I have a plastic tank in my garden to collect rainwater. It is general purpose IBC container (https://en.wikipedia.org/wiki/Intermediate_bulk_container). The tank itself is made of plastic, within a metalic cage. I have always heard that it is good practice to empty the tank during winter when temperature can go below 0°C. I know that frozen water has a volume about 9% bigger than liquid water, so a full tank will be damaged if the water freezes.

What I don't know however is what happens if the tank is only partially full. Will the water level rise to accomodate the ice's volume ? Or will the tank be damaged anyway by lateral pressure ?

I'm asking this because it seems like there will be a chronic lack of rainfall where I live, so it's best if I can start my gardening season with some water in stock.

  • $\begingroup$ Check out crystal-lattice formation - it starts at about 4°C. It's this that causes water to start expanding. I imagine that you'll find that below zero it expands in all directions and will not flow like a liquid anymore. $\endgroup$
    – Transistor
    Dec 6, 2022 at 17:58
  • $\begingroup$ Materials flow from pressure gradients. Air will shrink from cold. If the container is open to atmosphere, that would allow it to vent, but geometry of container will also matter. Envision what happens in increments. What happens as the tio layer of ice is pushed up by more ice forming below? $\endgroup$
    – Abel
    May 5, 2023 at 13:40
  • $\begingroup$ I idly wonder whether, by insulating some faces of the tank and not others, one could control the direction of freezing so as to ensure that, throughout the freezing process, there will always be a liquid region that can relieve any lateral compressive loads by being extruded upwards. $\endgroup$ Jan 1 at 16:32

1 Answer 1


If the inside of the tank is rough and the ice cannot move then it can burst the tank as the ice gets thicker, it also depends on the wall thickness and uv damage etc.

One "trick" I have seen is to have a foam block or two floating that can absorb the compression reducing the strength of the ice. However, not stayed around for days to check...

  • $\begingroup$ Thank you. Unfortunately that's not an option, as the hole of the tank is too narrow to insert any significant volume of foam. $\endgroup$ Dec 7, 2022 at 19:23
  • $\begingroup$ You only need to keep one surface free of ice. If the ice forms from the outer diameter, it will fit the tank. Water in the center will rise as the ice forms. Maybe top insulation will help? Alternatively, freezing from the center out will not break the tank: maybe insulate the sides and drop a steel pipe through the hole? $\endgroup$
    – david
    May 7, 2023 at 2:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.