Calc relief valve thrust per API-RP520 para 2.4.1. The
procedure is presented both for open-discharge and
closed-disch. system. API is silent regarding about
the dynamic load factor so does B31.3. For this, refer
to B31.1 (i guess its 1.2, check it out). Multiplied
the load factor by the calculated thrust load.
I agree with Christopher regarding about the external
load as far as the tightness of the flange is concern.
The B31.3 if im not mistaken refers you Appendix 2 for
flange design. However Appendix 2 only consider
internal pressure to det. the flange adequacy.
The method the Chris has described (external forces
and momments)is similar to Section III Subsection NB.
Another method is Appendix IV of B31 Mechanical Design
Document
For the pipe disch. or tailpipe support, the Crosby
suggestion will fairly give good result.
- Christopher Wright <chrisw@skypoint.com> wrote: >
>On my project , I have the problem on relief valve
> system. Now the bolt on
> >the upstream and downstream all is loosen due to
> vibration. The set pressure
> >is 68.9 barg and the size is 6" x 8".
> >How to calculate the force reaction when the relief
> valve popping ?
> It works like a rocket. The force on the relief
> valve equals the change
> in momentum of the fluid going out. Check your fluid
> mechanics textbook.
>
> >How to design a good support to protect the loosing
> bolt cause by vibration
> The relief valve flange has been designed for the
> necessary loads by the
> manufacturer. The bolt probably came loose because
> it wasn't torqued
> properly. If the flange bolts don't need to be
> removed periodically, you
> might want to stake the threads, otherwise use a
> thread locking compound
> (Loctite) when you assemble it or one of the many
> locking fasteners.
>
> >How to check the strength of flange on bolt due to
> reaction force on relief
> >valve?
> Figure the thrust load then and calculate the
> pressure equal to the
> thrust divided by the area of the flange bore. Add
> this equivalent
> pressure to the flange design pressure and use the
> result in the ASME
> Code flange calculations. This is conservative, but
> not enough to hurt.
>
> Christopher Wright P.E. |"They couldn't hit an
> elephant from
> chrisw@skypoint.com | this distance" (last
> words of Gen.
> ___________________________| John Sedgwick,
> Spotsylvania 1864)
> http://www.skypoint.com/~chrisw
>
>
[Non-text portions of this message have been removed]
Received on Wed Aug 22 10:44:00 2001