UNSYMMETRICAL SADDLES 1

[altoid3000]

I have a 15 foot long, 6 foot ID vessel. Initially the saddles were symmetrically placed at D/4 from right and left tangent lines. Latter due to nozzle interference issue, I moved right saddle in by 1 foot thus reducing saddle to saddle separation by 1 foot. The left saddle was not moved. Is this ok? Compress calculations are showing all the stresses way below the allowables. I want to know if this is done often? Is unsymmetrical saddle placement ok? Please comment and tell this newbie if the decision he made was a wise one?

 

[JoeTank]

There is nothing wrong with unsymmetric saddle placement; however, the load distribution is no longer 50/50, So, you must somehow fool the design program into increasing the saddle LOAD to account for this.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[codeeng]

Compress will automatically design for the saddle with the higher loading. I generally try to avoid unsymetrical saddles. If there are no obstructions on the other side you could still make it symetrical i.e. 2.5-10-2.5 ft. This is still within the guideline of 0.2L (3.0 ft) max from tanline to saddle.

 

[altoid3000]

well, the vessel has a boot which is closer to right saddle, thus actually moving the right saddle by 1 foot as I said earlier, I think has evened the load out between the two saddles. Would you agree? The boot is about 4700 lb out of 27627 lb operating weight (new) for the vessel. Does compress consider the boot location when it distributes load between the two saddles?

It appears that Zick method which is very conservative is still used in the industry. Compress also used Zick method. I keep hearing that FEM has proved that Zick method is too conservative, but is any one using FEM for saddle design? Are people using minimum saddle dimensions based on FEM analysis? I guess saddle dimensions suggested by FEM must be way lower than those given by Zick. Please direct me to some reading about this topic.

 

[altoid3000]

codeeng,

where is the guideline of 0.2L ?

 

[LSThill]

Team Member and altoid3000 (Mechanical)

RE: Reference check after CODEWARE COMPRESS analysis use SADDLE PRO by Paulin Reseach Group

http://www.paulin.com

Leonard Thill

http://www.thill.biz

 

[codeeng]

altoid3000
The 0.2L max guideline comes from our company standard. Its also mentioned in PV Design Manual by Moss. Compress should take into account the boot weight on the saddle. I use Zick for most designs and FEA when justified. Skimping on saddle design is generally not worth the extra effort.

 

[altoid3000]

I agree that using FEM for every saddle design is not the right way to go, but my point is why have the industry standards not changed based FEM studies. Why is Zick method still the only thing available when it comes to saddle placement decision?

 

[JoeTank]

Altoid,
Zick provided a solution to this problem many years ago and it takes a lot to move folks away from a solution that is proven to work. FEM just adds a lot of technical activity to an already solved problem. I find it difficult to justify the expense of an FEM solution that ends up with esentially the same result. Also, using advanced analytical technics usually ends up with reducing materials and details to the point that the robustness of the final design is gone... no margin for the practical problems of owning the vessel... corrosion, unanticpated settlement, piping loads, etc. Bottom line... put the owner's money in the vessel, not the analysis.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[waskillywabbit]

Much less of a problem to reroute the piping to a nozzle than make unsymmetrical saddles IMO.

Brian

 

[mechsurd]

FEM COSTLY, ZICK ANALYSIS CHEAP.EXTRA MEAT PUT ON BY ZICK ANALYSIS IS CHEAPER THAN FEM.