I have about 12 bottles of water. If I want to minimize the energy consumption of my refrigerator, should I put these bottles in the fridge or leave them out?

  • 4
    $\begingroup$ Short comment here - I started a longer answer but bed calls - I may add it sometime. Placing water in bottles in the fridge will probably slightly reduce power consumption, but not by a large percentage of normal. Energy loss rates are proportional to temperature differential and not to amount of contents. But adding say 12 litres of water will reduce the volume of cold air in the fridge. When the door is opened less cold air will be lost and less warm air will enter and need to be cooled when the door is shut. A very rough estimate is that 360 Joule less cooling would be needed . $\endgroup$ Commented Apr 6, 2016 at 13:51
  • 1
    $\begingroup$ Having more thermal mass inside the fridge might have a slight effect on the cool/idle cycle for the compressor. (With the bottles at thermal equilibrium, the compressor probably stays on longer when it does come on, but stays off longer between cycles because the thermal mass of the water means less change in fridge interior temp for the same energy loss through the insulation.) IDK if that makes any difference to efficiency, though. $\endgroup$ Commented Apr 7, 2016 at 6:08

5 Answers 5


The fridge works by pumping heat from its interior to its exterior. The amount of work it needs to do is proportional to the amount of heat entering the interior (and also the temperature difference between interior and exterior).

There are two ways that heat gets into the interior:

  • Putting warm objects into the fridge.
  • Heat leaking through the walls of the fridge.

When you put a warm bottle of water into the fridge, there is a transient load associated with cooling that water down. However, once the water is cool, it has no further effect on the work that the fridge must do.

Heat leaks into the fridge at a rate that is proportional to the temperature difference between outside and inside. This leakage is constant, and is not affected in any way by any objects that may be inside the fridge.

Therefore, if the fridge is empty, you should just turn it off. If you want the water cool, you should put it inside the fridge and leave it on. However, you should not turn the fridge off and on frequently — there's a huge transient load associated with cooling down the interior walls, etc. of the fridge itself.

Also, every time you open the door, some of the cool air inside is replaced with warm air from outside. This also represents a small transient load, but if you're opening the door a lot, it can add up.

  • 1
    $\begingroup$ This is a great answer. There are many approaches to answering this type of question but understanding how to analyze the system, and the flows of energy involved, is key to being able to apply the trivial problem of water bottles and refrigerators to less trivial problems, independent of the specific material properties of air and water. $\endgroup$
    – Air
    Commented Apr 7, 2016 at 0:17
  • $\begingroup$ Following on from your last part about the air change when you open the door, another answer notes that having more space filled with water bottles would reduce the amount of air in the fridge that can be changed when the door opens, so if you're going to be opening the door a lot, having the bottles in there would reduce the amount of warm air that can come in each time, and help keep it cold. Water has huge heat capacity, though, so it may be that by adding even room-temperature water, you're adding more heat than the little bits of air would. $\endgroup$ Commented Apr 7, 2016 at 8:57
  • $\begingroup$ What about heat radiation? $\endgroup$
    – ASA
    Commented Apr 7, 2016 at 13:02

If you have unused space in the fridge, then it's better to fill it with something than to leave it air. This is because every time you open the door, less air will be exchanged with the room, so less to cool again.

If you're not going to open the fridge, then it doesn't matter. As others have already pointed out, putting the water bottles in the fridge represents a one time energy cost. Once they are cold, they cost no further energy, and reduce the volume of cold air lost when the door is opened.

However, if you are really after less energy usage and you have excess space in the fridge, the best long term answer is to get a smaller fridge. Get one that is just large enough to hold what you need to keep cold. In smaller fridges, the cold box has a smaller surface area, which is where heat comes in when everything is at steady temperature and the door isn't opened. All else being equal, the fridge with the smallest cold box surface area will require the least power.

Of course all else isn't equal. There are some cost/performance/longevity tradeoffs in the cooling system, but mostly it's about the insulation. That is also a cost tradeoff, so different models have different power requirements for the same cold box. At least here in the US, sellers are required to list power consumption figures for refrigerators. Look for the one that requires the least power of those that have the minimum size cold box you need.

