Pipe Stress Analysis

[DCBII]

I'm primarily a structural engineer and am somewhat new to the world of pipe stress analysis. I have a few questions.

When is a pipe stress analysis typically required? Are there any piping systems for which pipe stress analysis is overkill?

Are pipe fittings typically stronger than the pipe itself?

 

[SNORGY]

Some of this is dealt with - weakly - in the governing Codes. I have a few rules of thumb that I use, but the work (separate) by folks like L.C. Peng and Charles Becht IV is a good guideline.

Regards,

SNORGY.

 

[curtis2004]

DCBII
Why are you asking this question? Just pure curiuosity or you have some "project" underway?
Curt

 

[if7005]

Anchored critical line pipe laid on your pipe rack or post may require stress analysis. You need pipe thermal trust load, slug load, or wind load to be considered in your design or your structure will be damage.

 

[BigInch]

2" diameter and less and other pipes operating below certain temperature differentials will not require stress analysis. Check B31.3 for the details.

We will design everything from now on using only S.I. units ... except for the pipe diameter. Unk. British engineer

 

[DCBII]

I've been asked to design some pipe supports (not the piping system itself). I asked my boss if a pipe stress analysis had been done (to determine the required supports) and it hadn't.

These pipes are 6" to 24" in diameter, and they are not at high temperatures. The largest pressure in any of them is about 120 psi.

I could do a quick finite element model to get a feel for the approximate stress in the pipe due to bending and axial load, but I don't know if something more substantial than a structural engineer making an rough estimate is required, especially at elbows, tees, and reducers.

 

[europipe]

Here are a few criteria to make stresscalcs,
this is a global incomplete list;

All safety pressure relieving systems (not including thermal
reliefs).

All lines operating above 315 degrees C (600 degrees F).

Lines 4” and larger operating above 150 degrees C

Lines 12” and larger operating above 93 degrees C

Lines 24” and larger.

All lines operating below -45 degrees C (-50 degrees F) which
require special “cold” supports.

Lines with Pipe type base supports and pipe type dummy-leg
supports with guides or fixed.

Lines with Base supports and dummy-leg supports welded to the
pipe or elbow without pad for pipe walls less than standard weight
for carbon steel/ low alloy steel and less than schedule 10S for
stainless/ high alloy steel/non-ferrous.

All lines connected to rotating equipment such as pumps,
compressors, steam turbines, and etc.

Lines connected to vessels on weigh cells.

All jacketed lines.

Lines connected to air cooled heat exchangers (fin-fans).

Lines located in pipe racks having exceptional length,
typically about 90M to 120M (300’ to 400’) or longer will need
review.

 

[SNORGY]

Augmenting europipe's excellent post:

* all buried pipeline and pipeline risers for which the arithmetic temperature difference from cold to hot in degrees Celsius is greater than the yield strength in MPa divided by 6.
* all tank farm piping for lines 6" and larger in which tank settlement relative to the piperack can be expected (though a lot of the times people just put in flex connectors to avoid the analysis)
* lines coming off the top of tall contactors down to a piperack

There might be other suggestions, but usually it's judgement and experience that determine what lines get looked at.

Regards,

SNORGY.

 

[BigInch]

Assuming you have no anchors to contend with, or specific loads having been given to you by the pipe stress engineers, a tyhpical customary practice is to take the dead weight of the pipes (pipe plus contents) x 20% as a lateral friction load (most friction loads are opposed by the same pipe on the other side of the beam, so they tend to cancel each other out) and use those as lateral loads (all in one direction) placed on the top flange of your pipe support beam.

You might want to add a separate lateral load for each large diameter pipes (>20") that you feel might not cancel each other. Long straight runs of pipe passing over your supports tend to cancel. Near pipe bends and turns, use separate lateral loads at 30% dead weight.

Dead weight loads from 1-16" diameter pipes on the pipe supports are usually assumed to be a 50 psf load on the rack space. Add larger diameter pipe's dead loads to that separately.

We will design everything from now on using only S.I. units ... except for the pipe diameter. Unk. British engineer

 

[curtis2004]

DCBII,
What is fluid inside pipes? Temperature?
Where is piping system located (jurisdiction, inside plant)?
Curtis

 

[DCBII]

The fluid is just process water I believe. It's not a hot process so I would imagine the water is around 60 degrees or so. This system is for a mining process in central Mexico.

 

[curtis2004]

DCBII,
What will happend if pipe will break supports and ruptures?
Is there any risk for people nearby, equipment, or mine?
Who is determining what kind of supports (saddles, anchors) piping should have?
Curtis

 

[if7005]

DCBII,
Sounds like you piping just subjected to weight load, and no complexity as like in processing plant. But, pipe rows included 24" fully water filled will be a big ton.

For general rule in process piping :
Use pipe sleepers with 6 m span.
Add weight pipe load margin about 10% ~ 30 %.
Put pipe axial+lateral stop at the middle of long straight lines. I guess no expansion loop required, but if you do for straight lines > 200 m in distance you have to put at the mid point between loop.
Put pipe guides every 12 m.
Apply pipe pad for 6" and larger.
Consider seismic load if any.

Finite element is not necissity, just use caesar ii or autopipe for specific software of piping stress analysis.