result the final tank should always have enough to output 10 kg/s on each of the 3 outputs and you no longer have a balance issue.

[Half a fork]

44 minutes ago, Nightinggale said:

High Flow Storage can solve that. With 3 inputs and 3 outputs you can set up the 4 tanks as one chain. As a result the final tank should always have enough to output 10 kg/s on each of the 3 outputs and you no longer have a balance issue.

Thanks, that really does simply solve the problem. I'm going to tinker a little without mods for a while though see if it can't be solved within the game already.