LNG On-Demand Energy Storage

[Xentik]

Hello all, I wanted to show off and get some feedback on my design for automated energy storage via LNG (liquefied natural gas).

It occurred to me while playing that the energy storage density of LNG is orders of magnitude higher than the current energy storage mechanism, batteries. Specifically, current batteries in game store 10kJ/tile, whereas LNG stores 13MJ (0.8kJ/s  * 1000kg / 0.060 kg/s). With that in mind, I was interested in seeing if it would be possible to produce a system that would automatically convert surplus natural gas into LNG, store it, and then convert LNG back to NG when extra power was needed. Without further ado, here’s the prototype system:

The system is broken down into four major subsystems, which I will describe below. Additional images are attached showing the various

1) NG Input Chamber – This subsystem stores NG and pre-chills it for conversion to LNG or direct conversion to power.  It acts as the central control unit for the entire LNG power system, directing excess NG to the liquefier and requesting LNG from storage as needed.  The system consists of three gas input pipes and one liquid input pipe (upper right hand corner), four atmo-switches (two for gas output to power, one for gas output to liquefier, and one for liquid input from LNG storage), and three gas pumps which are controlled by the aforementioned atmo-switches. In this prototype, the lower left gas pump/atmo-switch is connected to the on-demand power system, and the upper left  is unused.

In theory the lower left pump should be set to run as long as gas pressure is > 500g (to ensure efficient gas intake), but is set off due to oddities with the NG generators not ever shutting off even when batteries are full. The lower right gas pump is set to run when gas pressure > 1750g to divert surplus gas to the Liquifier. The upper right atmo-switch controls the liquid input and activates at pressure < 500g, causing LNG to be pumped in. The chamber itself is pre-cooled by wheezeworts to  ~-60 C to minimize the work done by the liquefier. As you can see in the picture, the wheezeworts here are stifled, but it is important to have excess cooling capacity for when gas is streaming in constantly during normal operation.

2) NG Liquefier – This is a simple hydrogen-based gas liquefier running off a single thermoregulator and a single pump.  It uses granite walls inside along with granite pipes to provide thermal mass, and abyssalite for all other components. A vacuum chamber is used for maintenance access, which is functionally equivalent to the surrounding abyssalite insulating tiles (Note: Vacuum created with debug because I was lazy at that point. It is otherwise a standard vacuum interlock design I use to keep cryogenic biomes and buildings insulated but accessible). The hydro-switch is set to 14kg to ensure efficient pumping.  A thermoswitch at the bottom functions as a kill-switch for hydrogen coolant temperature, preventing the system from liquefying the hydrogen coolant. This subsystem is in need of serious optimization, and is responsible for most of the power consumption in the system, since the thermoregulator runs nonstop.

 LNG Storage – This system is simply a cryogenic storage tank for LNG, in this case designed with a capacity of 25t (25000kg) of LNG for a maximum energy storage of 0.333 GJ.  Granite walls are used for thermal mass, abyssalite for insulation, and the hydrogen cooling pipe pre-cools the granite walls down to ~-170 C before being shut off (once prepared, it will maintain its temperature indefinitely). The outer wall is to ensure structural stability when the tank itself is at max capacity. A hydro-switch is present with a setting of just above zero to prevent the liquid pump from wasting energy when the tank is empty and the NG Input Chamber is requesting liquid delivery. A gas pump is placed within the chamber to vacuum it out before liquid delivery.  Finally, a flow regulator is placed on the output port, set to 4kg (2x the max consumption of the input system) guaranteeing that the LNG system will deliver sufficient liquid to the Input Chamber without bringing it over 1750g/tile pressure and guaranteeing it will shut off (2kg max consumption in the chamber for power gen).  

It is essential to minimize the distance between the liquid pump and the input storage, as the longer the distance, the more LNG will be pumped out in response to a shortage in the main chamber. If the distance is large, the lag may result in over-pressurizing the Input Storage, thus converting some of the LNG to gas and then feeding it back into the liquefier, which is a waste of energy.

Power Generation – The power generation system here is the emergency system for my base. A max output of 4.8KW in conjunction with a 0.64MJ battery buffer.  The NG generators are set to only turn on when batteries are < 10%. Unfortunately the generators seem to run non-stop if fed NG, regardless of whether the batteries are full or not, so currently they are not fed gas unless I see the base needs extra power specifically. In theory once the bug is fixed, however, the auxiliary generation should be 100% automated.

That’s the system, now lets talk about efficiency and how well the prototype works. Currently the system has 7.07t of LNG stored over ~200 cycles of operation. Given some downtime and testing of the automated withdrawls, the system has stored an average of 47kg/Cycle of LNG, with the rate varying between 33-60kg/cycle. Clearly we can see from this that the liquefier is the limiting factor here.  Currently the liquefier converts around 0.078 kg/s out of a theoretical input maximum of 1kg/s.  This limits the ability of the system to soak a NG surplus right now, and also determines the efficiency of the system.

