Science Questions

What is the most efficient way to heat water?

Sun, 3rd Apr 2011

Listen Now    Download as mp3 from the show Keeping the Conversation Flowing


Francis Tapon asked:

I'm in Croatia and I'm having a debate with my friend about water heaters.


I have a water heater in my bathroom. I take on shower per day. I set it at just the right level so that it gives me just enough hot water for a 5-minute shower, no more. I turn on the boiler about 1-hour before I want to take a shower and then turn it off right when I get into the shower. I don't turn it on until the next day, where I follow the same ritual.


My friend says I'm being inefficient and not saving any energy. He says I should leave the boiler on 24-hours a day, because it takes a minimal amount of energy to keep the water hot once it's hot.


If I turn it off, the water cools and then must be re-heated from scratch, requiring far more energy than if it had been on all along. Who's right?



We put this to Hugh Hunt, from Cambridge University's Engineering Department...

Hugh -   This is a very interesting and topical question.  If the hot water is stored in a perfectly insulated tank, then it makes no difference whether the water is heated one hour before it’s needed or one day before, or even one year before.   There's nothing at all to be gained from heating the water up 22 hours in advance, but that requires that the insulation is perfect.

In reality, insulation is not perfect and the water heater will lose some heat in advance and this heat cannot be recovered.  It’s the same as when we’re boiling water for a cup of tea.  If we boil the water when we need it, that makes sense because the kettle is not that well-insulated.  If you boil the water an hour in advance, the water would be cold when we needed it.

How much energy wastage are we talking about?  Suppose a 5-minute shower uses 60 litres of water, which is about right; and suppose the water is delivered at about 40 degrees, which is about right; and in the winter the water is cold, somewhere around 0, which is about right.  So we’re talking about 4 kilojoules/degree/litre of water.

Now suppose the tank loses 10% of its heat in 24 hours.  Well then that's a loss of getting up to 1 Mega Joule of heat.  That's equivalent to running an electric kettle for 6 minutes, enough to boil the water for several cups of tea.  So if you're the kind of person who’s bothered not to over-fill the kettle when making a cup of tea then you should certainly follow our Croatian friend’s example and only heat the shower water when it’s needed.


Subscribe Free

Related Content


Make a comment

It depends on how well insulated the tank that holds the hot water is. If it's really well insulated, you won't waste energy whether you leave it on all the time, or turn it off and on, so you might as well leave it on all the time if you find that more convenient.

If the tank is not well insulated, you are going to waste energy whether you leave it on all the time, or turn it off and on, but you'll probably waste a bit less if you only turn it on just before you use it.

We have a well insulated electrically heated tank that is always on. I know it is not wasting energy because it is in a cupboard under the stairs. If the tank was losing heat, the air temperature in the cupboard would be appreciably warmer than the air outside the cupboard, but it isn't. Also, on the few occasions when I have disabled the heating elements, it has taken a very long time for the water in the tank to cool down.

That said, probably the most efficient water heating systems are the type that don't store hot water at all. They only heat the water as it is being used, but I doubt they are all that much more efficient than stored hot water systems that have well insulated tanks. Geezer, Sun, 27th Mar 2011

Ironically, in some countries, they call the immersion heater the "geyser", Geezer.

Chris chris, Sun, 27th Mar 2011

standby heatloss is dependent on the temp difference between the hot tank & the environment around it. I do the same= drain/use the hot water storage & leave the tank cold so it cant lose heat while standing by= cold tank has no heat to lose. CZARCAR, Sun, 27th Mar 2011

Yes - my granny had a gas geyser, or maybe that was my grandfather? Geezer, Sun, 27th Mar 2011

BTW, and drifting off topic a bit, there are some really cool (well, hot really) heat pump water heaters. They extract heat from the ambient air in the home and use it to heat the water. This works really well in warm climates because the water heater is actually helping to cool the home.

