Hi, I have searched the forum but I wasn't able to find a clear answer to my question:
how do you check piping stresses due to tank loads under B31.3 code? I am not speaking about nozzle allowables.
Tank loads arise for soil settlement under tank weight, pressure head of contained fluid which makes the tank nozzle rotate (mainly near the bottom) and thermal displacements.
In my opinion all these loads are secondary loads, since they all cause nozzle displacements which stop at a given point.
But only thermal and pressure head loads can be considered truly cyclic.
I think that using Sh (hot allowable) is too much conservative, while using expansion allowable can be unsafe.
What are your thoughts about it?
Many thanks

You describe a strain-based load imposed on the nozzle. I agree that the Sh limit (used to ensure against collapse) is too severe. I've seen some limits to one-time settlement loads as high as 3Sh (nuclear code). B31.3 offers no such limit at this time. Other displacements (such as rotation associated with a full tank) are cyclic and would fall in with the expansion stress range evaluation.
That's the pipe side. I have no suggestion on evaluating the tank side except to point you for standards such as API 650 which provide maximum loads on these nozzles.
I've seen expansion joints out there too.

Staying on the pipe side, which is what bothers me at the moment, are you advicing to split loads (tank settlement and rotation associated with fluid head)? How do I have to configure load cases?
Many thanks

In my opinion,
I would evaluate three stress ranges (installed to operating): 1)no settlement and no bulge, 2)with settlement, 3)with settlement and bulge. If all three SEs are below SA you're OK. If not, then more work is required (for example, calculate equivalent cycles to adjust f for the highest stress range as defined in B31.3). I believe this would be conservative.
Again, that's my opinion with very little attention to details. You and others may change it...

Is SA the EXP case allowable??

Yes. SA is a B31.3 term.
And when I say all three, I mean all three - individually.

I agree with David statement. Nozzle displacement is secondary as per code (settlement included). I would say that the second case should be settlement at ambient temperature which can be more conservative than settlement at operating temperature (due to the first support near tank).

This is how I set my load cases:
W HGR
W+D1+T1+P1 HGR
WW+HP+H HYD
W+D1+T1+P1+H OPE
W+D1+T2+P1+H OPE
W+D1+P1+H SUS
W+P1+H SUS
L4-L6 EXP
L5-L6 EXP
L4-L7 EXP
L5-L7 EXP

T1 is 48°C, T2 is 80°C
D1 is the displacement vector which gets all the displacements related to the tank

So, W+D1+P1+H checked against Sh is the most conservative check I can do, I think. I am at 97%, but it verifies. So, though it is wrong pushing SUS cases to such percentage, if I understand what you said, the correct allowable (which I don't know exactly what it is) is higher than Sh. For this reason I consider my design safe.
Is it a good way of thinking?

W HGR
W+D1+T1+P1 HGR
WW+HP+H HYD
W+D1+T1+P1+H OPE
W+D2+T2+P1+H OPE
W+D1+D9+T1+P1+H OPE
W+D2+D9+T2+P1+H OPE
W+D9+P1+H OPE
W+P1+H SUS
L4-L9 EXP
L5-L9 EXP
L6-L9 EXP
L7-L9 EXP
L8-L9 EXP

In fact should be something like this.

D1 -Thermal + bulging
D2 -Thermal + bulging
D9 -Settlement only

L8 is still operating since contain displacements, it is usefull for evaluating the nozzle loads if you have a rigid support near the tank.

Regards,

Hi Dan

I just wanna reconfirm does HGR case have settlements value D9 or only operating D1

And
this load case should contain WNC case with operating and sustained case

Kindly reply

Thanks