In terms of efficiency, we are looking at a “wall-plug efficiency” of 71%, i.e. if we use 1kg of LNG for power generation, we paid roughly 29% of the energy of that LNG to get it to the generator. The majority of this cost comes from running a thermoregulator nonstop (estimated at ~3.0 kJ/kg = 240 kJ/s *600 s/cycle / 47kg/cycle). The gas pumps contribute an additional cost of 0.24kJ/kg each, and the remaining components are negligible (e.g. the liquid pump for cooling the thermoregulator flows 25g and runs 1.5 ticks a day = ~8e-3 kJ/kg). Efficiency can be improved by either somehow lowering the duty cycle of the thermoregulator (50% duty cycle, i.e. running every other tick, improves efficiency to 82%) or increasing the rate of LNG creation by the system (doubling to 94k/cycle would yield 82% as well). By way of comparison, the wall-plug efficiency of non-liquefied NG in my system is 96.4%, due to the use of two pumps (geyser->holding, holding->generator) and no filters.

I am interested in ideas to improve the system overall, as I believe LNG will be essential for advanced power systems even in the presence of higher-quality power sources that may be introduced in the future (e.g. nuclear or fusion). The reason for this is that NG can be produced using surplus power with a fertilizer/air scrubber combo, allowing excess power to be converted into LNG for long-term storage. Additionally, LNG pipelines can deliver 133 kW of power across long distances, roughly 6.7 times the rate of heavi-watt wires.

TLDR; Liquid Natural Gas can be produced for long term energy storage at 71% efficiency. With atmo-switches, diversion of excess NG into LNG creation and gassification of stored LNG can be automated, creating an on-demand energy storage/extraction system. The 5x5 tank used in the prototype has an energy capacity of 0.333GJ, an energy density ~1000x greater than the equivalent space worth of batteries.

 

[rezecib]

This is amazing. I don't think I'd build it, but it's still amazing.

Perhaps an optimization you could make for the cooling system is to throw an atmo switch in the cooling chamber, and have that control the thermoregulator in addition to the thermo switch. Although perhaps this would be better achieved by just setting a higher threshold for the thermo switch.

[absimiliard]

Wow!  I'm afraid I'm far too much of a barbarian to offer any improvements.  But I really wanted to say that this is some very impressive engineering.  Alongside the various scientific testing regimes some of the players do this sort of thing is what draws me to ONI.

 

Just ... damn fine work.  Damn fine.

[LeadfootSlim]

I can genuinely say I'm flabbergasted. How and where do you convert the liquid natural gas to a gas again for processing?

[Kasuha]

Okay, I think the machine is not very useful with relatively steady natural gas sources in ONI but it certainly is impressive.

As an optimization, I would probably recommend simple one tile thick walls made of abyssalite for both the liquefier and the storage. Even if some of the gas evaporates when you send it to the storage, the bulk of the following gas will cool it and liquefy again as long as it is not too warm. It should be slightly more than 5 C below its boiling point to be safe. I have never found a good use for "thermal mass" in freezers or cold storages, they only increase the amount of time needed to get the thing running.

Another "optimization" could be to keep the gas in gas form in the storage. Just add a space heater and run it if the chamber becomes too cold or if you detect liquid on the floor. The walls will certainly sustain the pressure in current release and you don't have to have a separate evaporation chamber.

 

[The Flying Fox]

Yeah, very impressive!  Not terribly practical at the moment, but who cares?  It's fun to do stuff like this, particularly since it mirrors real-life, in a way.

50 minutes ago, Kasuha said:

Another "optimization" could be to keep the gas in gas form in the storage. Just add a space heater and run it if the chamber becomes too cold or if you detect liquid on the floor. The walls will certainly sustain the pressure in current release and you don't have to have a separate evaporation chamber.

You could go one step further.  Why even have an insulated chamber?  Just make it out of granite, let the liquid naturally evaporate (Cooling your base!)  Then, use the gas pump you left in the chamber to make a vacuum and re-purpose it to pump out the gas.  As Kasuha said, the game doesn't destroy blocks due to over-pressured gases.  Just too much liquid and, its possible this might not change.  Unless they decide to fix polluted water producing polluted oxygen with no limit.

[rezecib]

The reason for leaving it in liquid form is so that you can have a LNG line transporting the energy, that's how you get the extraordinarily high energy transfer rate at long-distance.

But I'm not sure that's really that practical, considering that you have to warm it up again. Although I suppose it doubles as a cooling agent?

[The Flying Fox]

12 minutes ago, rezecib said:

The reason for leaving it in liquid form is so that you can have a LNG line transporting the energy, that's how you get the extraordinarily high energy transfer rate at long-distance.

But I'm not sure that's really that practical, considering that you have to warm it up again. Although I suppose it doubles as a cooling agent?

Well, since a single gas pump can feed up to 8 natural gas generators, it's just as practical to have a long gas pipe for long distance transport.  Unless you need to run more then 8 backup generators.  But yeah, you could use it as a cooling agent.  Maybe use it to lliquidize/freeze CO2 and chlorine so you can get rid of it?