B Why does hot water rise to the top of a water heater?

[sevensages]

TL;DR Summary

Why does hot water rise to the top of a water heater?

I am enrolled in a plumbing program with Stratford Career Institute. In the chapter in my plumbing textbook about Water Heaters, the textbook authors assert that hot water in a water heater rises to the top of a water heater. The authors assert that cold water sinks to the bottom of a water heater.

Here is an exact quote from the textbook:

"The design of the storage tank takes advantage of water's natural properties. Figure 18-2 describes how a water heater functions. The cold water supply pipe typically connects to the water heater at the top of the tank. A pipe on the inside of the tank directs the cold water to the bottom of the tank. As the water is heated, it expands. Since expansion makes it lighter, it rises to the top of the tank, where it can be drawn off. As heated water cools, it becomes denser and sinks to the bottom of the tank, where it is heated again. This circulation of water within the tank ensures that a reservoir of hot water will be available at all times."

Before I read this textbook, I never would have thought that if a given quantity of water was heated, that it would expand. I don't understand this. Does water expand when it is heated? Let's say that I have an empty 5-gallon size bucket. I pour exact one gallon of 50 degree fahrenheit water into this 5-gallon bucket. Then I put the 5-gallon bucket with one gallon of 50 degree fahrenheit water on a stove, and I heat the water in the 5-gallon bucket up to 100 degrees fahrenheit. Would the water in the bucket start occupying more than one gallon of space in the bucket when the water reached 100 degrees?

 

[Lnewqban]

Water expands when it is heated.
That is the reason behind the pressure relief safety valve that you see at the upper section of each water heater.

Please, see:
https://www.engineeringtoolbox.com/water-density-specific-weight-d_595.html

https://www.watts.com/resources/references-tools/t-and-p-relief-valves

 

[sevensages]

Water expands when it is heated.
That is the reason behind the pressure relief safety valve that you see at the top end of each water heater.

Please, see:
https://www.engineeringtoolbox.com/water-density-specific-weight-d_595.html

Well, my textbook says that water expands when it is heated.

So if a gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will be more than one gallon?

 

[jbriggs444]

Well, my textbook says that water expands when it is heated.

Your textbook is correct.

So if a gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will be more than one gallon?

Let us see what Google can dig up about the density of water as a function of temperature...

Search phrase: density of water versus temperature

First hit: https://www.engineeringtoolbox.com/water-density-specific-weight-d_595.html

You care about the blue curve on the bottom that cuts off at 212 degrees Fahrenheit. The pressure in the water heater will be a bit higher. So maybe you care about a curve in the gap between the blue and yellow.

 

[Lnewqban]

So if a gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will be more than one gallon?

Yes.
The occupied volume at 100° will be more than one gallon.

Note that the gallon is a unit of volume in British imperial units and United States customary units.
In both cases, at 50° and 100°, your mass of water will still be approximately 8.34 pounds.

 

[sevensages]

Both my textbook and Lnewqban have told me that water expands when it is heated. But I am still confused.

The idea that if I heat one gallon of 50 degrees fahrenheit water to 100 degrees fahrenheit, the water will expand to more than one gallon is VERY counter-intuitive to me. In fact, it is so counter-intuitive that I need someone to spell this out for me because I am unsure if it is true. I think that the statement that "water expands when it is heated" might mean someone OTHER than if I heat one gallon of 50 degrees fahrenheit water to 100 degrees fahrenheit, the water will expand to more than one gallon.

In other words, I think it might be true that water expands when it is heated, but I think that it somehow might also be true that if one gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will still be one gallon (not more than a gallon). Can someone help me understand this?

Just telling me "water expands when it is heated" ad nauseum is not going to make me understand this.

 

[jbriggs444]

Both my textbook and Lnewqban have told me that water expands when it is heated. But I am still confused.

The idea that if I heat one gallon of 50 degrees fahrenheit water to 100 degrees fahrenheit, the water will expand to more than one gallon is VERY counter-intuitive to me. In fact, it is so counter-intuitive that I need someone to spell this out for me because I am unsure if it is true. I think that the statement that "water expands when it is heated" might mean someone OTHER than if I heat one gallon of 50 degrees fahrenheit water to 100 degrees fahrenheit, the water will expand to more than one gallon.

In other words, I think it might be true that water expands when it is heated, but I think that it somehow might also be true that if one gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will still be one gallon (not more than a gallon).

You are wrong about this. A "gallon" is a unit of volume. It is not a unit for a fixed "quantity of water".

If you heat one gallon of water from 50 degrees to 100 degrees, you will then have more than one gallon of water. If you have it in a graduated cylinder, the cylinder will be more full. The water you have will be hotter and less dense. It will still weigh the same as before you heated it.

This assumes that the water is free to expand.

