API650 Wind overturning stability check and App F moment definitions
API650 Wind overturning stability check and App F moment definitions
(OP)
I know this has been covered before but not this specifically and with 11th ed. Addendum 1.
Assuming we are using the 5.2.1.k pressures 0.86kPa on the shell and 1.44kPa uplift. If D=diameter, r=radius and H=Height, Dp=design pressure and using Fig 5-27, the formulas from 5.11 do we use:
1/. Mw = 0.86*D*H*(H/2)+1.44*pi*r^2*r
2/. Mpi = Dp*pi*r^2*r
These formulas work out that most thin wall tanks need anchors, seems far more so than in earlier editions.
Also in appendix F. F.4.2 is the Wind moment refered to:
3/. M=0.86*D*H*(H/2)
Also the term DLs for a self supported umbrella type roof does not take any account for the roof is that correct? which gives a very low Pmax by F.4.2 again meaning I need anchors.
Assuming we are using the 5.2.1.k pressures 0.86kPa on the shell and 1.44kPa uplift. If D=diameter, r=radius and H=Height, Dp=design pressure and using Fig 5-27, the formulas from 5.11 do we use:
1/. Mw = 0.86*D*H*(H/2)+1.44*pi*r^2*r
2/. Mpi = Dp*pi*r^2*r
These formulas work out that most thin wall tanks need anchors, seems far more so than in earlier editions.
Also in appendix F. F.4.2 is the Wind moment refered to:
3/. M=0.86*D*H*(H/2)
Also the term DLs for a self supported umbrella type roof does not take any account for the roof is that correct? which gives a very low Pmax by F.4.2 again meaning I need anchors.





RE: API650 Wind overturning stability check and App F moment definitions
According to the current method, a piece of steel plate 5/8" thick flat on the ground will be blown straight up into the air by the wind. I find that rather hard to believe.
One thing that is overlooked is that on a rafter-supported roof, if the uplift pressure on the roof exceeds the roof plate weight, then all of the roof plate weight should be counted as resisting uplift, not just the portion supported by the shell.
I assume Mw in Appendix F should be the same Mw that you show being calculated. Or at least, that's the way the standard reads; that is not a logical way to do things, however.