When freezing a copper pipe with a refrigerant, how does one know when the pipe has enough ice to "cork" the pipe and halt water flow? Is there an indicator or a rule of thumb?

  • What's the scenario where you don't have a valve to turn off the water, and need to freeze the pipe? – Steve Sether Apr 30 at 3:45
  • 1
    When the pipe spits it should be done. – crip659 Apr 30 at 10:56

There are numerous water pipe freeze kits on the market. You can do a computer search for "water pipe freeze kits" or try "pipefreezekits.com". I however have never used a commercial freeze kit due to their cost. Instead of this method, I always used dry ice which was available at many ice plants. I have frozen water pipes up to 3" diameter in buildings. Since the size of the frozen pipe length is very short, and the discharge length was very long, there was never a chance for the pipe to burst.

How I did this was by using "shaved dry ice" put in a bath towel and wrapping the towel around the pipe using several outer wraps of the towel to act as insulation for the dry ice. Then you wait for the pipe to show an ice layer on the pipe just past the towel. As long as you maintain the dry ice the pipe will remain frozen. I once froze a 3" supply water line in a school where all the upstream valves would not close completely, including the utilities street shut off valve and kept it frozen for 10 hours.

Make sure that you protect any exposed skin and eyes with heavy work gloves and eye protection. The dry ice is unforgiving and will freeze skin and damage eyes in seconds.

  • 1
    I almost commented that you didn't actually answer the question, then I reread it and realized you did. I emphasized the actual answer part. The rest is excellent info, it just kinda hid the answer itself. – FreeMan Apr 30 at 11:30
  • I'm largely curious why this tool+technique exists. Is it largely when the valves won't work? I can't think of any other reason you'd use freeze a pipe, unless you're somewhere you simply can't turn off the water for some reason (commercial space maybe?) – Steve Sether May 1 at 4:23
  • @SteveSether you hit it. It provides a valve where none exists. – K H May 1 at 4:30

Do a flow test. Water will no longer flow. Note that freezing water inside a copper pipe can easily damage it. The technique in the video could conceivably be useful in some complicated emergency situation that justified the risk, but valves are put at convenient locations in water lines for a reason, and if you find you really need a valve somewhere, you're probably better off shutting off the main and installing one than freezing a pipe.

Also note that copper is an excellent thermal conductor, so if you're thinking of freezing a pipe "just enough" its likely that your plug won't last very long and the moment the layer of ice touching the pipe melts, the pipe is typically straight and smooth enough to allow the plug to move down the pipe.

  • The real problem with freezing a pipe is when the water has nowhere to go, since water is in-compressible. So if you close a valve, then froze he water between two closed valves, you'd likely burst a pipe somewhere since ice will expand, and the water won't have anywhere to go. If you leave the feed open, and for good measure let the other end drip a little, there's almost zero chance you'll do any harm since the water will be pushed back out your house, and anything freezing on the other end will just just push more water out the drip. – Steve Sether Apr 30 at 3:22
  • @SteveSether yes I agree there could be careful techniques and situations in which an unconventional technique could be valuable. Simply freezing a long enough section solid enough would likely crack the pipe though. I would also imagine such a device could be built into a system risk free with the correct type of some other pipe. Strong disadvantage that you would need extremely low temperatures to attempt to stop a pipe with even a little flow though. – K H Apr 30 at 3:26
  • I'd say it'd be pretty easy to freeze a section of pipe with a refrigerant like that, and a slow drip. It's not simply just the temperature involved, but applying a refrigerant directly to the pipe, and quickly absorbing the heat of the water to vaporize the refrigerant. – Steve Sether Apr 30 at 3:30
  • @SteveSether I definitely don't see it as impossible, just limited, risky, challenging to do right in a failsafe and useful way and unnecessary in most cases. An example application would be when you had a system you wanted to drain fail-safe on power or coolant failure like a nuclear reactor. They use a plug of working fluid maintained by constant refrigeration and if the coolant loop fails the plug melts and drains the reactor. I couldn't think of an example where this would be necessary in a common water system other than troubleshooting some complicated assembly. – K H Apr 30 at 4:20
  • @SteveSether water is not incompressible - it does have a bulk modulus of compressibility and that becomes significant at pressure. Just depends the pressures one is working at... – Solar Mike Apr 30 at 4:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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