If you lock the water in a tight box with no air space anywhere and do not allow the box to expand, the volume of the water will be unchanged, of course.

 

[sevensages]

You are wrong about this. A "gallon" is a unit of volume. It is not a unit for a fixed "quantity of water".

If you heat one gallon of water from 50 degrees to 100 degrees, you will then have more than one gallon of water. If you have it in a graduated cylinder, the cylinder will be more full. The water you have will be hotter and less dense. It will still weigh the same as before you heated it.

Ok I believe you. You and Lnewqban have made me understand this. Thank you very much.

 

[sevensages]

Yes.
The occupied volume at 100° will be more than one gallon.

Note that the gallon is a unit of volume in British imperial units and United States customary units.
In both cases, at 50° and 100°, your mass of water will still be approximately 8.34 pounds.

I believe you. Now you have made me understand this. Thank you.

 

[sevensages]

I thought that I was through with this thread, and then jbriggs edited his post to write this:

This assumes that the water is free to expand. If you lock the water in a tight box with no air space anywhere and do not allow the box to expand, the volume of the water will be unchanged, of course.

What if the water is in a water heater? Wouldn't there be a little bit of air space in a typical water heater? If the water is in a water heater, would there typically be enough air space for the water to expand when the water was heated?

 

[vela]

In other words, I think it might be true that water expands when it is heated, but I think that it somehow might also be true that if one gallon of 50 degree fahrenheit water is heated to 100 degrees fahrenheit, the 100 degree fahrenheit water will still be one gallon (not more than a gallon). Can someone help me understand this?

When you say "water expands when it is heated," you're saying the volume the water occupies increases. But when you say "if one gallon of 50 degree water...the 100 degree fahrenheit water will still be one gallon (not more than a gallon)," you're saying the volume of water doesn't change. Both can't be true.

 

[sevensages]

When you say "water expands when it is heated," you're saying the volume the water occupies increases. But when you say "if one gallon of 50 degree water...the 100 degree fahrenheit water will still be one gallon (not more than a gallon)," you're saying the volume of water doesn't change. Both can't be true.

That is why I had to create this thread.

 

[jbriggs444]

What if the water is in a water heater? Wouldn't there be a little bit of air space in a typical water heater? If the water is in a water heater, would there typically be enough air space for the water to expand when the water was heated?

I am not a plumber, so I do not know for sure whether there is air space or just a pressure relief valve. In my area, code requires a check valve so that expanding water cannot force its way back "upstream" toward the water supply.

We can look up the bulk modulus for water. It is about 2.2 gigaPascals (GPa). That is about 22,000 atmospheres. You can think of that as how much pressure it would take to change the volume of some water by 100 percent. [More properly, it is 100 times the pressure it takes to change the volume by one percent]

If we look up the density of water at 50 degrees Fahrenheit, that's 0.99965 in grams per cc. At 100 degrees, it is .99308. That is a change of about 0.69 percent.

So we would need to apply $22000 \times \frac{0.69}{100} \approx 150$ atmospheres to prevent the water from expanding and keep it in the heater. This matches what I can read off the graph I'd posted up-thread. Without a pressure relief valve or some air space, your water heater may have problems.

 

[sevensages]

The bulk modulus for water is about 2.2 GPa

I am not a plumber, so I do not know for sure whether there is air space or just a pressure relief valve. In my area, code requires a check valve so that expanding water cannot force its way back "upstream" toward the water supply.

We can look up the bulk modulus for water. It is about 2.2 gigaPascals (GPa). That is about 22,000 atmospheres. You can think of that as how much pressure it would take to change the volume of some water by 100 percent. [More properly, it is 100 times the pressure it takes to change the volume by one percent]

If we look up the density of water at 50 degrees Fahrenheit, that's 0.99965 in grams per cc. At 100 degrees, it is .999308. That is a change of about 0.69 percent.

So we would need to apply $22000 \times \frac{0.69}{100} \approx 150$ atmospheres to prevent the water from expanding and keep it in the heater. This matches what I can read off the graph I'd posted up-thread. Without a pressure relief valve or some air space, your water heater may have problems.

I don't know what bulk modulus means. I've never seen or heard the word bulk modulus in my life.

I just looked at the picture in figure 18-2 in my plumbing textbook, and it does show an air gap at the top of the water heater. So I suppose that most water heaters do have an air gap at the top of them, which would allow the hot water to expand.

 

[jbriggs444]

I don't know what bulk modulus means. I've never seen or heard the word bulk modulus in my life.

Which is why I tried to explain how a bulk modulus works.

I just looked at the picture in figure 18-2 in my plumbing textbook, and it does show an air gap at the top of the water heater. So I suppose that most water heaters do have an air gap at the top of them, which would allow the hot water to expand.

Thanks. That makes sense since water will release dissolved gasses when heated. Though I am not certain how one prevents the gasses from building up in the water heater over time.

