Design of pressure vessel with leg support 2

[GD_P]

Hello forum,

We have vertical vessel to be designed as per ASME Sec VIII Div 1. The vessel is supported on vertical unbraced legs, which are attached to bottom torispherical dish end. The vessel will be subjected to internal pressure and lateral force due to earthquake. Pressure calculations are not the issue, but we are stuck in the leg calculations and its implication i.e., local stresses on the dish end. So it will be very helpful, if you can share some references for manual calculations. EN 13445-3 does give reference for this in clause 16.11.2, but it is applicable only if the overturning moment is less i.e., net force does not become tensile in oneside leg. But in our case this is not applicable as overturning moment is high. So kindly share some other reference.
Regards

GD_P

 

[HTURKAK]

how did you calculate the period of vibration?.. I will suggest to look = Pressure Vessel Design Manual (by Dennis

R. Moss & Michael Basic ) .Good Luck..

 

[SnTMan]

...local stresses on the dish end.

I believe WRC-107 methods would be applicable, assuming your arrangement meets the assumptions, i.e. the loads are applied to the spherical portion of the head, shape of attachment, etc, etc.

Moss may have some guidance...

Regards,

Mike

The problem with sloppy work is that the supply FAR EXCEEDS the demand

 

[MJCronin]

WRC-107 methodologies will not help..

Moss does not give enough detail...

For seismic design of vertical vessel, consider changing the support design from individual legs to a robust skirt.

Legs can be used in many seismic designs, but a detailed evaluation of the legs and the vessel pads commonly require a robust and typical construction .... as well as protracted and involved calculation angst...

DO NOT TELL US THAT YOUR CLIENT INSISTS ON LEGS ....

Share with us all details of the vessel dimensions, wall thicknesses and bottom connections.

This topic has been discussed on these fora in the past

MJCronin
Sr. Process Engineer

 

[JStephen]

I can't think what difference it makes if the leg has tensile force. It'll need to be anchored, and if you evaluate forces in the head using WRC 107 or similar, it's just a negative number instead of positive.
You can attach legs to the shell instead of the head as another option. That doesn't solve problems but gives you different problems upon which to exercise your calculation angst.
As noted, skirt is far simpler and more reliable.

 

[MJCronin]

https://www.eng-tips.com/viewthread.cfm?qid=419343

MJCronin
Sr. Process Engineer

 

[r6155]

Legs are preferred in food industry, pharmaceutical……..etc.
Best design is centerline of leg coincident with centerline of shell plate.
Calculations are easy, see Pressure Vessel Design Handbook, by Bednar

Regards

 

[JStephen]

Calculations are only easy if you fit Bednar's assumptions, specifically, short squatty legs with minimal lateral force and neglect any radial growth, etc.

 

[oknow]

Be sure to read UG-54(a)...often overlooked Code requirement.

 

[GD_P]

Thank you all for your comments.
Kindly accept my apology for not mentioning complete details which are as follows,
Application: Food industry
Pressure: 6 Bar
Capacity: 250 ltr
Height to Dia ratio: 0.75
Dia: 730
Shell thk: 4 mm
Material: SS304
Length of legs: 550 mm

As indicated by MJCronin, WRC 107/537 both are applicable to "Local Stresses in Cylindrical & Sperical shells" and Dennis Moss & H. Bednar don't give details on evaluation of local stresses on dish heads due to external loads (due to legs).
As the application is food & the vessel is very small, it will be beneficial to use legs as indicated by r6155.
Method given in EN 13445-3 is applicable provided,

So what could be the option other that FEA software?



GD_P

 

[r6155]

Forget FEA.
3 legs NPS 1 sch 40 are sufficient, welded to knukcle of head (without reinforcing pad)
Normally no seismic calculation is required for this very small pressure vessel, but you can do it.

Regards

 

[Ehiman1]

When I design legs for vertical vessel I check the axial load on each leg by N/A +- M/W , if you are in the uplift condition you need to check bolts to axial tensile load, otherwise don't.
The bending moment depends on seismic load applyied on the C.O.G. of vessel.
In order to check stress on shell or head you can use WRC 107/297/537 or EN 13445-3.
Remember to calculate legs to instability load because they are compressed.
About period of vessel you can use T = C*H^0.75 if you want something to use without detail calculation. For C usualyy you use 0.085.

 

[GD_P]

Thank you r6155 & extragol1A for your comments.

Yes we can consider the leg size & ignore its stresses on dished head. But the reviewer may not agree with this judgement. As the legs are located on spherical portion of the dish end, I think it will produce the same results as that given in WRC107/537 for spheres with radius identical to dish crown.

GD_P

 

[r6155]

See AD-2000 Merkblatt S3/3 Vessel with domed ends on feet.

Regards

 

[MJCronin]

I still maintain that WRC 107 and it's daughters DO NOT APPLY to the analysis of vertical PV supported by legs that are attached to the vessel shell.

Most leg supports of vertical PVs are of this configuration...

https://www.researchgate.net/public..._of_Vertical_Pressure_Vessel_Using_ASME_Codes

https://www.eng-tips.com/viewthread.cfm?qid=421558

MJCronin
Sr. Process Engineer

 

[GD_P]

@MJCronin, i really forgot PV Handbook by Eugene Megyesy, which gives equations to evaluate the stresses on dish due to inclined legs projecting towards centre of crown rad. But this procedure is vague, since it talks about the wear plate dia but not about its thickness. Anyway my vessel don't require wear plate, but when required for other designs, it will sensible to add the wear plate thickness to shell thickness IMO.
@r6155, AD2000 is something new for me. I went through it, really good document. One thing i noticed it considers the legs as Fixed support at foundation (moment & disp. constraint). I was not able interpret the Table-1 i.e., stress table. What does the sign (- & +) in front of the stresses indicate and what is the meaning of internal & external (- & + sign column heading)?

Also I went through many handbooks/codes/guidelines for this subject of this post, I noticed even if the over turning moment is high and leg/s are under tension, no handbook/code/guideline consider it's tension for the evaluation of local stresses. Hence as indicated by JStephan in earlier post, I can even use EN13445-3 for evaluation of local stresses on dish due to max compressive force on dish.

As of now my problem is solved, but if someone can share their expertise to get more clarity on the topic, it will be really helpful for all young engineers. And I apologise, if I am stretching the topic too much.
Thank you all for your help.

GD_P

 

[r6155]

@ GD_P
1)Try to be simple when the problem is not clear for you- Several times we need to do calculations for a equivalent situation (rules of thumbs)

AD 2000: “A simplified limit load method is used as a dimensioning criterion for thin-walled spherical shells with local points where the loads is introduced. AD 2000 works with WRC.

i.e.: do calculations with “one leg” in center of head with filled PV (without design pressure). Buckling of leg is Ok.
I assumed your PV weight is 500 kg (filled with water). Stresses in the head are low

2) In AD 2000 Table 1 Nº 1 (+ sign) is the stress due to design internal pressure. Nº2 (- sign) is external stress due to external load (leg). Nº1 and Nº are superimposed.

3) Why do you insist with legs in the spherical section of torispherical head?

4) Why do you insist with torispherical head, why not semiellipsoidal?

Finally: Read (slowly) again AD 2000 S 3/3 and try to understand.

Regards

 

[CuMo]

Why is your client asking for seismic analysis?
Is this really necessary for such small vessel especially if mounted on the floor?