Topic Options
#60277 - 09/02/14 02:27 AM Local stress analysis on pressure vessels without supports
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Dear all,

I am designing a U-stamp pressure vessel which is an integral part of a piping system (i.e., no independent supports for vessels & since vessel is in line with the piping, it is supported by piping). Our client has advised us to perform local load calculation considering local loads caused by equipment weight.

So I have decided to consider the longitudinal force and corresponding circumferential moment due to the weight of the vessel.

Now my question is whether WRC107 or WRC297 analysis can be done in this case?? Or should I go for FEA??
Also d/D ratio in this case is more than 0.5; which is out of limit for both WRC107 (<0.33) & WRC297 (<0.5)

Somebody please come up with some quick valuable responses.


Bixindas
Mechanical Design Engineer

Top
#60280 - 09/02/14 05:31 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Also please advise whether wind & seismic load analysis are to be considered for these type of unsupported vessels.

Do wind & seismic loads hold any other effects other than on the calculation of forces and moments at supports base, on pressure vessels without any external attachments like platforms ????

Top
#60325 - 09/04/14 07:49 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Not even a single reply??? Moderators????

Top
#60351 - 09/07/14 06:52 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
You say "I have decided to consider the longitudinal force and corresponding circumferential moment due to the weight of the vessel". Such approach is adequate only when you have "anchors" at nozzles (i.e restrains able to take not only displacements-forces but also rotations-bending moments) and consequently the rest of piping system has no influence on the equipment.
You may have not this case and you rather need analyze your piping system having an in-line component which needs a special attention.
A piping system usually has valves as insensitive in-line components and consequently they can be modeled as rigid elements, your piping system has a special in-line component, (which basically is made by three cylinders- two nozzles and vessel body, quite sensitive at junctions nozzles-to-body). However, usually piping systems are analysed with software which considers beam elements instead "cylinders".
So you (or your stress department familiar with Caesar or similar software) need to model the piping system around your equipment, extend it "enough" as the boundary conditions have no influences on your equipment; the equipment itself would be modeled with three Caesar "stick"/beam elements - and 2 CNodes- in order that the final report will show loads in junctions. To have accurate results it is recommended to include also the local flexibility in junctions (reported by a FEA software as FEpipe)
A Caesar run of your piping sysyem, considering sustain (weight), expansion (loads due to restrained displacements due to temperature change) and occasional (wind, seismic) will clarify the loads in nozzle junctions. If your case does not fit with WRC, you need a FEA analysis of these junctions, considering the loads reported by Caesar analysis.

Top
#60358 - 09/08/14 06:48 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Hi dear,

Thanks for replying and keeping this alive. smile

The reason why I have decided to consider the forces and moments due to the weight of the vessel is that the nozzles are bolted to the piping flange, which gives an anchoring effect to the nozzles. So this will enable me to analyze the local stresses at the nozzle pipe & vessel shell interface. And these local stresses can arise only due to the weight of the vessel (unlike in the cases of supported vessels - for supported vessels local stresses are of course caused by piping loads only)

Am I correct??

And as you said, the piping stress analysis can be done in normal way considering the vessel as an in-line component and entering its weight as rigid in CAESAR II or any other piping stress analysis software. My concern is, what is happening inside that 'in-line component' as seeing typically through the eyes of a static equipment engineer, and the vessel is U-stamped as well.

Top
#60362 - 09/08/14 03:35 PM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
Bixin,

I think you must familiarize more with piping. Pipe flanges are not anchors. True, you can bolt-up a vessel support and that connection acts as an anchor because the foundation is stiff enough. But in piping, for a flange bolted-up to other flange, you will never get an anchor status there, in fact that joint may be weaker than the rest of piping.
IMO your equipment will be influenced by the piping beyond the nozzles as every equipment connected to piping by flange joints. How much- it depends on the piping configuration.

Best regards.

Top
#60386 - 09/10/14 07:56 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Dear mariog,

But unlike every equipment, this equipment is unsupported. If we goes deep in to finite element analysis at the interface of nozzle and shell, I strongly suspect there will be some local stresses acting due to the weight of vessel. The reason for this I would say is still because nozzle is attached to the piping and vessel is unsupported.

Piping department assures that piping can bear the load of the equipment. So it is evident that piping including the equipment as a system would be safely designed.

I still wanted to know the extend of WRC analysis that whether it can tackle this kind of analysis as I mentioned above.

In between could you please advise if occasional loads has to be analyzed separately for the vessel?? The piping stress analysis team would have already considered this since it is a single system including the vessel.

What do you say??

