Failure on Frozen Piping 3

[leont]

I am not familiar with failures due frozen conditions. Currently I am performing an assessment for a piping failed due frozen condition, and I would like to know how I can get the maximum pressure developed during the frozen process until piping gets the burst pressure.
Im was wondering if the piping material gets brittle and it contributes to the failure, or it is an additional factor ?.

The failure happen in a SA 106 3" piping.

Thank you in advance for you support

Regards

leont

 

[BigInch]

The pipe can get brittle, but if I recall correctly for A106 it will be at around -20F. That's probably just an additional fact, because the freezing process at 32F will develop enough pressure to burst the pipe assuming no change to normal ultimate yield strength.

Answers here might help you to guess what might have happened in your system. Continued freezing of liquid water trapped between two initial ice plugs seem like a plausible explanation, given that ice contracts with further temperature decrease once below the freezing temperature.
ttp://

http://www.newton.dep.anl.gov/askasci/eng99/eng99358.htm

 

[waterpipe]

Here is a paper regarding what you are looking for. You can check the references as well. Hope this helps.

 

[Duwe6]

The above paper seems to confirm my experience: it requires multiple freezing episodes to rupture seamless [A106] pipe. The pipe is eventially stretched beyond its limit and a small "birdmouth" opening is formed.

This makes inspecting for affected areas fairly easy. If the pipe is visible and has been painted, there will be long, longitudinal cracks in the paint -- pipe has swelled some, paint didn't. Swelling can be confirmed using dial-calipers. My examination of freeze splits has shown them all to be ductile, rather than brittle fractures.

 

[BigInch]

One would expect those to be ductile failures, as that steel is afterall ductile at 32F. Freezing of cast iron, glass, some plastics, or concrete pipe would have very different failure consequences.

 

[zdas04]

I had a bottle of beer (that was overfilled) freeze and break once, definately brittle failure. Every pipe freeze failure I've investigated (both steel and HDPE) have been ductile failures. Since water only changes volume by 4%, I'm pretty sure that the failures I've seen have been multiple events (especially with the HDPE).

David

 

[kacarrol]

The article that Waterpipe posted was really interesting. I was kind of wondering about how this stuff worked when preparing my cottage for winter but didn't act on that thought at the time.

Thanks!
K

 

[BigInch]

This will date me a bit, but my first encounter with explosive freezing was opening the milk delivery box outside the front door one winter morning when I was a kid. Frozen milk on the bottom of the box full of broken glass.

 

[jte]

Never froze milk... Worse, when I was a freshman in college I put a can of beer in the freezer and forgot about it. Nice mess to clean up later.

On the topic of ductile vs brittle... If a water pipe was exposed in say a -40° environment and the flow was stopped, the water would freeze at 32°F / 0°C. The loading on the pipe is now determined. At this point, the pipe is ductile. But as the pipe and ice continues to cool towards -40°, the pipe may become brittle. Could we see a situation where the pipe sustains the load in a ductile regime but ultimately fails at the same loading when it becomes brittle?

jt

 

[waterpipe]

@jte:
it's true that I'm in -10 right now, but it doesn't mean that you can expose me to -40 environment and stop the supply!

regards,
waterpipe

(just to refresh your minds!)

 

[jte]

waterpipe-

I'm in about 50°F now. I assume your -10 is in °C?

Inadverdent stuff happens. I've designed a temporary repair for an air cooled heat exchanger header which was hydro- ice-tested. As far as I know, the piping didn't suffer, but then, I wasn't on site. This was in a location which averages 8°F / -13°C ambient air temperature in the winter. Exposed metal can get a bit cooler with radiative losses...

But, that doesn't change my question above!

jt

 

[leont]

BigInch & waterpipe,
Thank you so much for so professional answers.
Waterpipe, that paper tells me everything about the technical causes of these type of failure.

Thank you much

 

[metengr]

You can have either ductile or brittle fracture, it depends if the material exhibits a ductile to brittle transition temperature. Don't forget, copper water pipes will also fail except these are all ductile failures from tensile overload, caused by an over pressure condition. Copper has no ductile to brittle transition temperature.

 

[rconner]

Interesting discussion. To hear the "freeze pluggers" talk, they claim that at least that somewhat controlled length freezing solid of liquid pipe contents does not normally harm most pipes. It appears however (and perhaps as BigInch alludes) that some of these folks also warn however that any confinement or bulkheading of a pipe section wherein freezing occurs can definitely cause problems. In that latter case of freezing within an enclosed/confined space or strong, closed container, or e.g. trying to freeze plug a short section of pipe inadviseably near a bulkhead or closed valve etc., they say the inward expansion of the ice against the unfrozen liquid within can result in extreme pressures (some authorities say in some cases thousands of psi). I believe they say the extreme liquid pressure is what to worry about/what can/will break pipes in those cases. [Maybe yet another reason to normally design wherever possible loops instead of deadlegs or ends in piping systems?]

 

[BigInch]

As the link I posted clearly explains, ice contracts with decreasing temperature just like most "typical" solids, however water has a quite different behavior and expands just before freezing. I agree with the contractors that use freezing as a means to isolate segments of pipe, in that it does not apparently harm the pipe to freeze only a short length. Logical; if the water is freezing in 3 dimensions where only one, or two of which are restrained, there is no increase in pressure. The liquid water just before freezing expands freely in the direction of least resistance. When there is a development of a third restraint in the longitudional direction of the pipe, due to freezing in a second segment where liquid water is trapped between, the process becomes restrained in the 3 dimensions and pressure builds. Inermolecular forces driving the expanding water are indeed very high.

3% of water on Earth is freshwater, 2/3rds of that is ice.

I'm sure anyone reading this thread will find these additional facts about water useful, if just not plain interesting,

http://www.chem1.com/acad/sci/aboutwater.html

Don't miss this part of the page,

http://www.chem1.com/acad/sci/aboutwater.html#ANOM

 

[waterpipe]

@jte: Indeed I'm in -10°C weather but only one hour flight away from -10°F! You made me jealous mentioning your 50°F.

The experience I have regarding freezing condition was a designer who had assumed that water wouldn't freeze in a pressurized pipeline because of the pressure! So there was no provision for freezing of an exposed 25 bar pipeline (relatively high pressure in water works) connecting surge tanks to the main line. Obviously he had not seen the charts provided by BigInch and was not aware about the "weired" behavior of water. This ended up to an extra item for thermal isolation of the connecting pipeline with heater jackets.