How should I heat our household water?
August 22, 2022 5:05 AM   Subscribe

Our instantaneous gas hot water system has a thermostat that goes to 55oC (131oF). All of our taps are mixers. If I want, say, a 43oC shower, which will use less energy: leaving the thermostat at 55oC and mixing in cold to get to 43oC, or setting the thermostat to 43oC?

This seems to be a divisive and somewhat complicated issue (among my decidedly-not-physicists friends and family, anyway).

I insist on dialling the thermostat back to 43oC. Everybody else seems to think I'm mad.

Some say that at the end of the day you're getting the same volume of 43oC water coming out of the shower, so it must use the same energy.

Others say it depends on the temperature of the cold water, and possibly the target temperature of the mixed water and/or the starting temp of the hot water, and nobody can possibly work it out so why not just leave it at 55oC so they don't have to keep changing it back again, and by the way, you left the toilet seat up again.

Your sage adjudication would be much appreciated, especially if I am right.
posted by some little punk in a rocket to Science & Nature (24 answers total)
 
Your shower is a mixer too, right, not electric?
posted by lokta at 5:18 AM on August 22, 2022


Response by poster: (In Summer, where I live, you might need to wait a few minutes to even get cold water out of the 'cold' tap, but I suspect that's not your issue)

(Just up the road in Canberra. Also, the PWEI graphic on your profile made me feel like an old man. Best take another shower.)
posted by some little punk in a rocket at 5:39 AM on August 22, 2022


You could probably make an argument about heat loss from the pipes between the heater and the shower being higher when the water inside them is hotter, as well as the amount of water remaining in the pipes which ends up cooling when you turn the shower off containing more energy... But the difference is going to be marginal. And you should insulate your pipes regardless.
posted by alexei at 5:41 AM on August 22, 2022


Presumably if you set the thermostat to 55C, then water will enter the heater at 'cold' temp and be heated to 55C, before being sent to your mixer. At your mixer it will arrive at just under 55C and be mixed with water at 'cold' temp to reach 43C.

If you set your thermostat to 43C, then water will enter the heater at 'cold' temp and be heated to 43C, before being sent to your mixer. At your mixer it will arrive at just under 43C and no water at 'cold' temp will be mixed in.

To me, I think that heating the water to 55C rather than 43C will use less energy. However, it is possible that (depending on heat loss in the pipes), if you set your thermostat to 43C you will end up with a shower that is at a slightly lower temperature than that. If your pipes travel a distance and are uninsulated it may be much lower.

The reason for not setting the water heater lower in general, is usually that legionella and other bacteria are more likely to survive in water that is below 50C. If you have an on demand heater, that's perhaps not as much of a problem.
posted by plonkee at 5:48 AM on August 22, 2022 [1 favorite]


Also, many people believe that setting the thermostat affects not only the temperature water is heated to, but how quickly it heats up. (That is, they believe that setting the thermostat to 55C will make it reach 43C more quickly than if you set the thermostat to 43C) I believe that is a fallacy.
posted by plonkee at 5:52 AM on August 22, 2022


Some interesting information at "How that temperature should hot water be?" - basically you will save money by not heating it so high. There are some applications where water hotter than the low 40s is useful however: water around 60C is better for cleaning and supports fewer germs, for example.
posted by rongorongo at 5:55 AM on August 22, 2022


All sorts of little variables affect this. Is the heater gas, which sends some heat out in the exhaust, or electric, which doesn't? Do the pipes between heater and shower pass through living space, an unfinished attic, or a concrete slab? Is heat lost into the home welcome (because the weather is cool) or does it increase AC usage?
posted by jon1270 at 5:58 AM on August 22, 2022 [1 favorite]


To me, I think that heating the water to 55C rather than 43C will use less energy.

