Hello to everybody. This is my first question on discussion forums, I hope I could find some help from you.
Is it appropiate to model restraints with gaps under seismic conditions?
Our client says that it is not acceptable because supports with GAP will generate shock effect (dynamic effect) not taken into account in a quasi static analysis.
Other question. I have used the following load cases. Load cases from 1 to 11 will be used for loads on restraints, and load cases from 34 to 43 for code stress check.Do you find them correct?
Thank you very much

L1 W+T1+P1 OPE Normal Operating Conditions
L2 W+T2+P2 OPE Operating at Design Conditions
L3 W+T3+P1 OPE Operating at Minimum/Maximum Ambient
L4 W+T1+P1+WIN1 OPE Operating + Wind 1
L5 W+T1+P1+WIN2 OPE Operating + Wind 2
L6 W+T1+P1+U1 OPE Operating + U1
L7 W+T1+P1-U1 OPE Operating + (-U1)
L8 W+T1+P1+U2 OPE Operating + U2
L9 W+T1+P1-U2 OPE Operating + (-U2)
L10 W+T1+P1+U3 OPE Operating + U3
L11 W+T1+P1-U3 OPE Operating + (-U3)
L12 W+P1 SUS Sustained with Operating Pressure
L13 W+P2 SUS Sustained with Design Pressure
L14 L1-L12 EXP Installation to Normal Operating Temp.
L15 L2-L13 EXP Installation to Design Temperature
L16 L3-L12 EXP Installation to Min./Max. Ambient
L17 L15-L16 EXP Design Temp. To Min./Max. Ambient
L18 L4-L1 OCC Wind 1
L19 L5-L1 OCC Wind 2
L20 L6-L1 OCC U1 (Seismic acceleration X direction)
L21 L7-L1 OCC -U1 (Seismic acceleration (-X) direction)
L22 L8-L1 OCC U2 (Seismic acceleration Y direction)
L23 L9-L1 OCC -U2 (Seismic acceleration (-Y) direction)
L24 L10-L1 OCC U3 (Seismic acceleration Z direction)
L25 L11-L1 OCC -U3 (Seismic acceleration (-Z) direction)
L26 L20+L22 OCC U1+U2
L27 L20+L23 OCC U1-U2
L28 L21+L22 OCC -U1+U2
L29 L21+L23 OCC -U1-U2
L30 L24+L22 OCC U3+U2
L31 L24+L23 OCC U3-U2
L32 L25+L22 OCC -U3+U2
L33 L25+L23 OCC -U3-U2
L34 L12+L18 OCC Sustained + Wind 1
L35 L12+L19 OCC Sustained + Wind 2
L36 L12+L26 OCC Sustained + (U1+U2
L37 L12+L27 OCC Sustained + (U1-U2)
L38 L12+L28 OCC Sustained + (-U1+U2)
L39 L12+L29 OCC Sustained + (-U1-U2)
L40 L12+L30 OCC Sustained + (U3+U2)
L41 L12+L31 OCC Sustained + (U3-U2)
L42 L12+L32 OCC Sustained + (-U3+U2)
L43 L12+L33 OCC Sustained + (-U3-U2)

All this many load cases to have mind, if you can found also this load case which the highest stress at restraint structure induced.
This is not easy to found.

Gaps under seismic conditions .... You might add a dynamic amplification factor to your quasi static results (DAF = 2 would be worst case). Unless it has changed in v5.1 (I am using v5.0), you cannot include gaps in a dynamic analysis anyway, since the process is a linear solution.

Load cases ..... good grief, I wish I had the time.

Hello MoverZ, thanks for your answer.
I have considered that possibility, but where should be applied the DAF? Only for the loads on restraints? What about the ocasional stresses?

The worst case woould be a DAF=2, but how could be calculated a more accurate value? If I try to consider a DAF=1.5 for example, it's possble to justify a lower value than DAF=2?
Thanks

Javgarloz,

There have been many posts on DAF values .... use the Search facility to find them.

The actual value of a DAF is a function of the natural frequency of the piping and the duration of the applied load.

Thanks again MoverZ,

I have found some post about DAF / DLF for relief valve discharge reaction forces, but nothing for this shock effect that could produce seismic loads in supports with gap.
Cuold you please give me some more information to find those post, investigate them and try to learn?
Thanks for your help...and patience

MoverZ,
I think DAF=2 is not the worst case. It could be 1,2,3,4 and more.
It's classic problem for body falling down on elastic beam. DAF=2.0 when H=0

DAF=1+sqr(1+2*H/D)

H - height from what the body is falling from
D - static deflection

If D=0 we have DAF=Infinity (when infinite rigid body falling at infinite rigid surface)

If we have Velocity:

DAF=1+V/sqr(g*D)

V - velocity of falling body

For horizontal impact:

DAF=V/sqr(g*D)

So to calculate DAF you have to know static deflection of support D. I think H=gap value

You can look the pictures here (It's in Russian, sorry):
http://mysopromat.ru/uchebnye_kursy/sopromat/udar/vertikalnyi_udar/
http://mysopromat.ru/uchebnye_kursy/sopromat/udar/vertikalnyi_udar_vsledstvie_ostanovki_dvizheniya/
http://mysopromat.ru/uchebnye_kursy/sopromat/udar/gorizontalnyi_udar/

Dear Javgarloz,

Why did you use W+T1+P1?
L6 W+T1+P1+U1 OPE Operating + U1
L7 W+T1+P1-U1 OPE Operating + (-U1)
L8 W+T1+P1+U2 OPE Operating + U2
L9 W+T1+P1-U2 OPE Operating + (-U2)
L10 W+T1+P1+U3 OPE Operating + U3
L11 W+T1+P1-U3 OPE Operating + (-U3)

Instead of W+T2+P2? I gues the worst comes from condition 2 = design
L6 W+T2+P2+U1 OPE Operating + U1
L7 W+T2+P2-U1 OPE Operating + (-U1)
L8 W+T2+P2+U2 OPE Operating + U2
L9 W+T2+P2-U2 OPE Operating + (-U2)
L10 W+T2+P2+U3 OPE Operating + U3
L11 W+T2+P2-U3 OPE Operating + (-U3)

but if not sure we can do such a load case L44 = L1+L2+L3 (OPE)using max combination methode.

For restrain check.
L1 ~ L11
L45 = L44 + Max Wind
L46 = L44 + earthquake.

For stress check, I think you miss SUS+ SRSS [U1+U2+U3] (OCC)

With so many load cases you may want to use CAESAR II's SignMax/SignMin load case combination to pick out max & min values.

Add two more load cases at the end of your list:
L44: L1,L2,...,L43 (SignMax)
L45: L1,L2,...,L43 (SignMin)

Replace the "..." with all the load cases you wish to include in the search.

Under Load Case Options, select the combination method shown above.

E.g., for SignMax, CAESAR II will list the largest value for each reported item over the group of load cases selected.

Regarding the DLF=2...

There may be some initial velocity present but using a max of 2 is documented. It will be published in the B31E document. It is also listed now in American Lifelines Alliance document on Seismic Design of piping systems. Google that and download.