 

[sevensages]

Which is why I tried to explain how a bulk modulus works.


Thanks. That makes sense since water will release dissolved gasses when heated. Though I am not certain how one prevents the gasses from building up in the water heater over time.

I don't know anything about water releasing dissolved gasses when heated.

I think it makes sense that the water heater would be designed to have space at the top for the water to expand because that if the heated water expands and rises to the top, and if the cold water sinks to the bottom, that would allow a circulation of water in the water heater, which is desirable.

 

[Drakkith]

Per the 2nd link in the 2nd post:

When water is heated it expands. In a 40-gallon water heater, water being heated to the thermostat setting expands by approximately 1/2 gallon. The extra volume created by this expansion usually pushes back into the water tower of public water supplies, or into the well tank in homes with a private well, resulting in negligible pressure increase. However, on public or shared water supplies were a backflow device is installed on the water main feeding the house, the pressure can no longer push back into the tower or well tank. With no place for the expansion to go, pressure will dramatically increase.

So in my house, where there is no check valve or backflow device installed, the water is free to expand into the house pipes and public water supply and there is little to no pressure increase inside my water heater tank. If the water expanded into an air space I would expect to see a slight 'burst' of high pressure hot water whenever I first turn on the hot water in my sink or bathtub as the air expands and pushes on the water, but I do not see this so I guess the pressure in the tank is the same as in the supply pipes. Perhaps others have this issue if they have check valves or backflow devices, but I do not know.

 

[sevensages]

Some water heaters have two heating elements, one at the top and one at the bottom. Some water heaters only have one heating element. If the water heater only has one heating element, the heating element is at the bottom of the water heater.

If a water heater only has a heating element at the bottom of the water heater, an air gap at the top of the water heater would be beneficial because it would allow the circulation of the water in the water heater as water is heated in the water heater. If the water heater only has a heating element at the bottom of the water heater, and if hot water rises to the top and cold water sinks to the bottom, the circulation of the water in the water heater would cause all the water in the water heater to get heated. If the water heater only has a heating element at the bottom of the water heater, and if the water did not circulate in the water heater, you might have hot water at the bottom of the water heater and relatively cold water at the top of the water heater instead of the entire tank being hot water.

 

[Lnewqban]

If the water is in a water heater, would there typically be enough air space for the water to expand when the water was heated?

Air can be compressed, water is incompressible.

Air in the system is desirable because it acts as a spring against which expanding hot water can move, avoiding excessive increase of internal pressure that could rupture the walls of the tank, which are much weaker than the pipes and fittings (due to greater diameter).

Nevertheless, as air is compressed by the expanding water, the internal pressure of the whole water system also increases.
Most codes require either a bladder expansion tank (located above the water heater) or a relief valve (located upstream of the main shutoff valve of the building) to protect the entire plumbing system (besides the built-in pressure-temperature relief valve of the water heater).

Please, see:
https://plumbingtoday.biz/blog/3-things-that-can-cause-your-water-heater-to-burst

https://www.freshwatersystems.com/blogs/blog/how-to-fix-common-water-heater-expansion-tank-problems

 

[sophiecentaur]

TL;DR Summary: Why does hot water rise to the top of a water heater?

Before I read this textbook, I never would have thought that if a given quantity of water was heated, that it would expand.

OMG this thread has really got out of hand. Why bring in how water behaves under vast and unrealistic (for plumbers) pressures? Let's walk before we run.

Your "given quantity" is too loose a term and is confusing. You need to say "given mass". Then you can heat it and cool it and the only thing to change is its volume. Your book says:

TL;DR Summary: Why does hot water rise to the top of a water heater?

Since expansion makes it lighter

and that is wrong because the word is not 'lighter' but 'less dense'. Many text books attempt to use too many friendly words and that can totally misdirect the reader.

Start with 10 lb of water, heat it and it expands, cool it and it contracts but you still have the same mass of water. There's no way to restrict the volume of the water you've got in your tank . (Not measurably).

What happens in water heater tank is convection.
Start with a tank, full of cold water. Put a heating coil near the bottom and the water around the coil will warm up and expand. It's density will be lower than the surrounding water; it will move upwards and the cooler water will fall towards the bottom. That's Convection and the water will carry on circulating and mixing until it's nearly all at nearly the same temperature.

You may have come across 'Immersion Heaters' that have a long heating element, which reaches right down to the bottom and also a short element that only reaches 1/3 of the way down. Convection doesn't work downwards so the short element only heats the top 1/3

 

[russ_watters]

I am not a plumber, so I do not know for sure whether there is air space or just a pressure relief valve.

I think it makes sense that the water heater would be designed to have space at the top for the water to expand because that if the heated water expands and rises to the top, and if the cold water sinks to the bottom, that would allow a circulation of water in the water heater, which is desirable.

