AnEngineer;
Is this for a boiler or pressure vessel? If this is for a pressure vessel, what is the design pressure? This will help determine which particular section of the code is applicable.
If this is for ASME Section VIII, Div 1 pressure vessels, refer to the following sections for nozzle load requirements and strength calculations
UG-40 "Limits of Reinforcement",
UG-41 "Strength of Reinforcement" and especially
Figure UG-41.1 "Nozzle Attachment Weld Loads and Weld Strength Paths to be Considered".
ASME VIII Div1 can only provide allowable stresses and not allowable nozzle load. Nozzle "load" is due to the reaction from attached piping which will apply a force and moment at the connection. Many different combinations of forces and moments could attain a "maximum allowable value" for nozzle load.
For new vessels most client specifications nominate required nozzle loadings for design. Otherwise, one needs to analyse the piping connected to a nozzle to determine the loading applied, then use the methods in WRC107/WRC297 (or codes such as PD5500, or one of the handbooks around) to determine the resulting stresses from the particular force/moment combination to check whether they are less than allowable.
If ASME allowable stresses are exceeded, then for new vessels additional reinforcement can be added to reduce stresses, or for existing vessels the piping layout may need to be altered to reduce the loading on the nozzle.
in addition, you might also want to review Appendix L of ASME Section VIII, Div 1 (Specifically L-7.4). This appendix is useful to see how code calculations are applied, and checking adequacy of design assumptions.
I need to perform WRC107 calculations for nozzles on my dss horizontal vessel. The problem is our work is not big enough to buy a software like Compress... Do you know some alternative, like sub-contracting this analysis, Renting a hour-based software, buying a software that performs this analysis only... other ideas?
It is a common problem that the "allowable loads" on a nozzle are to be found. This question is often posed to vessel fabricators by the vessel owner or the contractor, etc, looking for whether the piping loads will create excessive stress in the vessel. It's a simple question.
Unfortunately, there is not a simple answer. And fabricators who have the misfortune to fall into the trap of providing this engineering work for free can find themselves vastly overrunning their engineering time.
The issue is that there isn't a single "allowable load" that may act on a nozzle. There are 3 possible forces acting on the nozzle and shell (2 shears and a radial force) and 3 possible moments (2 bending moments and a torsional moment); call these Fx, Fy, Fz, and Mx, My, and Mz. These will not in general resolve to a single set of forces and moments that are the "maximum".
In fact, there is an infinite number of combinations of these forces and moments that will all act to produce the maximum allowable stress in the shell (this is the true meaning of the "allowable load"...that which stresses the shell to the highest allowable condition).
Since there is an infinite number of load combinations fabricators who are asked to find the "maximum allowable load" can be frustrated when they find one of the sets that produce the maximum load but their customer asks "but what if...?" and they want to change the loads slightly, perhaps then saying that there is no VL or no Mt, etc. Then the fabricator has to perform the exercise all over again. I hope they get paid for each additional cycle.
Anyway, by placing some judgment on this one can come up with a methodology to establish some maximum loads. ie: you might define that the forces and moments will always be in some given proportion to each other or some other relationship, then run your WRC-107 program or spreadsheet based on these rules to find the particular set of maximum loads. Some further refinement can be performed too since some of the loads and moments do not combine directly to produce membrane or through-thickness bending stress.
You can find a simplified calculation for loads on nozzles that can fit very well with Excell spreadsheets(though it´s very tedious anyway) at this books :
- " Pressure vessels handbook". by H. Bednar
( Van Nostrand Reinhold )
- " Pressure Vesses Design Manual" by Dennis R. Moss
( Gulf Publishing Company )
A lot of this concern can be alleviated if you put the squeeze on the piping design engineer to keeps the piping from using your vessel as a anchor point. It was always our design policy to design for zero loads from the piping at a vessel nozzle. This was considered imperative due to our use of mostly SS process piping, even more so when we transitioned to larger heavier schedule piping. We used the same philosophy in the design of heavy wall jacketed pipe. Just the same we did use a lot on integral pads and forged components.