There is no way this is true. Think about it like this: to heat the water to 55C, you first have to heat it to 43C, then you have to heat it another 12 degrees. Heating to 43C takes the same amount of energy whether you stop at that point or not, then there is the additional energy needed to heat it to 55C.
posted by number9dream at 6:08 AM on August 22, 2022 [3 favorites]


To add one more thought: if you heat the water to 55C and then cool it down to 43C by mixing, you are essentially wasting all the energy needed to get it from 43C to 55C.
posted by number9dream at 6:10 AM on August 22, 2022 [1 favorite]


To add one more thought: if you heat the water to 55C and then cool it down to 43C by mixing, you are essentially wasting all the energy needed to get it from 43C to 55C.

This is not true; you will use less 55C water than you would 43C water. If you heated the water to 55C and then let it cool to 43C, then you would be wasting all the energy needed to heat it from 43C to 55C. But 1L of 55C water will have as much impact on the end temperature of e.g. the shower as, idk, 1.3L of 43C water.
posted by mskyle at 6:19 AM on August 22, 2022 [3 favorites]


But 1L of 55C water will have as much impact on the end temperature of e.g. the shower as, idk, 1.3L of 43C water.

Yep. That's why the answer here hinges on water heater efficiency, pipe insulation, and downstream effects of the energy that leaves the water before it gets to the shower head.
posted by jon1270 at 6:27 AM on August 22, 2022 [1 favorite]


I think the problem minimum temperatures are meant to address is the problems inherent in storing warm water. Warm water is more conducive to bacterial growth than colder water. To avoid the risk of nasties multiplying in your hot water tank, they recommend minimum temperatures for these. If you heat water as needed as opposed to storing hot water for later use that is probably less of a concern.
posted by koahiatamadl at 6:30 AM on August 22, 2022


The reason for not setting the water heater lower in general, is usually that legionella and other bacteria are more likely to survive in water that is below 50C.

The recommendation is to keep your heater at 60° C. Keeping stationary water at 43° C is in the prime growth zone. I'm not sure how that applies to on-demand systems.
posted by JoeZydeco at 6:40 AM on August 22, 2022 [2 favorites]


The amount of thermal energy that you need to put into a given amount of water to raise it from 20°C (say) to 43°C is the same no matter what you do. In other words, it takes the same amount of energy to:
  • Raise a given amount of water from 20°C to 43°C, versus
  • Take a certain fraction of that water, raise it to 55°C, and then mix it with the remaining amount of water at 20°C so that all of the resulting water is at 43°C.
Since the same amount of energy needs to be added to the water in both cases, the answer will hinge on how efficiently the energy can be added to the water in each case — in other words, how many units of energy your heater needs to expend to impart one unit of energy to the water coming out of the shower head. This will depend on the many factors listed by jon1270 above.

My gut instinct is that heating the water to a higher temperature will increase the rate of heat loss to the environment; generally, the rate of heat loss is greater for warmer objects. And heat loss to the environment will decrease the efficiency of the system. So, for example, if the water flows from the hot water heater through a long stretch of uninsulated pipe before it gets to the shower head, it will lose more energy to the air surrounding it if it's hotter, meaning that your water heater has to burn more gas to get the same effect at the shower head. (The lesser amount of water wouldn't make as much of a difference here, because the pipe will still be full of hot water the whole time.) But it's also possible that your water heater is more efficient at bringing water to 55°C than it is at bringing water to 43°C (I'm not really familiar with how these instantaneous systems work), which would cut the other way.

Personally, I would propose an experiment: keep the thermostat at 43°C for one week and record your gas consumption. Then keep it at 55°C for one week and do the same. Keep alternating weeks until a clear trend emerges, or until it becomes evident that the difference is minimal and that there are better things to worry about.
posted by Johnny Assay at 6:44 AM on August 22, 2022 [2 favorites]


This is a usability concern rather than an energy-efficiency concern, but what about the tankless heater's ability to meet the hot water demand in your household?

If you live in a one-bathroom home without a dishwasher and will never try to shower while doing laundry, then it's kind of a moot point. (Side note: a dishwasher, if you have one, is not going to run effectively at bathwater temps, it'll need 50-60C input.)

