Defeciences/Warnings in COMPRESS 1

[keens]

Hello All,

I am in the process of designing a Sour Water Stripper that is 6'-0'' I.D X 95'-0'' T/T, and all the nozzles have external loads. I have a few questions regarding the Warnings/Defeciences Report, if anyone has experienced the same issues.

1. For one of my Nozzles located in the bottom head:
Warning, WRC-107:Rigid insert assumption questionable
(tn<0.4375).
Q: Does anyone know what COMPRESS is referring to here?
And if so, how would you correct this?

2. For all of my Nozzles:
Warning, ASME B16.5/B16.47:External laods have not been
considered in the flange pressure rating.
Q: Does anyone know if the external loads should be
considered in the flange pressure rating? If so, is
there certain conditions when and when not to
consider external loads in the flange pressure
rating?

3. For my Lifting Lugs and Tailing Lug:
Warning, This vessel has eccentric loading which is out
of plane with the lift lugs. To prevent tilt during lift
and reduce lug weak axis bending, lugs should be at
104.77 degrees and 284.77 degrees.
Warning, Tail lug is out of plane with top lugs. Vessel
may tilt during lift and lugs may experience weak axis
bending. Tail lug should be at 104.77 degrees or
284.77 degrees.
Q: I have moved the lifting lugs to the angles the
warning gives me, and have moved the tailing lug
90 degrees to the lifting lugs. Is there anything
further I can do to correct this problem? Or does
this situation require calculations outside of
COMPRESS?

4. For one of my Nozzles:
Warning, External loads may cause the pad to fail
by buckling unless intermediate pad welds are
used - See Bednar 7.5 paragraph 3.
Q: What is the best way to correct this?

I thank you all for your continued help.....

 

[TomBarsh]

Hello Keens,

Please feel free to continue to send inquiries for technical support to support@codeware.com. You will get the best results that way. This forum is intended for more general discussions. But since you asked...

(It will be interesting to see if some of your COMPRESS-customer colleagues will chime in to this thread with their views!)

COMPRESS assumes the nozzle attachment to be a rigid (solid) attachment when performing the WRC-107 analysis. If the nozzle nominal thickness is less than one-half of the vessel nominal thickness COMPRESS will issue the warning &quot;Rigid insert assumption questionable&quot;. This warning means the assumption made by COMPRESS to perform the WRC-107 analysis is not satisfied. Consequently the calculation results may or may not be conservative.

Write to support@codeware.com and ask for copy of document &quot;wrc107-rigid attachment.doc&quot; [or see

http://www.codeware.com/support/faq/1227.html

] for a discussion of why this assumption must be made.


Tom Barsh
Codeware Technical Support

 

[TomBarsh]

And for your question # 2:

COMPRESS performs an analysis for stresses in the shell when external loads are specified on nozzles. The analysis is performed using WRC-107. Loads must be specified at the nozzle-shell junction following the conventions of WRC-107.

COMPRESS assumes that external loads are transferred through the nozzle neck to the shell and performs calculations for stresses in the nozzle neck due to internal pressure, bending, and axial load P. If stresses in the nozzle neck are excessive then a deficiency is produced.

Loads applied to the nozzle must also transfer through the flange. COMPRESS will optionally apply the loads to ASME/ANSI flanges by selecting the check box on the nozzle flange dialog. COMPRESS determines equivalent pressures for the &quot;P&quot; load (WRC-107 nomenclature) and for the resultant moment due to &quot;M1&quot; plus &quot;M2&quot; or &quot;MC&quot; plus &quot;ML&quot;; note that the positive sense of P is such that it will counteract internal pressure. For design purposes the flange MAWP rating is reduced by the liquid static head at the nozzle elevation plus the equivalent pressure due to external loads.

The equivalent pressure due to radial load (axial to flange) and bending moment is found using the method described in &quot;Pressure Vessel Design Manual&quot;, Second Edition, by Dennis R. Moss. Appendix G, page 273.

COMPRESS provides the option to apply external loads to the flange design. This permits the designer to decide whether or not to apply external loads to the flange design. Historically, this analysis has rarely been done for Section VIII Division 1 vessels unless the vessel has been in a critical service. However, it is relatively simple to add these calculations to computer software and so this feature was added to COMPRESS.

Some designers or vessel operators may question the validity of applying external loads to the flange design, claiming that vessels have successfully operated for years. It is fundamentally clear that any structural loads are passed through a connection and the nozzle flange is a structural connection. Determining the true loads applied to the shell through the flange is very difficult and the specified loads may not reflect the true loads.

Vessel owners may question &quot;Why haven't the flanges failed in service?&quot;. Some responses are: The true loads may be much less than the calculated loads. Similarly, ASME/ANSI flanges may have structural strength beyond that given by their pressure-temperature rating. At best, overstressed flanges may be prone to requiring frequent maintenance. At worst, an overstressed flange may fail during service.


Tom Barsh
Codeware Technical Support

 

[TomBarsh]

For your question # 4:

It is important to consider the load transfer through the nozzle reinforcing pad. The WRC-107 analysis considers the pad to be integral with the shell. If the pad is too flexible it may fail. The &quot;Pressure Vessel Design Handbook&quot; by Bednar recommends that the pad be no wider than 16 times the pad thickness (section 7.5, paragraph 3, page 209 of Second Edition) unless there are intermediate welds. COMPRESS reports a warning message if the pad exceeds 16 times its thickness.


Tom Barsh
Codeware Technical Support

 

[TomBarsh]

For your question # 3:

Please send send your COMPRESS file to me at tom@codeware.com. I will review and respond.


Tom Barsh
Codeware Technical Support