rafter supported cone roof on ASNSY WORK BENCH

[inammanj123]

Dear ALL,

I am new to ansys workbench. I am working on a project in which i am checking rafter supported cone roof. Kindly tell me how i should do the modelling.As per my understanding, one way is to do the modelling, by drawing the UPN rafter using line body.One end of this rafter i want to connect with a cleat, as shown in the attached sketch. The other end will also be bolted to a cleat welded to circular drum. MY question is if I drawing the cleat as surface the how I can connect the cleat with rafter. For starter i am assuming that i will be welding the cleat with rafter. i have been told that that i cannot join surface with line body, so i modeled the rafter as surfaces, and joined surfaces of rafter and cleat as one part. Kindly advice

Thanks,
Inam

 

[IFRs]

Your tools are more complicated than needed for a simple structure as this. If you are new to modeling you will take way more time dealing with the software than checking with hand calcs using basic moment and shear diagrams and relatively simple steel construction code. You don't need Michaelangelo to paint your kitchen wall white.

 

[inammanj123]

well said, actually i am sharpening my grip oven workbench. Anyways, i have performed hand calculation for raftter size against UDL, But i still grappling, to some concrete formulation to check the support bracket on shell side. I want to know that clip size, thickness and the shell thickness is suficient for the UDL.

 

[IFRs]

For modeling the cleat - perhaps simply applying the external loads to the cleat is enough.

 

[carthago]

I came across this problem before. And could not find something serious in the literature !!
ANSYS was conducted upon client request as the tank is an existing one, and it has to be re-assessed...

I used ANSYS workbench. Building an asymmetric model. I focused on Minimum Principal stress as this will show the most negative ( compression) stress.

See attached files, for example: maximum displacement (about 6 mm outward) and Minimum principal stress ( about 40 N/mm2) which is higher than allowable for compression ( about 6 N/mm2). The tank is with D= 47 meters , and H= 17 meters

My understanding/ worry is: due the distance between bolts and the tank wall ..then we have an eccentricity,
This means an additional moment onto the tank wall.
The ugly part created by this moment is the compressive stress in the tank wall, which might cause a local buckling if too high !!
As known, for instance from API 650 or API 620 the allowable for compressive stress in a tank wall is very low. Much lower than the standard allowable (Sd).
There is also a similar limitation in EEMUA 159..

The finite element is a big tool , not always needed and could be quite cumbersome.
A good design starts by making this eccentricity as small as possible, and if needed make the upper shell a bit thicker ( at least locally), the later will also increase the allowable. Also use a stiffener under the bracket is quite helpful.


The point is how to model ? meshing ? boundaries ? loads ? calculated in corroded conditions or not ) ?
Also:
- taking into account the non linearity ( as the shell will move outward, which will redistribute the loads, and change the stresses.
- Need of sensitive studies: i.e. see the influence of some governing parameters ( wall thickness, eccentricity, stiffener size )


I wonder If some of you guys have a better idea on how to model this , and if one should really care about this eccentricity !!




take care and thanks

 

[carthago]

....ooops the other attachments

take care and thanks