But if you/family will often double up beyond your system's capacity or you want to fill a hot bath expediently pretty often, it makes sense for the system to output x liters per minute of piping-hot water that can be mixed down to 1.3x liters per minute at comfortable human temp rather than outputting x liters per minute at 43C and losing water pressure and needing to mix it with cold water anyway.
posted by fountainofdoubt at 6:55 AM on August 22, 2022 [1 favorite]


So, this is an instantaneous heater, which means that warm water doesn't sit around in it for anywhere near long enough to get festy; the 60°C recommendation for hot water storage tanks doesn't apply.

At the end of the day, the amount of heat energy being delivered to the water that flows through the heater is going to be determined by two things: the vast majority of it is going to end up coming out through the shower head and flowing down the drain, and the rest is going to be wasted in assorted kinds of loss.

If you've got X flow rate of Y temperature water coming out of your shower head, and the water was at Z temperature before it entered your premises, the amount of heat energy you're losing in the shower is the same regardless of the peak temperature any given parcel of water achieved on its trip from the supply mains to your shower head. If you had the instantaneous heater cranked all the way up to 99°C then you wouldn't be burning multiples of the gas you'd need to raise it to 43°, because the mixer would be mixing in proportionally less of the hot-side stream for any given shower flow rate and the flow rate through the heater itself would be much lower. So the total amount of energy actually required to supply any given shower volume at any given temperature, ignoring losses, doesn't depend on the heater's thermostat setting.

But if you take losses into account, you're better off lowering the thermostat. Raising it reduces the temperature difference between the water flowing through the heater and the combustion products heating that water, which in turn reduces the rate at which heat energy moves from those combustion products into that water, so you lose more energy up the flue than you would with a lower thermostat setting. Note that this specific loss does not apply to an instantaneous electric heater, because those have no flue to divert heat energy away from the water.

Also, the hotter you make the water flowing out of the heater, the more losses you will see to atmosphere from the heater's own surface and from the pipework between the heater's output and the final mixer - both because the temperature difference between what's inside the pipe and the air outside it will be higher, and because the flow rate of hotter water from heater to mixer will be lower so any given parcel of heat energy is going to be spending more time in that hot pipe. These losses will be seen for both gas and electric heaters.

These effects won't be major, but they will always be there. Best efficiency is always going to involve having your instantaneous heater make your hot water only just as hot as it needs to be for its intended end use. Your understanding of the world is correct, and that of your friends and family is not.
posted by flabdablet at 7:20 AM on August 22, 2022 [4 favorites]


Incidentally, lowering the heater's thermostat can help in another way as well: instantaneous gas heaters shut their burners off entirely when their water throughput drops below a threshold level, both because that's how they know when to fire up in the first place and because there will always be some minimum heat output from a working burner below which the flame won't stably self-sustain and you do not want domestic pipework turning itself into a high pressure steam vessel under those conditions.

If you've got a heater thermostat cranked well up, and are consequently relying on cold mixing downstream of the heater to get you the end-use temperature you need, and the total end-use flow rate is set by rate restrictors as many modern building codes require, then the flow rate through the heater itself can get low enough that the burner just shuts off entirely and your shower suddenly goes cold. This is surprising and uncomfortable.

We saw this happen after a bathroom remodel that involved a new shower head with a much lower flow rate than the old one. The only way we could get a hot shower that stayed hot, without running outside in the cold to faff about with the gas heater, was to run the handbasin hot tap wide open for the whole time we showered, completely defeating the purpose of the low flow head.

Dropping the gas heater's thermostat from 60°C down to 45° and showering with the cold tap fully closed fixed the problem, but was a pain in the arse of its own when it came to doing dishes. We ended up upgrading the heater to a more efficient unit designed after low-flow shower heads became a thing.
posted by flabdablet at 7:43 AM on August 22, 2022 [5 favorites]


Your understanding of the world is correct, and that of your friends and family is not.