There should be no air in the system. Oxygen + metal = corrosion. Plus, if there is air at the top, it's just going to go out the outlet pipe. There should only be an expansion tank, which has pressurized air on one side, water on the other, and a flexible membrane in between.

 

[sophiecentaur]

There should be no air in the system. Oxygen + metal = corrosion.

If the system doesn't 'breathe' then the amount of air is limited.
Closed systems have an expansion vessel (as above) at a high point which is pressurised when installed. There's an overflow which will let water out if the pressure get too high but we're only dealing with around a Bar.
We're getting confused with several different factors here. I've lost where we are going with this.The actual expansion of water is important in that it causes convection but only in an indirect system with a 'hot tank' and a heat exchanger. Combi Boilers predominate (in UK at least) which heat water on demand and the circulation is by pump and not by convection.

 

[Lnewqban]

We're getting confused with several different factors here. I've lost where we are going with this.

From post #1:

Before I read this textbook, I never would have thought that if a given quantity of water was heated, that it would expand. I don't understand this. Does water expand when it is heated?

 

[russ_watters]

Closed systems have an expansion vessel (as above) at a high point which is pressurised when installed.

The expansion tank could be anywhere but it is typically installed by the water heater, which can be anywhere - high or low. Mine's in my basement.

There's an overflow which will let water out if the pressure get too high but we're only dealing with around a Bar.

3-6 barg, typically. We're talking about household domestic water here; that's the pressure you get from the city service.

 

[Tom.G]

Hi @sevensages, lets see if this helps explain the water expansion with temperature.

You have probably heard of molecules, they are two or more atoms bound together. For instance water is two atoms of Hydrogen bound to one atom of Oxygen, this leads to the chemical formula for water, H₂O... and they are not in a straight line, The H atoms are bound the the O about 105° from each other; that results in a water molecule taking up a bit more volume than expected

When the temperature of something rises the, molecules jostle around and bump into each other more often that at a lower temperature. This, on the average, keeps the molecules slightly further away from each other.

You can see that same effect in a crowd of people. If they are standing still they can/will be rather close to each other, rather tightly packed. On the other hand, if they are dancing to some wild music they wil be further apart.

With both water and people, their may be the same number of molecules/individuals, but when they are randomly moving around they will take up slightly more space.

Hope this helps!

Cheers,
Tom

 

[Baluncore]

In this discussion, the only "solid" parameter that does not change, is the mass of water involved in the experiment. A fixed number of molecules, has a fixed mass, and so weighs an amount determined by the acceleration due to gravity, wherever the experiment is being conducted.

The state variables, volume in gal(imp/US), temperature(°C/°F), or the confining pressure, (Pa/psi), are all interdependent.

 

[256bits]

Note that the gallon is a unit of volume in British imperial units and United States customary units.
In both cases, at 50° and 100°, your mass of water will still be approximately 8.34 pounds.

I have to point out a slight correction, and some unit trash.

The two gallons comprise different volumes in the two systems.
As do the two gils, which is 5 fluid ounce in Imperial, but only 4 fluid ounce in the American system.
Four gils make a pint for both systems.
This makes the cup size different for both = 10 oz Imperial vs 8 oz American.

Imperial gallon = 4.546 litres = 4 quart = 8 pint = 16 cup = 160 fluid ounce == weight 10 pound water.
American gal = 3.78 litres = 4 quart = 8 pint = 16 cup = 128 fluid ounce == 8.34 weight pounds water.

The American gallon is based on the wine gallon, based on 8.34 pounds of water at 39 F.
The British adopted the imperial gallon, based on 10 pounds of water by weight at 62 degree F, in the early 1800's.

Watch out in recipees. Whether they are using Imperial or American units, your cake may not quite come out expected under an incorrect assumption.

 

[sophiecentaur]

Watch out in recipees.

Whatever the recipe, volume is not conserved. It may be a convenient unit for measuring everyday fluids but it's hardly a good basis for talking Physics.

The expansion tank could be anywhere

True. I seem to remember a bleed screw on mine. Mine is actually at the highest position and the boiler is on the ground floor wall. But, of course, the boiler is closed and uses its own pressure; I'd forgotten that. The hot tank is upstairs .

 

[snorkack]

Generally, nearly all substances expand when heated.
Nearly all, because there are some very significant exceptions. Water is one of these few exceptions.
As stated, most substances expand on heating - but different substances expand to a different amount.
Substances can also expand differently in different directions. This happens to some solids and some composites. However, many solids and all liquids (and gases) expand equally in all directions.

I mentioned that water is one of the few and significant exceptions that do not always expand when heated. Water sometimes does expand when heated.
When you warm ice Ih from absolute zero, then it first shrinks slightly. Reaching about 934 g/l at -180 C.
On further warming, ice expands, significantly, from the said 934 g/l at -180 C to 917 g/l at 0 C.
And then on melting, shrinks drastically - to almost 1000 g/l of water at 0 C.

