Is it more economical to leave the air-con on?

What is more economical - leaving the air-con switched on all the time (keeping the room cool) OR switching it off when not required (when you leave the house, say to work or the shops)?

See if you can crack this case and enter our MythDefied competition. Leave an answer with your reasoning in the comments below.


No. Unless your absence is only for a short period (minutes) then turn it off.

If your A/C (reverse cycle or swampy) can’t cool the house, or room in the case of a split, in a few minutes its inefficiency is probably killing your power bill anyway making it a moot point. Typically A/C is at its best when running flat-out rather than ‘cycling’ - and leaving it on all day doing that will also wear it out quicker.


The only time it could save a very small amount of energy is if the airconditioning unit is turned off for a short time, and this time is within the normal cycling time of the compressor running (when the thermostat cuts off and then on again). The amount of energy being saved would be very small and would correspond to the energy used during compressor standby and potentially from the fan running. To save energy at such times, one would need to closely monitor the airconditioning cycling, which in reality is impracticable and unlikely.

Otherwise no. If an airconditioner turn on from the thermostat being triggered, it means that heating/cooling is being lost from the air within the aiconditioned area. The longer the duration the airconditoner is off the more one will save as the amount of energy to maintain a particular ambient temperature will be significantly greater than turning off and then airconditioning the room back to the same ambient temperature.

Leaving the airconditoner on to ‘save money’ does not make any financial sense.


I can’t bust this myth, but am very interested in the answer!

I recently had my aircon fixed, and the repairman recommended I do exactly this (leave all units in the house on, 24 hours a day) to reduce my electricity bill! I haven’t been game to follow his advice, though.


As said above, generally no.
Many modern ACs don’t cycle on and off, they use inverter technology to run at a lower power setting, providing the right amount of cooling to keep pace with heat gain from outside.

However, if you happen to have a PV system, it can be worth cooling the house with essentially free (marginal cost anyway) energy from your roof in the afternoon when no one is home, so it is already cooled when you arrive home, rather than using grid (ie mostly filthy fossil fuel generated) power, which will cost more, especially if paying peak late afternoon rates for it.
You will miss out on some feed-in tariff (not everyone gets paid though!), but since the FiT paid for solar power is less than the cost of power in most cases, you are better off using your own generation when it is available.


A fringe comment, but when the A/C is turned off during the day the inside heat load will build up in many dwellings and much less so in others. The furniture, walls, and floors heat up.

If the A/C is left off, when it is turned on it is individual whether it is comfortable in a cool room on a warm lounge or in a warm bed. For me leaving the A/C on to keep it all cool provides a better comfort level.

The ‘cost’ to cool that heat load each time could be where @SarahAgar’s repairman was coming from.


Still doesn’t make sense…an airconditioner has to ‘fight’ heat transferred from outside the airconditioned room. The amount of energy absorbed by the room will be the same irrespective of whether the room is airconditioned or not (in fact, the amount of energy absorbed and retained within a house will be less than that being fought by the airconditioner in longer periods…otherwise a room will continue to get hotter and hotter and never cool until the outside temperature decreases/night time…in reality, a room will heat up until such time the amount of heat being radiated will equal the amount being absorbed). The factors which will affect heat absorption are things like roof and building colour, insulation, construction materials, curtains etc.

If the room heats up, it is still the same heat that the house would otherwise have absorbed and the airconditioner cools. There becomes a point where the temperature within the dwelling won’t change much through heat being absorbed and retained (the heat equilibrium outlined above)…depending on the type and insulation in the house, this could be less to an hour to several hours. Anytime after this there will be considerable energy savings. Before this point, any saving is likely to be minuscule and due to air leakage within most (all?) Australian homes, the cost to run the airconditioner will be greater as cool/hot air will be continually lost and replaced with external air, in effect, increasing the amount of work the airconditioner would otherwise have to do…

The other thing to consider is many businesses/office complexes will have night airconditioning cycles where the airconditoner is either shutdown or runs at a more efficient setting (either at higher/lower temperatures or at lower intensity). I have worked in a large office building where, outside normal business hours, the airconditioning shut off completely and one had to contact security should one need airconditioning at such times (e.g. running meetings with external clients on a weekend). If it was cheaper to run the aircon all the time, business would not be doing the above.

The other thing to consider is coolness/warmth comes from air temperature and radiated sources. A chair/bed or other furniture is unlikely to have a high embodied energy and radiate out much heat to make one feel warm.

Also, the higher the differential temperature between two bodies, the quicker the energy loss. If one runs an airconditioner, it is highly likely that the airconditioned room temperature would be significantly greater than a non-airconditioned room. This greater temperature differential would increase running costs and would be better to turn off the aircon when not in use. The only way this factor would not come into play is if the insulation was near perfect, and there were next to zero losses from a airconditioned room. In reality, this is impossible as there will always be come loss unless one lives deep under the ground or in a vacuum.