That said, it may well be that your understanding of your friends and family is not correct, and that the efficiency gain from your "rational" supported-by-Science! way to use your heater is so minor that it really isn't worth its cost in reduced domestic harmony. This has been a bitter pill that I've needed to swallow myself, in related fields involving electric room heaters used as substitutes for adequate bedclothes by teenagers.
posted by flabdablet at 7:47 AM on August 22, 2022 [4 favorites]


To me, I think that heating the water to 55C rather than 43C will use less energy.

There is no way this is true.

Yes, I said this completely the wrong way round!! What an idiot I am!
posted by plonkee at 8:11 AM on August 22, 2022


many people believe that setting the thermostat affects not only the temperature water is heated to, but how quickly it heats up. (That is, they believe that setting the thermostat to 55C will make it reach 43C more quickly than if you set the thermostat to 43C) I believe that is a fallacy.

At startup, the heater's burner will run flat-out until the heater thermostat's set temperature is achieved, at which point that thermostat will back the fuel flow off to maintain that set temperature at the heater's output.

The mixer's thermostat, in turn, will set the mixer's hot-side flow demand according to its own set temperature. So at startup, when the water that the mixer sees coming from the heater is still below the mixer's set temperature, the mixer will be running 100% hot-side input and 0% cold-side.

The mixer will not start adding cold water until after its hot-side input hits 43°, which it will do as fast as the burner is capable of making it do regardless of whether that burner then backs off at 43° or 55°.

So even not ignoring losses, the heater's thermostat setting has zero effect on the initial time taken for the shower head to start delivering 43° water.
posted by flabdablet at 8:53 AM on August 22, 2022


As so often, "it depends". Let's ignore insulation and efficiency for now.

If you have a constant flow through the heater, temperature control will be through gas control; dial in a lower temperature and you use less gas. You also have less total flow at the shower, because less cold is mixed in near the outlet.

If you have constant power at the heater, temperature control will be through water control; dial in a lower temperature and you get more water. The total flow at the shower will be independent of the heater setting.

I think the first one is more likely. So, lower heater temperature, less total water, less gas. But legionella; you'd be unlucky but severely annoyed.
posted by StephenB at 12:18 PM on August 22, 2022


At shower's end, if no use of hot water down that supply line occurs in the time it takes the water in the line to reach room temperature, all the energy used in heating the water left in the supply line is wasted (except as heat for the house). So, another vote for 43°.
posted by diodotos at 2:54 PM on August 22, 2022


The mixer will not start adding cold water until after its hot-side input hits 43°, which it will do as fast as the burner is capable of making it do regardless of whether that burner then backs off at 43° or 55°.

Only true if you have a thermostatic mixer. If you don't, you need to run the hot tap on full until it stabilizes, then add cold water manually. If you set it to be mixed from the start, it'll take longer and waste more energy with the higher hot water temperature.
posted by alexei at 3:41 PM on August 22, 2022


Yes, my explanation above did assume a thermostatically controlled mixer. Whether the sensing and control elements involved are mechanical and built into the mixer, or conscious and downstream of the shower head, doesn't really matter though.

As long as the mixer does start out drawing water only from the hot side, the shower will come up to temperature at the same speed regardless of the heater's thermostat setting.

If you have a constant flow through the heater, temperature control will be through gas control; dial in a lower temperature and you use less gas. You also have less total flow at the shower, because less cold is mixed in near the outlet.

Not usually the case with modern plumbing, which usually includes a flow rate restrictor inside the shower head itself. Usually that restrictor is the dominant factor in setting the final flow rate, and what's happening upstream of it makes very little difference.

You can find out whether or not your shower is controlled by such a restrictor by opening up one of the feed taps slowly and steadily. If there's a restrictor in place, there will come a flow rate beyond which opening the tap further makes no difference; turn the tap up to this-goes-to-eleven! and the shower will just keep on running at 3.
posted by flabdablet at 7:28 AM on August 24, 2022


« Older Gift voucher gone missing or red herring?   |   Air Force wives in 1971 South Korea? Newer »
This thread is closed to new comments.