Even in a cold climate, these systems heat the water very economically. The only problem is that they are extremely expensive, although I suspect the prices will come down a lot if there is sufficient competition. I don't think there is anything fundamentally expensive about the technology. It's a lot like a refrigerator in reverse. Geezer, Sun, 27th Mar 2011

The energy to heat the water to a temp is the same whether you do it immediately prior to shower or several hours prior. But it's the thermal isulation which is critical to the overall efficiency of the system. If the system has 100% effective insulation so the water stays at temp then there is no advantage to one system over the other. But if the water in the tank is able to cool before being used then more energy will be used by using central heating to keep it constantly upto temp.
If you use your bathroom heater to only heat enough water for your shower, then that should be more efficient than having it on all the time except the problem is that as you use the water from the water heater tank it refills itself with cold water which cools the already hot water which is inefficient.
So the most efficient way to heat water would be a system which only heats water on demand without storing it and in fact these systems are becoming increasingly popular for that very reason. Called Combi heaters which heat the house but also heat hot water on demand.
But the absolute most energy efficient way to shower is to use cold water. Hot water is for wimps.
percepts, Mon, 28th Mar 2011

Yes, well good luck with that! I don't know where you are, but the water here comes out the tap at a blistering 46°F. Geezer, Mon, 28th Mar 2011

@NakedScientists It depends entirely on the size of your boiler. Problem sorted :) was tweeted by @geo2ward @geo2ward, Mon, 28th Mar 2011

Yes, well good luck with that! I don't know where you are, but the water here comes out the tap at a blistering 46°F.

I lived in a town where the cold water was rather warmer than that. I was having a shower one day when my neighbour walked in and asked if they had fixed the boiler. I had to tell him I was only using cold water, as it was near to boiling just from solar heating in 20m of pipe on the roof. SeanB, Mon, 28th Mar 2011

The answer is turn the water off when not being used.

The rate of cooling i.e. the energy lost is proportional to the excess temperature.  That is the temperature of the water heater above the room.  This rate of loss is always highest if you maintain the temperature of the water at its highest i.e. that which you use it for your shower.  The fact you need to reheat the water from the lower temperature is not the dominant factor here because by keeping the heater on you are constantly doing this anyway as the thermostat cycles on and off.

The only way to lower your energy input is to insulate the boiler as best you can and try and retain as much heat as possible when you next switch it on. PeterBroad, Thu, 31st Mar 2011

@NakedScientists Is the answer the most common one in science, 'It depends...'? was tweeted by @bangscience @bangscience, Thu, 31st Mar 2011

When you've had a shower, the water taken out of the tank is replaced by an equal quantity of cold water - which needs to be heated up - just as in the case of only heating it when you need it.  Factor this in, and only heating it when you need it is far and away the most efficient thing to do. Gwernogyss, Wed, 20th Apr 2011

Actually, he's being slightly inefficient in one way. I *think* he should turn the water heater off right before he has the shower.

That way it will avoid the heater coming on during the shower- which will probably largely waste heat, you would be mostly heating up water that you're not going to use in the shower. Also he should not turn it on too long before hand- the optimum temperature for the water is hot enough to just last for one shower.

But there is one catch worth mentioning- if the water doesn't get hot enough, then you're at risk from Legionaires disease; hot water systems run at the uncomfortably hot temperature they do to kill off the bugs...
wolfekeeper, Wed, 20th Apr 2011

Of course, there is another factor that should be taken into consideration here. The water in the tank is not all at the same temperature. It actually stratifies so that the hottest water is at the top of the tank, and the feed to the shower is taken from the top of the tank.

So, if you have a crappy tank with lousy insulation, what you want to do is heat precisely the amount of water that you need for the shower then turn off the heater and give it just enough time for all the hot water to collect at the top of the tank. If you do this correctly, the shower will suddenly cool down just as you are about to get out.

Of course, if you invest in a tank with decent insulation, you can forget about all this nonsense, because you will waste a lot less energy by just leaving the heating on and allowing the thermostats to do what they are good at. Geezer, Wed, 20th Apr 2011

bingo CZARCAR, Fri, 22nd Apr 2011

Contrary to popular belief, electricity does not leak from power outlets when there isn't a plug in the socket. Remarkably, neither does much heat escape from a well insulated water tank. If the tank is well insulated, the financial penalty for heating the water before it's actually needed is negligible.