Interlude: how can you even measure expansion of a liquid?
Most liquids expand on warming - but so does the solid vessel which the liquid is in!
Well, you can make a careful comparison of vessels with rulers which you try to not let heat up with the vessel. It´s possible, but it took a lot of work, and a lot of mistakes.

Water shrinks to almost but not exactly 1000 g/l on melting:
https://www.engineeringtoolbox.com/water-density-specific-weight-d_595.html
1 l of water weighs 999,85 g at 0 C. And water shrinks further on warming - weighs 999,975 g at 4 C.
For the measurements: the British and USians use their (different!) gallons. (Which gallons are in Canada?). Much of the world uses metric... which has the advantage of some more rounded measurements!
Not all of them exact.
When the French revolutionaries back in 1790s enacted the metric system, they intended to define metre so that the circumference of Earth around poles - they knew that Earth is flattened like an orange and the equator is longer - should be exactly 40 000 km. They also intended to define kilogram so that 1 cubic metre of water at 4 C should weigh exactly 1000 kg.
They made measuring errors with both. When the etalons of metre and kilogram have been measured again, it turned out that the circumference of Earth was 40 008 km, not 40 000, as it should have been, and that 1 cubic metre of water at 4 C and 1 bar weighed just 999 kg 975 g, not exactly 1000 kg. The metre ruler had been made 0,2 mm - 0,02% - too short, and the kilogram weight, given the metre ruler, 25 mg - 0,0025% - too heavy.
The French decided against correcting their measures (they would have had to tinker with measures all the time as they got increasingly improved measurements!). So they settled at Earth being approximately but not exactly 40 000 km around and water weighing approximately but not exactly 1000 g/l.
In any case, water above 4 C starts expanding. It is 999,1 g/l at 15 C, 997 g/l at 25 C, and 958,4 g/l at 100 C (all at 1 bar).

 

[256bits]

(Which gallons are in Canada?).

We are on metric. Before it was the Imperial; as the commonwealth countries followed Britain.
It would be interesting to have a synopsis what the French colonies used as a system of measurement prior the Louisanna purchase and formation of Upper and Lower Canada, but never talked about.
I found this difficult site
https://www.bing.com/ck/a?!&&p=025e...zc2V0cy9kb2N1bWVudHMvTWV0cm9sb2d5LnBkZg&ntb=1

 

[256bits]

Generally, nearly all substances expand when heated.

A reference for the knowledge hungry
https://en.wikipedia.org/wiki/Negative_thermal_expansion

 

[snorkack]

Air can be compressed, water is incompressible.

All substances are compressible.
Otherwise, if you knock at one end of an incompressible item, when would its other end move? Answer is, instantly (because the item is not compressible). Therefore, before a light speed signal reaches the other end.
That said, expressing the bulk modulus as "2,2 GPa" is unhelpful. You don´t appreciate how big it is.
2,2 GPa is 22 000 bar. For comparison, the bulk modulus of air, and of any other gas, is equal to their pressure up to normally a few tens of bar.
22 000 bar is unrealistically high pressure for plumbers (and the bulk modulus is nonlinear on the scales of itself, most of time it grows). But it is useful to calculate the fractions.
If you have a 100 l water tank containing 90 l water and 10 l air at 1 bar and you add 100 ml of water into the tank, what happens to the pressure?
Answer: since the free space for air has decreased from 10 l to 9,9 l (you have not let any out), the pressure of air, and therefore water, will increase from 1,00 bar to 1,01 bar. Just 0,01 bar overpressure
If the same tank is completely full with no ullage left (but pressure relaxed to 1 bar) and you tried to add another 100 ml water, what will happen to the pressure?
Answer: since the added water is 1/1000 of the water already in, and the bulk modulus of water is 22 000 bar, you would need to increase the pressure by 22 bar.
That is, if your tank does not burst under these 22 bar, and if your tank is altogether incompressible.
The latter it physically cannot be because of special relativity.
Say you use a more realistic pressure. For example, mains pressure 5,5 bar above ambient, total 6,5 bar.
When you open a tap and let pressurized water from mains to flow into your tank, completely full but under just 1 bar, will any water flow in the tank?
The answer: 5,5 bar is 1/4000 of the water bulk modulus. Therefore, 1/4000 of the tank volume will be added due to compressibility of water - 25 ml will flow in the 100 l tank. However, the elastic expansion of the tank under internal pressure will be an amount varying depending on the geometry and elastic properties of the tank - and may well be much bigger than the elastic compression of water!