Top
#60403 - 09/10/14 02:10 PM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
I made a sketch, hoping we can have a common "strength of materials" language.
Try to evaluate whether the bending moment in flanges (and in connections to that in-line box) is the same for the two cases shown. Note that the supports are beyond flanges in both cases, so you can say that the in-line box "is unsupported". Consider also that the box weight is negligible vs. forces F.
With this example, just hope you can discover how the piping around can influence the in-line component connections.

For your actual case you say "The piping stress analysis team would have already considered this since it is a single system including the vessel."
In this case, why not use the results already reported?


Attachments
Sketch.pdf (585 downloads)


Top
#60413 - 09/11/14 01:20 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Thank you very much for the sketch and your valuable time.

I can understand your point; but my concern is not yet clarified.

To make my point more clear I have made some mark-ups and notes in the sketch you made and is attaching herewith.


Attachments
Sketch - Reply.pdf (638 downloads)
Description: Attachment



Top
#60414 - 09/11/14 03:00 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
I try to answer you.

1. As a static equipment engineer, under normal circumstances, you do not need to be worried about the piping and/or piping supports. However, you should understand that the piping would impose loads considerably higher than the loads given by the weight of the in-line equipment. Consequently you need to team work with piping/stress specialists.

2. I hoped that the sketch as example will convince you about point 1. The associated bending diagram is as in the attachment; you can see that, in my example, the bending diagram is constant between supports and is given by force F acting on cantilevers of length a. Consequently, in my example, the bending moment in J1 and J2 is given by the "external" force F and supports position rather than the weight of equipment.

Hence, as a general rule...no, the local stress in these 2 points is not developed by the the weight of equipment alone.

3. The results depend on how stress engineer will model your equipment in his piping analysis. As I said in the beginning of our conversation, the optimum approach is to model the equipment by three elements introducing the local flexibilities (by Vendor, by FEA) in J1 and J2. The stress report shall show accurately the loads in J1 and J2 and you or Vendor can go ahead by WRC or FEA.
However, sometimes (often?) this approach is not possible and it is conservative to consider in piping stress analyze just one rigid element having weight and to report to Vendor the loads in equipment flanges given by stress report; it remains Vendor shall evaluate the consequences (local stress) in junctions J1 and J2 based on information given.


Attachments
Bending diagram.pdf (646 downloads)


Top
#60463 - 09/13/14 05:33 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Dear,

Actually I meant to say that I am not worried about the piping or piping supports since this vessel is unsupported. I believe that since the vessel is free to move with piping, there can't be any local loads getting accumulated in the piping-vessel interface. And hence there can be any so called 'external piping loads' in this case. However as in normal piping case, the forces and moments acting at any point along the length of piping would be acting here also, but this can be transmitted through evenly spaced supports in the piping after proper stress analysis of the system considering the weight and location of the equipment.

In all other cases (for supported vessels) piping would impose loads on vessel nozzles and this has to be considered for a safe design and I agree 100% with that.

I hope my point is understandable; or am I going in the wrong direction??

Please advise.

Top
#60465 - 09/13/14 09:06 AM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
Bixin,

I guess we can speculate for ever. Having no actual information about the piping around I cannot guess what is there or to interpret "free to move with piping" as no stress in piping and/or piping components (including your in-line equipment). And, as I tried to explain, in my understanding the piping stress calculation is made using "beam theory" and refers rather to statics chapter....

But I think your problem has a simple solution.
Step1- look into the piping stress report and collect the loads in nozzles or junction points (if they are shown), for all cases considered by stress team;
Step2- decide what to do with these values reported (WRC, FEA... you or Vendor).

My best regards.

Top
#60466 - 09/13/14 10:52 PM Re: Local stress analysis on pressure vessels without supports [Re: Bixin]
Bixin Offline
Member

Registered: 05/15/14
Posts: 10
Loc: Dubai, UAE
Yeah.. I think that is the better solution. I will ask the piping guys to model the equipment in the line and will collect the loads at the junctions after analysis. The way they model the equipment is the crucial point here I guess.

Thanks mariog for your support and advises.

Meanwhile, In WRC analysis if curves are out of limit, can we do extrapolation when stress ratios are not so high?? And the results obtained, are they admissible??

Top



Who's Online
0 registered (), 124 Guests and 0 Spiders online.
Key: Admin, Global Mod, Mod
May
Su M Tu W Th F Sa
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Forum Stats
12065 Members
14 Forums
16973 Topics
75151 Posts

Max Online: 303 @ 01/28/20 11:58 PM
Top Posters (30 Days)