Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Will my water filled pipe freeze?

  1. Apr 30, 2007 #1
    I have a problem that my technical engineer training can't solve.

    We have static water pipes installed under a canopy of a building. The coldest air tempreture recorded is -8.5 Celius during winter. We can assume that the air tempreture stays at -8.5 Celius and there is minimal air movement. If we assume that the pipe and water starts at 6 Celius and is exposed to -8.5 Celius for 8 hours will the water in the pipe freeze?

    The pipe is 32mm OD, 25mm ID. The pipe is carbon steel.

    Part of the answer is a formula for a curve that relates heat loss and time to tempreture. The differcult problem becomes when the tempreture reachs zero then the latent heat starts to be given up.

    Currently me and the building inspector are having an arguement as to whether the pipe should be dosed with antifreeze. I need a copy of the calculation to show the inspector.

    Thanks for any help.
  2. jcsd
  3. Apr 30, 2007 #2
    What is the water for? (It seems the only factor for deciding on antifreeze should be whether it all recirculates.) Why "6" Celsius? It does seem a valid question (how much insulation does a pipe require?), although if you get to the point where latent heat matters, then you're outside of the building safety margin.
    Last edited: Apr 30, 2007
  4. Apr 30, 2007 #3

    The water is in a fire sprinkler pipe. From climatic data expect worst case a high during the day of 6 celsius followed by an over night frost of -8.5 celsius. Even if the point where latent heat is factored in that is still okay, so long as we don't have solid ice.

  5. May 1, 2007 #4
    Fire sprinkler? Just make sure the antifreeze is dissolved in properly, you'd be very silly to build an emergency system that doesn't work properly (or fails permanently) if the climate changes a little.

    I don't like the way you've approached your climatic data. Why do you assume the water will warm completely (to the ambient 6 degree temperature) during the day, and not assume it will cool to ambient temperature at night? If it doesn't have time to cool to ambient temperature at night, it can't have time to warm up to the maximum during the day, so it will be at about the average -1.25 celsius all week, which is completely useless (it would freeze solid). You also assumed minimal air movement, which cannot be relied upon (imagine there is a cold front moving through): you should assume the outside surface will be the same as ambient temperature. Provided your pipes aren't physically connected to anything acting like a big heatsink, you'd want enough insulation that it only cools a few degrees over the worst week, in other words you'd want the above-ground pipes to be indoors.

    I'd expect the building inspectors know something about safety, why are you so unwilling to follow that advice? How significantly does it alter the cost?
    Last edited: May 1, 2007
  6. May 1, 2007 #5


    User Avatar
    Science Advisor

    It's going to freeze. At the least it will cause you troubles. I have seen people have problems with water pipes in homes where the water is moving and the pipes still freeze. Are these pipes out in the open or are they in an enclosed area such as an eaves area or the like?

    Whether the air is static or not really doesn't matter. You're only going from 6C to -8.5 for 8 hours and it's a 1" pipe.

    I fully agree that if you are relying on the latent heat as your safety factor, you are not being safe enough. That system has to work whenever and every time. You need to insulate or run heat tracing down the pipes to ensure free flow.

    Where are you located? Don't they have specific code requirements for this?
    Last edited: May 1, 2007
  7. May 1, 2007 #6
    I fully agree. Moreover when the latent heat begins to "help", it is too late: water is freezing.
  8. May 1, 2007 #7


    User Avatar

    Staff: Mentor

    Right - that isn't safety factor you're talking about. It is already into failure.

    If I had to guess, I'd say you will only lag the ambient temperature by an hour or two with a pipe size like that.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook