Test Plate (Blind Flange) thickness - Ductile Iron - EN 12516-2/4 & ASME VIII Div.1

[Y-Shevy]

Hello,

We have a requirement to test a number of large nozzle check valves (60"-300# Series B) shell and seat test (1.5xDP & 1.1xDP respectively). We need a pair of blind flanges to seal the unit for testing, we have a pair of plates (9in thick) made of Ductile Iron A395 60-40-18.

I have done a plate thickness calc based on ASME VIII Div.1 UG-34:2015 'Unstayed Flat Heads and Covers' (fig. UG-34-j). Using UCD-23 & UG-24 for allowable stress I have a thickness (much thicker than the plates we already have).

I have done a second calc based on EN 12516 parts 2 & 4. Using the equation from 12516-2, 8.2.1 and the allowable stresses for both scenarios; 'design condition' and 'test conditions' from 12516-4, table 17 for EN-GJS-400-18 I have determined what the plate thickness would be based on this standard.

I have a significant difference, the EN standard for a 'design condition' is over 25% less than the ASME calc, does this sound right? I know the EN standard is one of the least conservative standards for valve wall thicknesses.

Additionally, do any of you know why there is such a difference in the safety factors between 'design' and 'test' conditions (EN 12516-4:2014+A1:2018) Table 17? Is this based on a controlled pressure (no surges), no corrosion, test conditions, etc?

Thanks in advance.

 

[JGard1985]

Why are you using ductle iron flanges/plates? Have you looked into B16.5 blind flange specification and considered steel vs ductile iron. A prequalified 60", 300lb flange could be 6.5" thick

http://www.piping-designer.com/inde...lange-blind-ansi-class-300-b16-47-series-a-in

Jeff
Pipe Stress Analysis
Finite Element Analysis

http://www.xceed-eng.com

 

[Y-Shevy]

I have noticed the standard thickness for this size and class in B16.5, and it would always be my first choise to use a steel plate. However, as I mention, we already have the ductile iron plates from a previous job a from a long long time ago. To buy new plates (when it's not absolutely necessary) would be a signifcant cost.

I have since adopted the EN 12516 standard which has three advantages; firstly it's less conservative (reduced FOS), secondly, part 4 accomodates a variety of ductile iron grades and thirdly it considers 'test condtions' which has a reduced FOS.

As we only use blind flanges for testing, we may consider using this standard in the future for the larger valves.

Thanks for your reply.

 

[gerhardl]

In my opinion no plate thickness will corrupt the test validity, as long as there is no leakage connected (given by) the blindplates. Why a required standard thickness? Be sure to quote a thickness that will hold under all circumstances?

If you should set a thickness and give a material thickness 'off the cuff' why not use flange spesifications ?

(Just tossing around some ideas)