The old guidelines were if you do a complete nozzle design per Code, get the piping design right, (especially thermal and support), and watch the installation and you were on the way to viable system.
I say this realizing when dealing with very high stresses and temperatures it requires a much more rigorous design. If the numbers get high in normal Code calculations, its time to bring in one well versed in the art. The same goes for piping.
All the analyses in the world will not help if the installation is not supervised. I’ve witnessed the best of design calculations go out the window when the pipe fitters use a crane or 5 ton chain falls to drift a line to a flange on a column or make up the line from the column then use chains falls to line up pipe for closure.
Heyner
If you go to the Compress website there is list of companies and individuals that do design work.
Also in the page banner above there is ad for Paulin Research and if you contact them they can put you in touch with someone that can help you.
Also check the allowable loading on the flange joint [i.e., Kellogg method - turn moment & force in to equivalent pressures & combine with line pressure - result must be less than flange rated pressure @ temperature per B16.5 chart] Frequently this is the limiting area - not the nozzle-shell.
Sietemares is right - Bednar's book is the easiest/quickest way to calculate the nozzle-shell joint.
unclesyd,
You know as well as I do you cannot physically design piping to give zero load on a nozzle. The only way is not to bolt it up and leave a gap. Even with all the will in the world and setting the piping up on installation before bolt-up as the vessel/piping changes temperature there will be interaction at the flanged joint resulting in loads onto the vessel.
I do agree with your comments re the fact that poor installation can blow all the extensive analysis out of the water. I have witnessed this myself, however the designer assumes that the fabrication contractor is not a "cowboy" and uses acceptable fit-up techniques. If this were not the case then "design" could not be performed.
Perhaps zero sum at the nozzle was the wrong word I probably should have used minimize. The point I was trying to make is that the vessel is not an anchor, even though with current software it could be designed as such, so take the line loads elsewhere if possible. Our design philosophy was to design flexibility in a system and allow same to minimize undesirable loads on a nozzle. That philosophy was taken because we didn’t know and didn’t have the tools to do the calculations and didn’t want to have a ridged line and flexible vessel. Prior to the advent of computer design programs and FEA this was about the only way to accomplish eliminating nozzle and vessel failures. If you could look back at some of the nozzle and vessel failures you would find that most nozzles held and the vessel shell failed. I have seen the investigation reports on quite few instances where the root cause was an inadequate design of the connecting line for operating conditions. The connecting line/lines were too ridged with the solution being to add flexibility to the line. There are many reasons for this to occur other than the original design, the main reason being a change in operating conditions without readdressing the piping design. As the reliance on computer analysis has progressed I have seen a trend to push the loads closer to the allowables especially on nozzles and flanges.
As stated I realize that with larger vessels and piping with higher temperatures and pressures this become harder to accomplish and some sharing of the load is necessary.
Since I’ve seen more poorly designed piping systems than I have poorly designed vessels and nozzles I still don’t think a vessel nozzle should not be used as an anchor for a poorly designed piping system.
Don’t take the construction for granted as there are a lot of “hard core unemployables” working in the industry, both in the skill trades and supervision..
We generally issue a table of forces and moments based on nozzle size with pressure vessel and heat exchanger purchase orders. That way, the vendor knows up front what the expectations are and the stress engineers can do their work. In doing so, we take the nozzle connections as rigid. If we meet the table loads with a rigid nozzle, then all is good. If for some reason, we can't meet the table loads, we'll do a more rigorous analysis of the nozzle.
arto - we also check the flanges for excessive bending, but the "equivalent pressure" method is an extremely conservative approach. The Caesar II pipe stress program has a built in tool to analyze flange loadings and it includes a comparison to the equivalent pressure method. It's a rare thing to see a flange connection that satisfies equivalent pressure, but most of these do pass the other analysis methods.
Edward L. Klein
Pipe Stress Engineer
Houston, Texas
"All the world is a Spring"
All opinions expressed here are my own and not my company's.