Also: it is relevant that it is the weight of water per volume that decreases on heating, not just its mass. Carry your water boiler to a spaceship, where water weighs nothing whether it is denser or less so, and switch on your heating element. Your boiler will not work. The hot water near the heating element will expand, but in absence of gravity, it will have no direction to rise, and it remains near the heating element - the cold water at the other end will not get warm.
Spaceships may not be so common issues, but portable boilers which may be turned altogether wrong way up might be a bit more common.

 

[russ_watters]

True. I seem to remember a bleed screw on mine. Mine is actually at the highest position and the boiler is on the ground floor wall. But, of course, the boiler is closed and uses its own pressure; I'd forgotten that. The hot tank is upstairs .

Are you talking about a hot water or steam heating system? This thread has been about domestic (drinking/bathing) water systems. I think in UK both are called "boilers"(and may be combined)? In the US a boiler is for heating your house(whether makes hot water or steam) and the thing that makes hot water for your shower is called a "water heater". Caveat: hot water for heating your house is fairly rare and archaic in the US, but house and domestic water heating may be combined in one device with separate circuits.

Anyway, a heating system needs air bleed (and runs at lower pressure), whereas domestic water doesn't because every sink/shower bleeds air out. Even in a mostly closed heating system air can collect because it is dissolved in and driven out of the water when heated.

 

[sophiecentaur]

Also: it is relevant that it is the weight of water per volume that decreases on heating, not just its mass.

= density?

 

[sophiecentaur]

This thread has been about domestic (drinking/bathing) water systems.

Yes. We don't use steam systems in UK in private dwellings, afaik. Steam has advantages in large /tall buildings with community heating. Is it still in use for large modern blocks?

The term "boiler" is not a good one but it's used as a matter of course in UK. Not the best of terms when the water is seldom anywhere near 100C. Nearly all CH systems use hot water radiators in UK. This is anotherTower of Babel; one of many in the domestic engineering world.

Even in a mostly closed heating system air can collect because it is dissolved in and driven out of the water when heated.

In a closed system, the amount of new water admitted is small (ideally zero). Hot water for washing etc is seldom much above 60C (for safety reasons and legionella). There doesn't seem to be much out-gassing at that temperature.

 

[snorkack]

= density?

Mass per volume has a nice standard term "density". Weight per volume does not have a good standard term I know of. Note that weight per volume goes to zero when gravity goes to zero even when density does not go to zero.
Also note that the derivative of water density with temperature gets very low and goes through zero at low temperature. If your water tank is full of 0 degree water then the density is 999,85 g/l. Turn on the heating element at the bottom of the tank, and warm water will not rise! It will shrink (to 999,97 g/l at 4 C) and therefore sink to the bottom. Only when the bottom water is slightly above 4 C will it start to convect upwards, eroding out the bottom of the icy cold water above.

 

[sophiecentaur]

Mass per volume has a nice standard term "density".

An ideal term for an OP who claims to have little knowledge of Physics. Surely this sort of thread should try to help an OP and not make life more complicated. It's so much more concise and than 'lighter' or 'heavier' yet still familiar.

Perhaps the behaviour of water around 0C would be better for another thread. We could give him Cold Feet?

 

[Ken Fabian]

@sevensages - I think that textbook description is wrong - mixing up what happens when the heater element is on with what happens when it is off. Stratification happens when the heating element is off.

As the water is heated, it expands. Since expansion makes it lighter, it rises to the top of the tank, where it can be drawn off. As heated water cools, it becomes denser and sinks to the bottom of the tank, where it is heated again. This circulation of water within the tank ensures that a reservoir of hot water will be available at all times."

You only get stratification - a hot layer above a less-hot layer - after the heating element is off for a sufficient period of time that the convection from it ceases.

During heating there is convection driven circulation that keeps mixing the contents, preventing stratification. Any stratification is not driven by the heating element making hotter water that rises - a vigorous process - it comes from much slower, less vigorous cooling from pipe inlets and outlets, that are made of metals like copper, brass and stainless steel and cooling of the (imperfectly insulated) tank itself.

The strength of the circulation from that convection when the heater is off is insufficient to mix the contents, allowing the cooled water flowing down to settle. Rather than cause mixing, the slow flow causes the top of that pool of cooler water to rise slowly, as a layer.

I note that our own hot water system has the outlet (and pressure relief valve) lower than the top of the tank, so there is an air space in the top of the tank. Not sure but I expect dissolved gases being released keep it from filling with water or replaced by water vapor. Any corrosion concerns are dealt with by use of non-corrosive materials and a sacrificial anode.

 

[Ken Fabian]

And (duh!) cold water coming in near the bottom will be the main cold water source and will tend to stay there, without much mixing - it won't make the sustained convection cycling that a heating element does, that breaks down any stratification.

 

[russ_watters]

And (duh!) cold water coming in near the bottom will be the main cold water source and will tend to stay there, without much mixing - it won't make the sustained convection cycling that a heating element does, that breaks down any stratification.

