1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Water Temperature Difference with Depth

  1. Aug 1, 2012 #1

    idp

    User Avatar

    I'm trying to find out if the water temperature in a hot water tank (household solar system) varies substantially from the top temp to the bottom. In essence in a stable system, what is the difference in temperature per centimeter drop in height of water in the tank? I assume that the tank size and shape should not matter too much (I stand to be corrected) ie a vertical tank vs a horizontal tank.

    Reason for asking is that to set the temperature of the water may depend/vary substantially on the position that the temperature sensor is placed in the tank. ie top, middle or bottom.

    Thanks
     
  2. jcsd
  3. Aug 1, 2012 #2

    mathman

    User Avatar
    Science Advisor
    Gold Member

    I can't give you a precise answer. However water tanks are heated at the bottom, so mixing is automatic and there shouldn't be a lot of variation.
     
  4. Aug 1, 2012 #3
    lots of different type systems.....see here for some discussion:


    http://en.wikipedia.org/wiki/Solar_water_heating


    general rule: hot water tends to rise.
     
  5. Aug 1, 2012 #4
    Also I am not aware of any research in that direction. However the http://earth.geology.yale.edu/~avf5/teaching/ResourcesGG535/Lecture5.PotTemp.Thermodyn.LapseRate.pdf [Broken] seems to be about 0.12C per 1000 meter. So the position of the thermometer in the tank may be irrelevant.
     
    Last edited by a moderator: May 6, 2017
  6. Aug 1, 2012 #5

    krd

    User Avatar

    My water heater has the heating element near the top of the tank. So, I switch it on, the top water can become very hot, and I can have a quick shower but the bottom of the tank is cold to touch (the top gets very hot).
     
    Last edited by a moderator: Aug 1, 2012
  7. Aug 1, 2012 #6

    russ_watters

    User Avatar

    Staff: Mentor

    It is pretty common -- dare I say typical to required for larger ones -- to have heating elements both at the top and bottom of the water heater. At the top only, however, would not work very well. Here's an example that discusses replacement: http://www.water-heater-repair-guide.com/waterheaterelementreplacement.html

    Here's a manufacturer's brochure picked at relative random, showing all of their larger ones with two elements: http://www.rheem.com/documents/professional-electric-rheem-electric-water-heating-brochure
     
  8. Aug 1, 2012 #7

    haruspex

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    I wouldn't have thought that was relevant here. That tells you how much a bucket of water would cool (due to adiabatic expansion) if you were to bring it up from depth swiftly.
    If the tank is heated from below, that's where the warmest water will be, with the gradient depending on convection versus rate of loss of heat at the top. If from above, it will depend on conduction versus rate of loss of heat from the bottom.
     
    Last edited by a moderator: May 6, 2017
  9. Aug 2, 2012 #8

    idp

    User Avatar

    Thanks for the replies, unfortunately none of them really address my question. Considering that the take-off point is at the top of the geyser, and on occasion while having a shower the water will become cold, (150l tank) indicating that within possibly 20 to 30l of water the temperature variation can be quite dramatic, obviously the whole 150l has not been heated. It could possibly indicate that there is a definite unseen barrier/division between the heated water and cold water??? as opposed to a defined gradient of change in temp with depth of water in the tank.
     
  10. Aug 2, 2012 #9

    krd

    User Avatar

    I'm pretty sure you're right about that. I'm used to boilers that get hot at the top, and are icy cold to touch at the bottom.

    With the heating element at the top, the hot water would have to push it's way down. The tank is under pressure from cold water at the bottom - so it seems, and just from touching these tanks (seeing if it's hot enough yet for a shower) , the hot water is trapped near the top. This is definitely the way my boiler works - I would prefer something fancier, but I am broke. The fancier, the more expensive to install and run. In Europe, the energy costs would give you a heart attack.

    My boiler heats from the top - it super heats the water (I'm deadly serious - it's very primitive - there is no thermostat, just an on switch). Which will get me a few minutes of absurdly hot water - hot enough to give serious burns if I wasn't careful. But when the super hot water runs out, it turns icy cold.
     