Other things you can do to minimize power consumption:

  1. Make sure the radiator has good ventillation. The cooling system is a heat pump that moves heat from the cold box to the outside of the fridge. The radiator is where it gets rid of that heat. If airflow to the radiator is blocked, then the heat pump has to work harder (make the radiator hotter) for the same cooling. This costs power.

    The radiator is usually a metal "mesh" sort of thing on the back of the fridge. Modern fridges are designed so that there is some minimum space for air flow with the fridge backed up all the way against at wall. Moving it out from the wall a little and making sure that air can flow in at the bottom and come out easily at top can help.

  2. Keep the radiator clean. Dust is a good insulator, which will make the cooling system work harder to push the exhaust heat thru the dust layer.

  3. Put the fridge in a colder place if you can. The colder the radiator, the less hard the cooling system has to work.

  4. Minimize the time the door is open. If you have several things to move in and out, open the door once, move everything in quick succession, then close the door.

  • 3
    $\begingroup$ I've never understood why people go to such lengths to hide their fridges away in closets and cabinets. It just impedes airflow around the coils, collects dust and reduces magnet real estate. $\endgroup$
    – Air
    Commented Apr 7, 2016 at 0:24

There are two aspects to this.

First if your fridge is filled with anything other than air then you will save the energy associated with cold air escaping whenever you open the door.

However if you put bottles of water in your fridge it will take a lot more work by the fridge to cool them down than the equivalent volume of air as a litre of water at room temperature contains a lot more heat than a litre of air at the same temperature so more work is required from the fridge to cool a litre of water than a litre of air.

If you intend to put the bottles in the fridge at some point it is certainly better to fill the fridge now rather than put them in individually. But if you don't actually need the water to be cool ever it would be better to fill the spare space with empty bottle or blocks of styrofoam.

This is because of the relative specific heat capacities of air and water. The specific heat capacities of air and water by mass are 1 and 4 KJ/Kg·K, respectively, but water is around 1,000 times denser than air so the heat capacities by volume are different by a factor of 4,000.

So it is obvious that the energy wasted by loss of air when you open the fridge is fairly trivial compared to the work needed to cool the same volume of water.

Typically a fridge will use around 1 unit of mechanical power for every 3 units of cooling power.

So overall there is a small advantage in minimising the volume of free air in a fridge but it is only worth doing this by filling it with things that you want to refrigerate anyway or solids (or containers or foams) which have a low volumetric heat capacity.

This only applies to losses from opening the door though, once the contents are at a stable temperature it makes no difference to energy consumption what is actually in it, assuming that is is reasonably well-sealed.



Many factors play in this scenario:
1- the shape of the bottles and their material thermal insulation index.
2-how you place the bottles on the shelves: do they promote air current or impede it.
3- the frequency of opening the fridge door.
As many have mentioned there is an initial load when room temperature bottles cool of. but after that point everything else being equal it boils down two cases; penetration of heat into fridge due to design and manufacturing exigencies and material thermal properties of fridge without bottles versus contribution of bottles acting like the vanes of a cold radiator letting the stream of transient air wash around it. So if the shape and placement of bottles doesn't encourage entrant air circulate easily it would help saving energy a bit.
ultimately if you fill up the entire available space with water past the initial pick of energy the chronic use of energy will be smaller.
we could look at this fridge as revers of a cooling tower. if we don't want to interact thermally with ambient air we make it hard by placing the bottles of water in a way that shields the other parts of the fridge from convection or ventilation. this concept has been used for centuries in middle eastern architecture, eg, passive air-conditioning.


The added weight of the water bottles will cause the contact patch between the feet of the refrigerator and the floor to deform ever so slightly, thus making better thermal contact between the feet and the (presumably warmer) floor. Hence, the addition of the water bottles will increase the heat loss through the feet by an amount so small as to be undetectable by even the most sensitive heat flux sensors.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.