Tank Design - Vacuum Loads

[sme75]

When checking vessels for vacuum loads I have used the ASME charts and formulas to determine the coefficients A and B and then calculate the allowable load using:

Pa = 4B/[3(Do/t)]

or

Pa = 2AE/[3(Do/t)]

where applicable. Can anyone point me to a good reference to use when the Do/t ratio is very large? Fig. UGO-28.0 has a maximum contour at Do/t = 1000. What have you used when the calculated value is greater than 1000?

The tanks I am looking at are not subjected to large pressures, however even a small vacuum adds up when the diameter is large. Any thoughts or tips would be appreciated. Thanks.

 

[Tankman650]

Use API 620.

 

[LSThill]

Hi sme75 (Structural)

Need more information: FIG. UG-28.1 Location of Stiffening rings.

UG-28 STEP 6. (f) do you need a code stamp?

What is the Diamater and Hight? What type of Heads?

Leonard@thill.biz

http://www.thill.biz

 

[sme75]

No stamp is required.

The tank is 20' diameter with a 44' straight side.

Flat bottom with a cone top.

The shell is 3/16".

I forgot about API 620, and think that will work, but any other suggestions are appreciated.

 

[sme75]

Tankman650,

In section 3.5.4.1 of API 620 it states that:

"These rules do not purport to apply when the circumferential stress on a cylindrical wall is compressive(as in a cylinder acted upon by external pressure)."

What section of API 620 were you refering to with your response?

Section 3.5.4.4 is the closest section I have run across with the meridional and latitudinal unit forces being compressive and of unequal magnitude. The worst load case would be if the tank was empty with the vacuum load applied. In that case the the shell would be in compression from the roof and shell weight in one direction and the vacuum load in the other direction. However, does this section apply in this case or is it only for tanks with walls of double curvature as section 3.5.4.1 suggests?

 

[Tankman650]

I think your cone roof will control.

If your stiffener at the cone cylinder is large enough you can use L=44' (528&quot

For the cylinder you can determine A from the following formulaASME)

A= (1.3*(t/D)^1.5)/(L/D)

 

[LSThill]

sme75 (Structural) and Tankman650 (Structural) all of the above:

Reference source:
STRUCTURAL ENGINEERING HANDBOOK
Edited by
EDWIN H. GAYLORD, JR. and CHARLES N. GAYLORD
McGraw-Hill Book Company

SECTION 23 STEEL TANKS:
John N. Pirom* and Robert S. Wozniak**
*Chief Structural Engineer and **Design Engineer
Chicago Bridge & Iron Company
Oak Brook, Illinois

Page 23-1 - 23-25

Mr. sme75 (Structural) for your tank design application the above reference is from a project of 6/3/85.

Storage Bin with a Flat Bottom and Cone Roof.

If you have STADD 2003, it would take 10 minits to the Design.


Leonard@thill.biz

http://www.thill.biz

 

[jte]

sme75-

You could consider using ASME Code Case 2286, "Alternative Rules for Determining Allowable External Pressure and Compressive Stress for Cyulinders, Cones, Spheres, and Formed Heads Section VIII, Divisions 1 and 2." Your tank has a Do/t ratio of about 1300 which would be within the range of the Code Case. The Code case is applicable for Do/t ratios not exceeding 2000. In particular, it sounds as if section 5.4, "For Combination of Uniform Axial Compression, Axial Compression Due to Bending Moment, M, and Shear, in the Presence of Hoop Compression" would be useful for you.

jt

 

[sme75]

Is ASME Code Case 2286 contained within the ASME pressure vessel code? If not where can I get a copy of it?

 

[TD2K]

http://cstools.asme.org/CodeCases.cfm

Enter code case 2286 and voila!

 

[jte]

TD2K-

The link you provided gave information regarding the Code Case, but as far as I could tell, not the 18 pages of the Code Case itself. Did I miss something?

sme75-

The Code Cases are supplemental to Section VIII. If you have a subscription to Section VIII I believe you should also be getting copies of the Code Cases. I pulled my copy from our electronic subscription to IHS at

http://www.ihserc.com.

jt

 

[TD2K]

Ish, sorry about that jte. I thought from browsing the site that the code cases were available to review online. And they may be, I may just not be finding them.

But yes, the link I gave you does not bring up the total code case. Not sure of the value of the information they do provide you ;-)

 

[tonycola]

See Maan H. Jawad. James R. Farr,"Structural Analysis and Design of Process Equipment",2nd ed.,John Wiley & Sons, page 468. The formual given in this reference is the basis for the formula provided by API 650 for placement of stiffners on the tank shell. Therefore this is the formula that should be used for API 650 tanks.

Pcr=(2.42*E/(1-u^2)^0.75)*((t/D)^0.25/(H/D-0.45*(t/D)^0.5))

where E= Young's Modulus, u= poisson's ratio, H= length of shell section (or distance between stiffners), t= shell thickness.
If P=25.6 psf and factor of safety = 1.5, then formula becomes
H=6*(100*t)*(100*t/D)^1.5
where H is feet, t is inches, and D is feet.
This formula is the exact formula used in API650.

Have fun!

 

[sme75]

tonycola,

I looked up the reference you mentioned. Have you tried to follow through the development of the API 650 equation? I agree with them until they say that P = 25.6 psf and FS = 1.5 is substituted to obtain equation 13.19. It seems like they left our a few steps or am I just missing several unit conversions? Any thoughts?

 

[tonycola]

sme75,

I have been using this formula for years and I am almost certain that I checked it out before I first used it (to calculate vacuum rating of tanks). But, since you ask the question, I went through the derivation again!
If you are very careful with converting inches to feet and psi to psf you will get the authors results, except instead of getting 100 times t under the radical you should get 98.4. This is close enough!

Have fun