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

Pipe Line Failure

  1. Dec 25, 2008 #1
    As a part of my Project I'm experiencing a serious problem, "Blow out of Pipes in Compressor". In that we are trying control the blow out. Let me tell you the real scenario. The outlet of the HP Compressor is connected to a pipeline which in turn is connected to the reservoir.

    HP Outlet - Flange(parallel thread)(Cast Iron) - Pipeline(Taper thread)(Alloy) - Reservoir (10 kg Capacity)

    During times there is a blow between the flange and the pipe. When we examined, the threads in the Pipe and Flange are Blunted. The compressor is coupled with High -Speed Engine from which it gets its power.
    I want to know the reasons for this blow out and how to proceed for solving this problem.
    Expecting your Reply
    Thank You,
    Last edited by a moderator: Dec 25, 2008
  2. jcsd
  3. Dec 25, 2008 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    A pipe ruptures when the local stress intensity exceeds the local critical stress intensity in a material. Flaws in materials allow for stress concentrations at the boundaries of the flaws, which then propagate slow until a critical size is achieved and the 'crack' propagates at about the 1/3 the speed of sound in that material. Cracks will stop where the material is softer or the stress drops below a threshold value that precludes propagation.

    Flaws may exist in engineered materials where the flaws are created during the manufacturing process. Some flaws may nucleate and evolve during operation, e.g. localized corrosion due to material incompatibility or localized erosion/corrosion in flowing systems.
  4. Dec 25, 2008 #3


    User Avatar
    Science Advisor

    After you saw the blunted threads, did you check the pipe for cracks? I don't just mean a visual inspection, as you may have a crack that would not be apparent to the naked eye.
  5. Dec 25, 2008 #4
    I did not inspect., Do you mean that crack in the thread will have a greater impact., Please let me know how to avoid this.,.,
  6. Dec 25, 2008 #5


    User Avatar
    Staff Emeritus
    Science Advisor

    One would have to describe the pipe at the connection with the flange, or show a picture.

    What are the dimensions? What kind alloy is the pipe? What type of connection - welded or mechanical?
  7. Dec 25, 2008 #6
    It's a 5cm Dia pipe attached with the flange which is connected to the HP side of the compressor. The Pipe is of Galvanised Iron one,,,
    The flanges are connected by mechanical bold and nut., The flange inturn is connected to the flange by threading.,,.
  8. Dec 25, 2008 #7
    It's a 5cm Dia pipe attached with the flange which is connected to the HP side of the compressor. The Pipe is of Galvanised Iron one,,,
    The flanges are connected by mechanical bold and nut., The GI-Pipe inturn is connected to the flange by threading.,,.
  9. Dec 25, 2008 #8
    “HP Outlet - Flange(parallel thread)(Cast Iron) - Pipeline(Taper thread)(Alloy)”
    You should not connect parallel thread to taper thread. They are not designed to be married and will gall.
    What pressures are you dealing with? You should provide numbers because what is high pressure in one industry is low in another.
    What fluid are you dealing with?
    What codes are applicable to your project?
    Cast iron can be dangerous if it breaks. It should be used in low-pressure systems only.
    Alloy: Alloy of what?

    I could go on but you need to provide more information about what you are building and your technical expertise etc.
  10. Dec 26, 2008 #9


    User Avatar
    Science Advisor

    Sounds like nucleus has already identified your problem: your using a parallel thread with a non-parallel thread. The two thread types will not mate correctly (galling is typically due to similar hardness in the material which can occur even with the correct thread types).

    If the threads are not mating properly and are then subjected to high pressure, then I would definitely expect to see the threads blunted since they are probably being pulled apart under pressure.

    Make sure you're using HP rated fittings as well (cast iron is not meant for that type of service).

  11. Dec 26, 2008 #10


    User Avatar
    Science Advisor

    The straight threads mixed with pipe threads is a definite no no. Also, you are using cast iron on the outlet of an HP compressor. What is the pressure ratio across the compressor? It doesn't take much of a ratio to produce some decent temperature rise.
  12. Jan 28, 2009 #11
    Also, flanges have either flat or raised faces. They can not be mixed. Doing so can cause cracking of the flat faced flange.

    Another reason for the leakage could be gasket blowout. Have you checked that? Flat faced flanges should have full face gaskets.

    Further, if the flanges have different materials and coef. of expansion, then a temperature excursion could cause stresses.


    Last, flange ratings are temperature dependent, and the flange class rating is not a pressure. You need to look at the ANSI standard to find the pressure rating, or contact the manufacturer.
  13. Jan 29, 2009 #12


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi nucleus. I don't think that's what karven said (or meant). He/she said:
    which I take to mean that the flange fasteners are conventional straight thread fasteners but the flange is a "threaded flange" which means the pipe has a pipe thread and the flange has a corresponding pipe thread such as is shown in this picture:
    http://www.chnflange.com/images/threaded%20flange.gif [Broken]

    Posting some pictures of your failure and providing photos is not just going to be helpful to others, but it is a courtesy to provide relevant information for people trying to help. Regardless, a threaded flange on a compressor is poor practice. The pipe thread represents a tremendous stress riser and the vibration from the compressor will commonly fatigue the pipe at this thread. There shouldn't be a pipe threaded flange on the compressor outlet. If you need to have the flexibility to orient your flange, use a lap joint flange instead. Otherwise, use a weld neck flange. If the break isn't through the thread (circumferential break following a thread root) then please provide sufficient information so folks here aren't spinning their wheels.
    Last edited by a moderator: May 3, 2017
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook