When to consider pressure thrust in WRC local load analysis?

When to consider pressure thrust in WRC local load analysis?As per pvelite june 2001 newsletter refered in pvelite help it depends on type of support arrangement of piping either it is anchored or expansion loop.please share your views. 9 days ago Like CommentFollow Flag More 13 comments

Follow John Z. John Z. Wang, Ph.D., P.E. Search this board with "pressure thrust". There is already some discussion on this topic. 7 days ago Like

Follow SHUKLA SHUKLA AMITKUMAR if anchor or expansion joint with tie rods or loop in radial plane of nozzle orientation then no need to consider thrust in WRCB calculation, however you need to be sure of nature of forces (tensile or compressive) 7 days ago Like

Follow Trevor Trevor Seipp As John said, this has been discussed already. The answer is "never". Read WRC107 (actually, WRC107 has been replaced with WRC537). The pressure thrust was considered in the development of the calculations. 7 days ago Like

Follow Donald Donald Shaw We recently looked at a case of taking Caesar piping loads and evaluating them with a finite element analysis and determined that Caesar output loads does not include the pressure thrust. we are adding the pressure thrust to the Caesar loads. I would be interested to know if others agree. 6 days ago Like

Follow Trevor Trevor Seipp Donald - are you talking about a WRC 107 (actually, now WRC 537) evaluation? If so, then please read the Bulletin. And then read my response above.

Oh, and don't even bother about an FEA comparison to WRC 537. This has been done many times in the literature (search for prior PVP papers and JPVT papers), and the general consensus is that the correlation between FEA and WRC 107/537 is poor. But, WRC 107 was pretty good for the day - and it was a general improvement on Biljaard. 6 days ago Like

Follow Donald Donald Shaw Trevor, thanks, No we are not talking about WRC 107 or 537. We are looking at very closely spaced nozzles with nozzles in a head, and therefore are relying on a finite element analysis instead of WRC. As part of this effort we have confirmed that Caesar does not include axial pressure thrust loading so we are adding that to the Caesar output loads. Sorry if I did not make that clear before. Appreciate your comments. 5 days ago Like

Follow John Z. John Z. Wang, Ph.D., P.E. http://www.linkedin.com/groupItem?view=&gid=1794527&type=member&item=95336990&trk=gr oup_search_item_list-0-b-ttl&goback=%2Egna_1794527 5 days ago Like1

Follow Tony Tony Smith Hi Trevor, where do you find in WRC-107 that the pressure thrust has been included? I find in 3.3.5 and 4.1 where it states that evaluation of stresses resulting from internal pressure have been omitted.

5 days ago Like

Follow Trevor Trevor Seipp Tony - I am traveling right now and don't have my copy in front of me. It's in one of the appendices at the end, after the graphs. 5 days ago Like

Follow Trevor Trevor Seipp Donald - if you are doing an FEA, then you absolutely need to include the pressure thrust load because, you are correct, no pipe stress analysis program will include that load as part of the nozzle loads.

I will definitely give you credit for knowing that it was a load that needed to be included and then checking/verifying/validating whether or not it was included. In my opinion, that demonstrates an excellent engineering approach. 5 days ago Like

Follow Luiz Paulo Luiz Paulo Faustini Can pressure thrust influence the overturning calculations when the tank is empty? 5 days ago Like

Follow Fernando Fernando Lidonnici Gentlemen,

let's try to deal with this complicated problem making some example.

1. Suppose that your nozzle is a short one, ending with a flange and closed by a blind flange, like in the case of a manhole. Of course in this case there is a pressure thrust, but nobody would require a calculation to WRC using the pressure thrust as an additional load: in fact the code calculation of the opening reinforcement takes already this load into account. Of course you have stress

concentrations around the nozzle due to the pressure thrust, that can be better detected using a stress analysis: but these stresses are not calculated using WRC, which only takes into account the external loads. Moreover, the external loads considered by WRC will cause a reaction on the vessel supports, and this is not the case of the pressure thrust, which is self-equilibrated.

2. Suppose your nozzle is connected to a straight pipe (of negligible weight) and the straight pipe is connected to another vessel (this is, for example, the case of two stacked horizontal heat exchangers). Then you may have 3 different situations.

a) No expansion bellows on the pipe, saddles of the two vessels are designed in order to allow movement in the vertical direction (for example with springs): the loading situation is the same of case 1.

b) No expansion bellows on the pipe, but saddles are rigidly bolted together: a thermal axial compressive load will arise when the temperature inside the exchangers will raise. You can take this thermal load into account using WRC: but you have to consider only the thermal load, not the pressure thrust. Note that also the saddles must resist this load.

c) There is an expansion bellows in the pipe. In this case the unbalanced thrust due to the pressure inside the pipe cannot be taken by the nozzle wall because of the presence of the expansion bellows: the pressure load (directed inside the vessel) will cause a thrust that will be transferred to the saddles, exactly as any other external load. In my opinion in this case a pressure thrust (opposite to the one you should normally imagine, because it is directed towards the inside of the vessel) must be considered from the point of view of equilibrium: the problem is that it should not be considered for the WRC calculation, because it is not transferred through the nozzle wall, and therefore its effect is simply to relieve some part of the pressure stresses, which in any case are to be considered separately, without using WRC.

3. In all cases where the piping connected to the vessel is more complex, you may have also bending moments on the nozzle (caused, for example, by the presence of elbows). I am not an expert of piping calculations, however, in my opinion, the anchor points on the vessels should show all the loads transferred from the piping to the nozzle wall, except the pressure thrust, because this simply means that the thrust of the internal pressure towards the inside of the vessel has been subtracted form the external loads. If you do so in cases like case 2c above (no external loads, but nozzle wall unloaded), you get an external load equal to the pressure thrust and directed towards the inside: if we had a precise method to take pressure stresses into account, you would probably find that the effect of this external load is to relieve stresses around the nozzle caused by pressure (and therefore by the nozzle reaction directed outside). Of course if pressure stresses are calculated (as it normally

happens) by the rough code methods based on area replacement, this superposition would lead to unreliable results.

The problem is rather complicated, further comments will be very much appreciated.

Best regards. 2 days ago Like Reply privately Flag as inappropriate

Follow Fumio Fumio Ando WRC 107 does not state the stresses due to pressure because of its scope for external loadings. If the general membrane stress existed at section in the nozzle due to pressure, it should be the thrust load by the sum of stresses to shell wall. Otherwise it is not necessary to consider the thrust load. As mentioned above, WRC 107 is out of scope, and just applied pressure thrust is not appropriate approach using WRC 107. It depends how the program implemented the pressure term in WRC 107 module. The better estimation is WRC 386/497 or FEM to analyze local stresses at discontinuity location between nozzle ane shell.