On the other hand, the financial penalty for using a tank that is not well insulated, regardless of any scheme to manually control the applied power, is significant. Geezer, Fri, 22nd Apr 2011

wolfekeeper, I do turn off the water heater right before I get in the shower. You're right to make the point. Naked Scientists, Hugh Hunt's answer is logical, but there's one assumption that is faulty. It assume that the heat loss is 10% per day. It's more like 50% per day with my water heater. I know this because if I heat up the standard level that I need for a 5-minute shower, then in two days, the water is cold if I don't use it. A 10% loss implies that roughly after 10 days, the water is cold/room-temperature. It reaches this state in just 2 days. Thus, the energy is not running the electric kettle for 6 minutes, but rather for 30 minutes. That's a bit more expensive and significant than originally assumed. Thank you for proving me right! Francis Tapon Francis Tapon, Fri, 6th May 2011

You should also consider what effect intermittent aperation of the water heater has on the longevity of the appliance. Thermal cycling through large temperature differences might possibly deteriorate the tank more quickly. Then you have to replace the appliance sooner. May not be a good bargain. Atomic-S, Mon, 9th May 2011

Good grief! Are you heating a cauldron on a wood stove? Geezer, Mon, 9th May 2011

This is all good discussion and the theory is correct but one thing is missing. In many countries where hot water is heated with electricity the electrical power to the heater is controlled remotely by the power company. There are two tariffs for the cost of power – one or power consumed during the ‘peak’ hours and one for ‘off peak’. The reasons for this are another topic. Peak is usually a few hours in the morning, lunch time and the evening. So . . . if the hot water heater is switched on at 2 am it will heat up but if it is switched on at 6am it will not heat up until the morning peak is over (albeit some installations have an override switch BUT the override uses electricity at the peak tariff). So . . . to save energy 1) take a shorter shower and 2) use a solar assisted or heat pump hot water supply or a well insulated hot water heater with the smallest capacity your household can manage with off peak demand heating. Clearly option 1 is the only option for the person with the original question. jo nurd, Sun, 29th May 2011

If my water came out of the hot tap at 46°F I would regard that as perishing cold not blistering hot. syhprum, Mon, 13th Jun 2011

I wonder if that was intended as Celsius.

That is about the temperature of the Metolius River. 
I will tell you that it is CCCCOOOOLLLLDDDD

I've been boating in it.  Get splashed with it and it bites.
I did tip over once, but I didn't stay in the water for very long. 

I believe well water around here comes out in the mid-50's. CliffordK, Mon, 13th Jun 2011

Sheesh! You scientists are soooooo literal. My comment was an example of irony.

Wiki - Irony "Ironic statements (verbal irony) typically imply a meaning in opposition to their literal meaning." Geezer, Mon, 13th Jun 2011

Most responses here have missed the point. The point is NOT whether heating in advance is economical, but whether switching on and off is economical. Lets assume there is some heat loss over a period of time. That would mean at some point the water would need to be reheated. Let's say I lived in Africa, where electricity and water were scarce and I only showered once a month. Surely it would be less expensive to turn on the geezer just before I had a shower, and as soon as the water was hot, I'd use it. Then leave it switched off for the rest of the month. Whereas keeping it on for the entire month, it would consume more electricity to keep the geezer on, so that any heat lost could be regained. And do the geezers remain on, at a low temperature? Or do they turn off completely when the water is hot enough? If the later is the case, then turning it off yourself is the same thing. Except it would never turn itself on again when the temperature dropped, so you'd being saving power. BTW, instant gas hot water systems have been around for decades. My Dad is using one that's at least 50 years old and still working perfectly well. JK, Wed, 3rd Aug 2011

See the whole discussion | Make a comment

Not working please enable javascript
Powered by UKfast
Genetics Society