Horizontal Cylinder Under External Pressure based on AS1210 Clause 3.9

[Ray2020]

Hi Everyone,

I'm a recent graduate currently learning pressure vessel design. Im building a spreadsheet which determines the minimum thickness for a pressure vessel under external pressure. I hope someone is able to give me some feedback on what I have done until now. In this thread I'm approaching designers who are familiar with AS1210 as I'm not sure if ASME carries out the same procedure when calculating minimum thickness for external pressure.

AS1210 Clause 3.9 mentions an iterative procedure to determine the minimal thickness for cylinders under external pressure. Next I will share what I've done step by step.

1. Firstly I assume a thickness, lets say t = 3 mm.
2. Use t to determine number of circumferential buckling lobes n from equation 3.9.3(4) for the effective length Le under investigation. The mean diameter Dm will be (I.D. + t) and Z is the external pressure factor (Z=Pi*D/2L).
3. Use n, t, Dm, Z to determine Aa the circumferential strain.
4. Use Aa, E (Youngs Modulus), Dm and t to determine Pe from 3.9.3(5). Pe is the pressure to cause elastic buckling.
5. Use t, Y (specified minimum Yield stress), and Dm to determine Py from 3.9.3(6). Py is to cause plastic buckling.
6. Compare Pe and Py and then either use 3.9.3(7) or 3.9.3(8) depending on the comparison result to determine P.
7. If P value is less than P_external then t shall be increased.

I created a spreadsheet which keeps increasing t by increments of 0.1 mm.

8. After reaching the correct P, apply under tolerance to thickness and repeat calculation for that thickness only.
9. If P is still more than P_external, then this should be the minimum thickness. If P became less than P_external after applying under tolerance, then go to next increment and iterate for with/without under tolerance.

I hope my description is clear enough. Please let me know if what I'm doing is right.

Thanks
Ray

 

[BJI]

The methods in ASME VIII-1 are not the same as AS1210 for external pressure, although it is an acceptable alternative method. The procedure up to 7 appears to be correct but 8 and 9 are not required. You are calculating the minimum required thickness, then tolerances are added to this thickness to determine the required ordering thickness. For pipe the design thickness is the calculated minimum plus corrosion allowance, which must be less than your nominal pipe wall thickness less under-tolerance. Have a look through AS1210 3.4.2. Are you recalculating the minimum Aa with each iteration of t? It is also worth nothing that your ID for calculation purposes is the corroded ID, if applicable.

 

[Ray2020]

Hi BJI,
Thank you for the response.
Yes, Aa is recalculated for every t and ID is considered to be the corroded ID.
You are right! I just have to compare that calculated value to the selected corroded thickness less mill under tolerance.

I have read AS1210 3.4.2 many times and I find it hard to understand why they are lowering the ordering thickness in equations 3.4.2(2) and 3.4.2(3) by a factor of 0.94 or subtracting 0.3 respectively. I usually don't go this route when selecting a thickness for vessel under internal pressure. I just check if the 'randomly' selected thickness minus mill under tolerance minus corrosion allowance is greater or equal to the calculated thickness.

Do you have any idea why they are lowering the ordering thickness? Thank you

 

[BJI]

It does seem kind of counter-intuitive but 0.3mm is the permitted plate under-tolerance for Australian boiler plate (ref AS 1548). If you are using this plate material then you don't need to make any adjustment for under-tolerance. If you are using another plate material with a higher under-tolerance limit, then you effectively have to add on the difference. The 0.94 gives 6%, which at 5mm is 0.3, so the allowance is reduced proportionately as the plates get thinner.