If the slender gusset plate in column bottom piping from bottom head kept for shipping protection is kept intact, what could be the repercussion on structural integrity of column and the piping - in cases where column is in cold service and in hot service respectively ?

My opinion is that column bottom is within skirt and in steady state, the slender gusset plate will attain same temperature of piping and no appreciable secondary stress will be developed in the gusset plate.

For cold service, the small temperature differntial between piping and gusset may put gusset in copression and gusset plate may buckle, but there will be not much load on pipe and column bottom head.

For hot service, gusset plate will be in tension.

How the load in gusset plate can be checked I am wondering.

It needs to be simulated by an equivalent pipe with stiffer direction moment of inertia of gusset plate for conservative
approach and the loads can be checked on pipe and head using nozzlepro with plate structure for local effects.

Have any of our forum members similar experiences ?

reg,
sam

Measure the maximum distances along the two welds from the nozzle / wall interfaces.

Calculate change in length of the nozzle and shell from this intersection point via dL=L•alpha•dT.

Apply this displacement in shear along the ends of the gusset plate. This is your strain.

Stress is equal to strain times shear modulus, which is young's modulus E divided by 2•(1+poisson ratio).

Stress is equal to force divided by area, which would be the gusset/weld interfaces.

This can be backed away from if you adjust the dT from the above equation as the maximum temperature differential between gusset plate and base metal.

Edit to add: This calculates maximum force on the gusset weld. If you measure the closest point of the gusset weld to the nozzle/wall junction you will instead get minimum force, and at the mid-point, average force.

Tension and compression will be similarly calculated except it'll be the SRSS of the two displacements = to PL/AE where L is length of gusset, mid-point to mid-point, A is normal cross sectional area, E is Young's modulus, and P is force.

Thanks; here, you have overlooked the point that the gusset is inside skirt and welded to the bottom head and bottom pipe after elbow - attaining same temperature of pipe and equipment in steady state.

In that case as stated above, there is no strain - only thermal expansion of gusset plate - no secondary load coming due to thermal expansion as there exists no thermal stain.


In your formulation, linear strain gets multiplied with shear modulus - not fair! - sorry, we were using this term frequently till this era of interesting time when people object to it. In your subsequent post, you have rectified it with Young's Modulus, which is right.

As gusset is a plate having length and width of similar order with thickness small enough, it can't be considered in simplified manner in Caesar-II, asit looks to me. I consider an equivalent pipe of same area and the higher side moment of inertia of the plate section - 1/12 tb^3, to be on safer side. But, without cirulation of air inside skirt and conduction through weld, I do not find any reason of different steady state temperature in gusset plate than the pipe or column bottom.

Thus, for me through the above justification, the concern for the gusset plate in undermining the pressure integrity of system does not look to be sound enough! If anyone looks worried about gusset temperature, let him/her put the gusset within insulation of pipe and bottom head of column envelope!

This is what I interpreted what you described. I suppose instead you're saying they connected a gusset plate from the vessel straight down to the pipe? If this is the case, then you could emulate in CAESAR by adding a Y support at the pipe connection with a stiffness value equal to P/dL as calculated from above.