C900 Pipe Burst 2

[Hayesmc]

On a project I worked on, a section of 12" C900 fire line completely burst while undergoing a pressure test. The attached image shows the length of pipe that blew up between two 45 degree elbows. Due to changes during construction, a high point existed at the elbow on the right, without an air release, and likely had an air bubble. As the tester introduced pressure to test the fire line (his goal was to test the fire line at 200 psi for 2 hours), the pipe section blew up at about 180 psi. Fortunately, no one was hurt. The tester said the air pressure at the high point was likely 500-600 psi and caused the explosion. Its also worth noting the section between the elbows was the last section to be installed - so its likely the contractor had to push and pull the pipe to make sure the connections could be made between the elbows. This likely introduced some extra stresses to the C900 pipe at the elbows. It also noteworthy that this section of pipe was left exposed during testing.


Does anyone have any experience with such an incident?


It seems prudent that as the fire line was pressurized, the volume of the air bubble decreased and increased air pressure. If the tester measure 180 psi water pressure, how could one calculate the air pressure at the air bubble?

It is common in our area to not draw up a profile for fire lines; therefore it would seem likely that high points and air pockets would existing in other fire lines and other bursting incidents would have happened - but they don't. This makes me consider the pre-loaded stresses in the pipe in order to match the elbows and whether this particular section of pipe material could have been substandard.

Any feedback or experience in this matter is really appreciated.

 

[LittleInch]

Ha,

"The tester said the air pressure at the high point was likely 500-600 psi and caused the explosion"

Utter and complete rubbish.

Pressure acts in all directions at the same time. There is no way on this earth that a flatish piece of pipe can have two radically different pressures in it at the same time when it is one piece of pipe. Against the laws of physics.

The tester is talking complete nonsense and is simply trying to get his company to avoid blame for what is in all likelihood a construction issues, though difficult to tell as the evidence is scattered to the 4 winds.

You will find your answer here "so its likely the contractor had to push and pull the pipe to make sure the connections could be made between the elbows. This likely introduced some extra stresses to the C900 pipe at the elbows. It also noteworthy that this section of pipe was left exposed during testing."

Leaving it open would allow both of the elbows to move, possibly in contraction as the pressure was applied, adding to the bending stress on the pipe. I can only guess how much force went into getting the flange faces to fit.....

So in short, your contractor is talking complete and utter Bullshit.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[cvg]

for buried waterlines, typical procedure is to backfill prior to testing and the contractor is required to bleed all air out of the line before it is tested. It doesn't sound like any proper procedures were followed

 

[GeoEnvGuy]

As others have stated you cant get a 600 psi air bubble in a 180psi water system. You can get a 180 psi bubble. Also the air bubble would be in the elbow unless they had the pipe bowed.

Retrieve a sample of pipe measure the thickness with a digital micrometer and compare it to the specs. Also look for markings on the burst pipe them compare to the spec. I would suspect that if it burst at 180psi it is probably rated as 100psi or 80 psi pvc pipe. If that all checks out then look at the elbow connections and if they are squeezing the pipe.

Unlikely the burst is due to a lack of confining pressure from being buried or thermal shock, or water hammer.

 

[Hayesmc]

Thank you everyone for your feedback- most helpful. I spoke with the pipe supplier further and he still contends the failure was an air pocket issue. I am supposed to get some data from them explaining the issue - will certainly share what I find out.

 

[itsmoked]

More likely the pushing and pulling may have resulted in scratches that affected the strength.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[LittleInch]

Can you post details of the flange joint.

It look to me like the flange is attached to the pipe using screws. Are these designed for PVC pipes?? I can see then being ok for DI puppies but not PVC.

PVC is very brittle and this whole set up doesn't look good to me.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[edwards1800]

The joint restraint is typical for C900. However, one thing that should be checked is whether or not the joint restraints were used correctly for that particular diameter. As an example, when using Megalugs with C900, there is a spacer on the bolts that must be left in. When using IPS OD pipe, that spacer is removed. If the contractor is using joint restraint for IPS OD, then the bolts would've been tightened too far and added stress to the pipe.

See this video at 00:32 for this important installation note:

https://youtu.be/nMIlusd1svo

Also, the points made about the air bubbles are absolutely correct. It was not the bubbles that caused it. Air in the line makes it harder to get up to pressure on a test. They do not get pressurized more than the water.

 

[Hayesmc]

Here is a closeup of the flange joint - or were you looking for a detail drawing? Also, the joint immediately to the south slipped - the joint restraint did not hold. Sure seems like this leverage on the stick of pipe between the elbows would be the culprit.

 

[SteveWag]

I worked on an existing 4500 PVC foot sewer force main laid parallel (in the same trench) to a gravity sewer line. Very low pressure at the pump discharge. The line design (Plans) showed fittings around the gravity manholes. Very low pressure at the pump discharge. Failures were monthly, some times twice a month. This was for a small firm and the president of the company handled this client. He insisted that I mitigate the water hammer issue. After a day at the pump station I returned with the conclusion that water hammer was not the culprit. After looking at a map of the failures I found out that the failures were all along the main, one being 19 feet from the open air discharge. I then asked that the damaged pipe and/or fittings be brought to the office from any additional failures. Soon I had enough to make a preliminary decision. I asked to be notified when the next repair was planed. This would result in some time delay as the site was 2 hours away. Sure enough, when I arrived, there was a failure next to a manhole, and NO fittings were in sight. Seems that the inspector (still our employee at that time) had ordered the contractor to warp the pipe and save the fittings, and their COST! Later,We, the design firm, contributed 75% toward re-laying the entire main.
This all being said:
1. The air bubble has the exact same pressure as the water pressure (in fact a little less because it is at the top) and,
2. The joints failed because of excessive deflection caused by either, probably both; deflection forces caused by installation and/or the increase in length of the PVC section caused by the slipping (extension) of the joint restraint.
Iron pipe should have been used. Not because Iron is better than PVC, just that it is better at handling joint deflection. With iron pipe it is almost impossible to break the pipe. The excessive deflection just causes the gasket to fail. With PVC, excessive deflection will stress the material and lead to circumferential failure.
BTW, nice photos, glad you stuck with it and made use of the opinions.
Steve

 

[djs]

Last month I had a similar issue. We were starting up a small bank of water pumps to supply washout water for some centrifuges. Pumps were about 30 Hp each self priming. PVC pipe was connected directly to the pumps without any type of flexible coupling. Ran for about two weeks before the PVC discharge pipe broke. The GC immediately accused us of over pressuring the discharge. Schedule 80 PVC pipe. I told the contractor that the failure was most likely the result of misalignment and vibration. The installed pumps could never exceed the burst pressure of the installed PVC pipe. PVC needs to be installed with absolutely no stress in the final installation. ANY misalignment will result in pipe breakage. PVC is not flexible and is very brittle.

 

[LittleInch]

So do you have a conclusion / told the tester he's talking rubbish??

PVC is not the most forgiving of materials and has the unfortunate tendency to fragment when it breaks.

Testing PVC in an exposed condition should not be allowed. Next time it will kill someone. Don't allow this to be brushed under the carpet. The consequences could easily have been very serious indeed.

I hate PVC for this reason.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.