There's a major design requirement at work when it comes to putting the cold water in at the bottom and drawing hot water off the top: supplying consistently hot water for as long as possible. If the cold water entered at the top and hot water was drawn off the bottom there'd be more mixing and the temperature would start to drop soon after you start using hot water.

 

[Klystron]

Caveat: hot water for heating your house is fairly rare and archaic in the US, but house and domestic water heating may be combined in one device with separate circuits.

While generally correct for houses in the US, I was somewhat shocked when buying a condominium (condo) unit to find it heated with the same hot water as used by a sink or shower. I have owned 3 condos in a warm climate (CA and NV) that use 'aquatherm' systems to heat the condo. Also a two story house ~1600 square feet with a 75 gallon water heater.

Hot water from the hot water heater outflow runs through a T-connector to a radiator when the thermostat trips activating a taco valve. An electric fan blows through the radiator driving warm air through HVAC air ducts. Sure enough, HVAC techs call a plumber to work on this hybrid heating system.

 

[sevensages]

@sevensages - I think that textbook description is wrong - mixing up what happens when the heater element is on with what happens when it is off. Stratification happens when the heating element is off.



You only get stratification - a hot layer above a less-hot layer - after the heating element is off for a sufficient period of time that the convection from it ceases.

During heating there is convection driven circulation that keeps mixing the contents, preventing stratification. Any stratification is not driven by the heating element making hotter water that rises - a vigorous process - it comes from much slower, less vigorous cooling from pipe inlets and outlets, that are made of metals like copper, brass and stainless steel and cooling of the (imperfectly insulated) tank itself.

The strength of the circulation from that convection when the heater is off is insufficient to mix the contents, allowing the cooled water flowing down to settle. Rather than cause mixing, the slow flow causes the top of that pool of cooler water to rise slowly, as a layer.

I note that our own hot water system has the outlet (and pressure relief valve) lower than the top of the tank, so there is an air space in the top of the tank. Not sure but I expect dissolved gases being released keep it from filling with water or replaced by water vapor. Any corrosion concerns are dealt with by use of non-corrosive materials and a sacrificial anode.

@russ_watters
@jbriggs444
@Lnewqban
@sophiecentaur

What do you all think about what @Ken Fabian wrote here about stratification?

Do you agree with Ken Fabian that my textbook's assertion that hot water in a water heater rises to the top is wrong?

Before Ken Fabian made this post, I was supremely confident that my textbook is right.

 

[Ken Fabian]

I will add that our own HWS has a temperature gauge; the owners' guide informed us (warned us) that when the heater switches on the gauge will initially show cooling, from the cooler water in the bottom and hotter water at the top becoming mixed. The temperature sensor must be near the top of the tank nearer where the outlet is.

When the heating cycle ends and the convection circulation from that ceases I would expect the contents to be well mixed, with little difference in temperature between top and bottom.

It isn't that the heating doesn't cause hot water to rise - clearly it does - just does so too vigorously to allow stratification; stratification arises from less vigorous colder water inflow and from downflow from cooling of the inlet, outlet fittings where there is no insulation, plus some cooling from the tank body and imperfect insulation.

 

[sophiecentaur]

Do you agree with Ken Fabian that my textbook's assertion that hot water in a water heater rises to the top is wrong?

Before Ken Fabian made this post, I was supremely confident that my textbook is right.

Your text book is clearly wrong where it doesn't specify what 'the amount' of water he discusses actually is. You examine the same mass of water as it heats up and cools. Using volume (gallons) is not good for the description od convection and lighter / heavier is also a bad choice because that gallon of water (in a bucket) will always 'weigh' the same, despite expanding and contracting. All you need to say is that hot water, being less dense than cold water, will float up, being displaced by the more dense cold water.

The topic of unwanted stratification is only relevant in a system that's been designed wrong; you would always have at least part of the heat source set low down in the water. In the UK, the hot water tank has a coil of tube (a heat exchanger) which goes pretty near the nottom of the tank. This establishes a gradient of temperature with no stratification - except right at the bottom where there can be a very small quantity of water which the convection doesn't reach. Of course, there's no mixing of water that flows through the 'boiler' (but it never 'boils') and the water that comes out of the taps.. Often, the water that passes through the boiler splits between radiators and water heating. 'Combi' boilers have a heat exchanger in the boiler so, again, the instant hot water from the tap does not flow through the boiler itself.

 

[sophiecentaur]

I was somewhat shocked when buying a condominium (condo) unit to find it heated with the same hot water as used by a sink or shower.

Are you sure it's done that way? In UK there is a heat exchanger to isolate the boiler water from the washing water. It's common to use additive in the sealed heat circuit and you wouldn't want to wash in that stuff.