  11. Aug 2, 2012 #10
    You want to read up on convection

    If the heating element is at the top of the container, it's probably mounted upside down.

    As warm water is less dense than cold it will just stick at the top.
     
  12. Aug 2, 2012 #11

    krd

    User Avatar

    In my boiler, it's not a mistake. It's just cheaper to heat some of the water and not all of the water. A tank full of lukewarm water would be more expensive than a hot top.

    I could have a hot tank all the time. But my electricity bill would be obscene.
     
  13. Aug 2, 2012 #12

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    There's one design of Solar water heating that uses a small solar-electric (thermostatically controlled) powered pump which delivers very hot water directly to the top of the tank. They claim that you can have small amounts of usefully hot water available very early in the morning but that, eventually the tank will fill (downwards) with hot water. The heat exchanger uses a 'direct' system which requires a water softener to be used.
    It sounded like a neat idea to me but I guess it may have disadvantages.
     
  14. Aug 2, 2012 #13

    krd

    User Avatar


    Heat pumps have been around a while. A problem is the commercial installations can be expensive and often very bad. It's a combination of manufacturers, designers, and installers not knowing what they're doing. It keeps failing to take off, because people get burned on a very expensive pump that doesn't work very well. Lovely brochure, but overpriced, inefficient, and it plainly doesn't work.

    Recently, I heard from a friend, his gym (in England) have a heat pump for hot water - the tank is on the roof, he knows the gym owners, and they say the pump has worked out really well and they get cheap hot water even in the winter.

    I have heard (and seen video) of people in central London boring their own holes and laying copper pipe for their own heat pumps.

    The water passing through the coil doesn't have to mix with the water you're trying to heat, you just have the coil run through the tank you're trying to heat (at that point you use your electric pump to compress out the heat into the water).
     
  15. Aug 2, 2012 #14

    sophiecentaur

    User Avatar
    Science Advisor
    Gold Member

    You misunderstood me. This isn't based on a heat pump. It's based on solar heating panels. The small (water) pump only needs to work when the Sun is shining so it uses a small PV panel. It pumps slowly enough to ensure that the water out of the solar panels is 'hand-washing' hot and, because it's at the top of the storage tank, it is ready for use before the rest of the stored water.
    They seem to have though about most problems because the system uses flexible pipes which can cope with the fresh water freezing in winter nights. It melts quick enough when the Sun rises and the pump kicks in when it's hot enough.

    Talking of heat pumps, I am surprised that refrigeration units don't have a good reputation for that application when you think that domestic freezer units work and work and work.
     
  16. Aug 6, 2012 #15

    idp

    User Avatar

    The system I have comprises 18 evacuated tubes with separate tank/geyser with backup electric element. The dc electric pump circulates the water by sucking out the cold water at the bottom of the tank and pushing the hot water into the top. Whilst the solar system is working (ie no electric back-up element) I have to conclude that the gradient in temp difference between the top and bottom of the tank will be dependent on the rate at which the water in the evacuated tubes is heated. ie the hotter the sun, the faster the circulation, the rate at which the heat from the top water diffuses into the lower band of cooler water is effectively the "unknown", but that factor will in essence determine the gradient of change in temperature with depth. I guess that there may well be 3 distinct temperature layers, a hot layer at the top with a small temp diff within that layer, a band of water in the middle where there is a steep gradient of change in temp with depth where the heat diffuses between the hot and cold layers, and a cold layer at the bottom.

    When the electric element is used (being at the bottom of the tank) the whole scenario will change as the heated water will cause convection phenomenon which effectively "stirs up" the whole body of water.

    any thoughts on the above assumptions?
     
  17. Aug 6, 2012 #16

    idp

    User Avatar

    If my assumption in the above is true for the solar system, then the knowing position of the temp sensor in the tank becomes critical because the electronic control unit settings will limit the circulation of the heated water if the "desired temp" is reached. If the sensor is at the top of the geyser/tank, then you may well have a very small amount of usable hot water, whereas if the sensor is lower down, the system will circulate and heat a far greater volume of water until the sensor temp limit is reached.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook