Momentum due to cold venting through a pipe ... 2

[somacast]

Hi ,

My question is about a 16 inch flare line, on which a branch was taken for cold venting in case the flare is out, this branch was added about 20 years ago, and its a 12 inch branch, the venting is said to be (55000 Kg / hr) of gas (density 1.1 kg/m3) , and an old man did a simple calculation back then and said its ok, I can not post his papers here due to confidentiality, but what he did basically was : he used Force = ( mass x delta velocity) / ( change in time) , and taking a time of 3 seconds, he found the force , however as it is a live load he just multiplied the force into 2, then added weights of valve and pipe and checked modulus of section if its safe, and said its safe, and it worked like that for 20 years....

now out of no where some big guy said you need to review this, it does not seem right ..... so he wants me to review it again with even assumption of higher flow (60000 Kg/hr) ... can I do the same ? it worked for 20 years and it was used many times for cold vent although it causes high noise ...

I attached a schematic of the case , hope this info is enough ...

thanks in advance..

 

[LittleInch]

I think you need a piping stress analysis. You have significant steady load and moment on the branch connection. What you describe sounds only like the additional transient load during a release event. That's important but isn't the only force here and it's the branch connection which is the weak spot. It may have worked for a long time but you might be fatiguing the joint very fast. Or it might be perfectly ok. Do the analysis and find out.

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

 

[somacast]

hi @Littleinch ,

thank you for the reply, but where to start, what to check over there ?? thanks

 

[3DDave]

For any analysis an analyst needs to understand both the loads and the structure. There are guesses that one can make about them to ensure the problem can have a satisfactory solution, but it requires a detailed analysis to verify that a particular solution is acceptable. The better answers will come from a CFD to determine a better estimate of the load distribution on the pipe wall and a Finite Element analysis would tell if there are any areas of other wise unexpected stress concentration.

I would be surprised if anyone had a parametric analysis waiting around to put the numbers and structure diagram you have into and get a result.

A lot can happen in 20 years. Corrosion and fatigue are at the top of the list.

An interesting book called Engineers of Dreams was written to deal with a similar problem. Bridge companies typically start with one or two people on simple designs and as they gain experience they move up. Eventually they get big enough they are gaining the most interesting contracts, but then the founders retire or die and the knowledge they have of what fails goes with them. The writer wanted to gather failures into a single book so that engineers would stop repeating them. Some failures had been repeated at intervals going back the Roman Empire and are still seen in recent construction.

It seems like the old man had enough experience to judge a good solution and avoid a failing one, but was unable to pass that along. It's a problem that most engineering only records what works, but not what was rejected and why it was rejected. I suppose it is job security and wanting to hide failures.

 

[somacast]

@3DDave , hi , thanks alot for your reply, however I really dont think that this all applies to the case over here, the old guy was not Mr. wisdom or anything he was just relied on because he was the only one, and his calculations are so rough that even the number plugged into the equations were wrong, and its not only in this case he always did, while on the other hand I do agree with you that the firm did not encourage him or even force him to pass much of his knowledge...

and regarding the 20 years issue I will just re-phrase it as it seems I did confuse everybody with my first description: its not being re-checked because they worried about its deterioration due to age, that is not the case at all, because our piping systems are under excellent PM system and checked thoroughly all over the year, but they want to check if the system which was installed 20 years ago was good at first place.

anyhow all of this argument wont help at all , I just want a few hints if possible please to start from, as this job is stuck with me and its not going anywhere else (as usual)

Thanks in advance

 

[LittleInch]

I understand that someone now wants to see if what you've got is actually within acceptable limits of safety or just hanging on by the merest small margin...

The attached is an example of the forces involved and what happens if it goes wrong....

To me you need a piping stress analysis of your system - something like Caesar II. Then add the loads from the exit nozzle.

To me this is applying the momentum force from the nozzle = mass/ sec x velocity/ sec ( you'll need to figure out the velocity) to the 45 degree elbow. This will give you quite a large moment on your branch connection and I would be quite surprised if it didn't "fail" without some sort of additional support or reinforcement to take some of the moment off that tee connection.

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

 

[somacast]

@littleinch, hi, thank you for the important investigation report i will go through it thoroughly,

regarding my question, velocity can be found as we do have volumetric flow and pipe area,I even added an assumed wind force acting along with the horizontal component of the force (f cos45) . i multiplied the f by 3 as factor of safety and still looks within limit, as i took an allowable bendinf stress for the vent bottom as 0.4 yield ..

i do understand that full stress analysis is the right way, but that was the all calculations the old guy did 20 years ago..

 

[LittleInch]

You may well be right, but I don't think the bending of the pipe is your issue.

However a 12" branch will stand a fair bit of force.

What force did you think you have at the 45 degree nozzle?

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

 

[somacast]

Thanks again , I will finish it then I will tell you the number I assume to be over here and tell me then please what do you think..