 

[Klystron]

Are you sure it's done that way? In UK there is a heat exchanger to isolate the boiler water from the washing water. It's common to use additive in the sealed heat circuit and you wouldn't want to wash in that stuff.

After owning and refurbishing several Aquatherm units, I am certain the same water heater tank provides the hot water source for all uses. You are correct and I was mistaken if I implied that used water returns to the basic house tank to be reheated. A much older system in my first condo in Santa Cruz County, California, diverted used water to a seconday "gray water" holding tank IMS.

My previous house featured a large water heater in the first floor garage piped to a 'maniblock' distribution device (also archaic) for household use and a separate distribution device for appliances. Copper pipes then directed hot water to the house Aquatherm heater located in a maintenance attic above a second-floor walkway above a twenty foot drop. I was apprehensive every time I crawled up there for inspection and maintenace on the air filters and fan. The hot water circulated quite a bit then dumped into municipal sewer lines. Mind you, Southern Nevada temperatures do not require much home heating. Apologies if a bit off-topic.

 

[sevensages]

It isn't that the heating doesn't cause hot water to rise - clearly it does - just does so too vigorously to allow stratification; stratification arises from less vigorous colder water inflow and from downflow from cooling of the inlet, outlet fittings where there is no insulation, plus some cooling from the tank body and imperfect insulation.

I don't understand this.

Why would heating the water be too vigorous to allow stratification? Please explain this using simple language.

 

[sophiecentaur]

Why would heating the water be too vigorous to allow stratification?

You are too worried about this, I think. Stratification is a big word but it doesn't mean it's a 'big idea' in your home. It's what you get when the situation is more or less in equilibrium - like an hour after the heat source is turned off. The 'eco' setting for electrical water heating uses an element 2/3 from the bottom of the tank so the bottom 2/3 won't be heated by convection. I wouldn't have used the word "vigorous"; The Power output of the heater will just affect the speed of circulating water

The heat affects the density. If there is cooler (more dense) water above then that will cause convection. Heating a pan on a stove is the most extreme situation because nearly all the pan base is heated and will cause small areas of convection cells.
Stratification is what you get when convection misses a region of water that's the same temperature or lower than the water above and higher than the water below.

Stratification is a phenomenon that can be seen in nature (In water and in the atmosphere) but the heat supply tends to be more complex and of course, the systems were not 'designed' like a domestic heating system. Stratification is one of the features of a Weather event.

 

[russ_watters]

@russ_watters

What do you all think about what @Ken Fabian wrote here about stratification?

Do you agree with Ken Fabian that my textbook's assertion that hot water in a water heater rises to the top is wrong?

Before Ken Fabian made this post, I was supremely confident that my textbook is right.

I don't agree. I think what your textbook says is fine. There's pretty much always more you can dissect out of how things operate, and the textbook doesn't go into excruciating detail, but everything it says is correct/true. Some specifics:

@sevensages - I think that textbook description is wrong - mixing up what happens when the heater element is on with what happens when it is off.

I think it's pretty clear. The textbook is very clear in talking about when the heating element is on:

"As the water is heated, it expands. Since expansion makes it lighter, it rises to the top of the tank"

Heating element on drives convection.

You only get stratification - a hot layer above a less-hot layer - after the heating element is off for a sufficient period of time that the convection from it ceases.

And it says:

"As heated water cools, it becomes denser and sinks to the bottom of the tank, where it is heated again."

That's less clear about whether the element is on or off, but clear enough for me. And probably because the water is always cooling when it's away from the heating element, whether the heating element is on or off. It's just that the convection is much more strongly driven by the heating element. But I see no reason to quibble over how the textbook explained it.

Note: since the water heater is insulated, what happens when the water is not being used is pretty minor/unimportant. Yes, you'll get convection and the heater will kick-on to maintain temperature, but it turns on rarely and the convection and stratification are minor.

I note that our own hot water system has the outlet (and pressure relief valve) lower than the top of the tank, so there is an air space in the top of the tank.

That would be very surprising if true, unless it's some sort of integrated expansion tank. There's no good reason I can think of to have an exposed airspace but there are bad reasons/reasons not to. Can you share the make and model?

 

[russ_watters]

Your text book is clearly wrong where it doesn't specify what 'the amount' of water he discusses actually is. You examine the same mass of water as it heats up and cools. Using volume (gallons) is not good for the description od convection and lighter / heavier is also a bad choice because that gallon of water (in a bucket) will always 'weigh' the same, despite expanding and contracting.

There's some colloquialisms there, but I see no need to come down so hard on the textbook's verbiage for this. The writer clearly understands since s/he points to density in the next sentence:
"Since expansion makes it lighter [than an equal volume of colder water], it rises to the top of the tank, where it can be drawn off. As heated water cools, it becomes denser and sinks..."