‘cost’ was parenthetical for a reason. Regardless the thermodynamic arguments do not reflect an individual’s comfort level in the two scenarios, dwelling dependent.

I challenge that assertion although it is furniture dependent. A hardwood chair will not have the same heat transfer as a king gel mattress for example. Not sure if you mean sitting on it by radiation, or you mean radiation into the rooms; I meant sitting on it. If you want to talk about it, I will send my partner to visit you. After a week dealing with her in hot weather I truly think you will yield the point, even if you don’t mean it :smiley:


This illustrates the question being very broad - along with factors like usage profile of the building, physical dimensions, construction material, building contents, insulation, and externals such as climate, ‘air conditioning’ itself is a very broad term that embraces a number of possible techniques to achieve the end result - evaporation, refrigeration, heat-exchange, etc - all these things have an effect on a good outcome. Air leakage is a very good thing for evaporative, indeed mandatory - while it is not such a great thing for refrigerative.


It sounds like someone is trying to keep themself in a job by increasing the maintenance on your unit!


I am not sure of the answer, but I am thinking fridges. (For this I am ignoring any effects on the food inside.)

If you turned the fridge off, it would maintain the inside temp for a bit and slowly the temp would rise as the room temperature leaks in. If the power is turned back on after a significant temperature increase, the motor would have to work long and hard to reduce the temp back to the presets.

If the fridge is left on, the motor just has short bursts to maintain the temp at a somewhat constant level.

The question is does the motor expend more energy to decrease the temp after being off, and conserving power, or does it use more for those short bursts to maintain the temp?

OK, now back to the air conditioner. I think the answer is it depends on how much temperature difference there is between the ambient temp and temp of the cooled area, and on how long the power would be off.

The greater the difference in temperature the more likely I would be to leave the power on. Equally, the longer I am away the less likely I would be to leave the power on.


It’s a common misconception that fridge motors/compressors have to work harder to bring the temperature down, but this is not the case for a regular fridge. The compressor motor only runs at a set (it’s maximum and only) power level - what varies is the time it runs, and the thermostat is what governs how long it runs for.

Inverter ACs can run at lower power levels as I mentioned earlier in this thread, but I’m not sure if that tech has appeared in domestic fridges yet.

In general, if the average temperature in the fridge, or house, over a period is higher, then less energy will have been used, cf maintaining average colder temps in the fridge or house for that period.


The other thing is the amount of ‘cooled’ energy in the air is not that much. Most of the cooling energy in the fridge is that in the produce and the interior of the fridge itself.


So this is the perfect spot for debunking yet another myth. Thank you for correcting my MYTH-take. :blush:


The first part of the problem with getting an answer to this question is that it is too vague. How long is long enough to turn off the air when leaving the room? Are we examining just the cost of running under different regimes or do we bring in questions of amenity, such as having to suffer in the heat while the room’s temperature is brought down, of the possibility that the house will grow fungus or your wooden artefacts will crack while the air is off?

A second more serious problem is that both the yes case and the no case are attempting to make a quantitative argument without producing any numbers. We have all sorts of pronouncements by supposed experts but no sums. Some claims are just silly. For example; leave it on and save money by turning up the thermostat 7-10 degrees F (presumably in cooling mode) because when you lower the difference between inside and outside temperatures you save energy. This saving is real in comparison to having the temperature set to 68F (20C) but not compared to turning the machine off altogether, which was the proposition.

There are many claims about the efficiencies and inefficiencies of refrigerated air-conditioners but no modelling or statement of assumptions that might make the position clear. Are we talking about machines with inverter tech? Does this matter? Unless we can find a clear well-run study that makes these uncertainties explicit it will just be gut feelings hurled at 20 paces. I can’t find such a study, that is not to say it doesn’t exist just that I have limited resources to look.

My gut says to save the most turn it off if you leave the room for more than 21.32 minutes. That could be absolutely right - or just borborygmus.


My new word of the day :slight_smile: Ta,


Interesting. I had only ever heard the plural, ‘borborygmi’ used.

I suppose it is possible to have just one rumble.


The answer is it varies, there is no clear cut process depending on your needs. But if it is cool enough or warm enough outside then yes turning off is more economical.

Several sites have guides to using Refrigerative Air Cons:

In a few (one an air con business) it does mention only cooling/warming the areas you need to and shutting off using the timer function or when it is warm/cool enough outside (depending on the season).


A modern inverter reverse cycle a/c has an occupancy sensor that will adjust room temp to a higher setting when the room is empty. If turned off during the day, any heat build up has to be removed. The unit will have to work harder to bring the temp down to a set temp. Measure kwatts used for both scenarios to find the best option.


There has obviously been a lot of discussion about this issue, but I think it is time to set aside all of those differing views and state, once and for all, the definitive answer to Brendan’s question:

It depends.

Thank you. Any further correspondence regarding this question and answer may be directed to @BrendanMays.