.
Originally Posted by maizo
How to determinate the distance between the two saddles of an horizontal pressure vessel?
Regards!
Regards!
Originally Posted by Ali366
maizo
Stresses in shell due to saddle loads need to be specified. You can use Zick method or do an FEA. If these stresses are within allowable limits, there is no problem with the saddles.
Take a look at "Pressure Vessel Design Manual" by D.R. Moss. If you have got the 2003 edition of this book, section 3-10 describes the zick procedure and is quite easy to follow.
To give you an idea about distance between saddles, they should not be farther than 20% of "Tangent to Tangent Lenght" from the nearest tangent line. e.g. if TTL is 10m. then each saddle should not be more than 2m away from the nearest tangent line. There is a interesting chart in above-mentioned section which suggests the maximum distance of saddles from the nearest tangent line for a particular vessel with known thickness, length and radius. (Source of chart is from AWS).
But don't forget that above chart is just a suggestion and after selecting the location of saddles, stress analysis must be performed (zick, FEA or other methods) to evaluate the shell and heads under support loads.
Cheers,
Stresses in shell due to saddle loads need to be specified. You can use Zick method or do an FEA. If these stresses are within allowable limits, there is no problem with the saddles.
Take a look at "Pressure Vessel Design Manual" by D.R. Moss. If you have got the 2003 edition of this book, section 3-10 describes the zick procedure and is quite easy to follow.
To give you an idea about distance between saddles, they should not be farther than 20% of "Tangent to Tangent Lenght" from the nearest tangent line. e.g. if TTL is 10m. then each saddle should not be more than 2m away from the nearest tangent line. There is a interesting chart in above-mentioned section which suggests the maximum distance of saddles from the nearest tangent line for a particular vessel with known thickness, length and radius. (Source of chart is from AWS).
But don't forget that above chart is just a suggestion and after selecting the location of saddles, stress analysis must be performed (zick, FEA or other methods) to evaluate the shell and heads under support loads.
Cheers,
Originally Posted by tony_black
Yes, Ali366 is correct - however, to describe the issue in a more "practical" sense - please see the attached excerpt paragraph 13.9.1 from "Coulson and Richardson's Vol 6"
In short - you want to start with an analysis where the bending moments across the saddles are equal to the bending moment in the mid-span. And (from the design of a uniformly, loaded, simply-supported beam) this occurs when the supports are placed 21% of the span in from the ends. Then, most oftenly, to avoid the need for stiffening rings at the saddle because the shell will want to "ovalize" at the supports (if far removed from stiffening effects) - many designs push the saddle supports farther outwards towards the heads to gain stiffening from them (and avoid extra stiffening all together).
In short - you want to start with an analysis where the bending moments across the saddles are equal to the bending moment in the mid-span. And (from the design of a uniformly, loaded, simply-supported beam) this occurs when the supports are placed 21% of the span in from the ends. Then, most oftenly, to avoid the need for stiffening rings at the saddle because the shell will want to "ovalize" at the supports (if far removed from stiffening effects) - many designs push the saddle supports farther outwards towards the heads to gain stiffening from them (and avoid extra stiffening all together).
Recent Threads
Random Threads
vBulletin Message