9 replies to this topic

[lafondejs]

Dear all,

Is there any special method to design pressure vessels with very low design pressure & quite large in size. For an example, design pressure is about 0.01 bar.g, & diameter is 3.5m and tangent to tangent length is about 8m. I can find thicknesses as according to ASME Sec VIII Div 1. But I feel that thicknesses are very small when consider the size of the vessel, specially when the dead weight & operating weights are considered. Is there any minimum thickness for a given diameter (and Length) ??.

Thanks.

[ankur2061]

lafondejs,

The design pressure suggested by you is not correct in case of vessels handling liquids. In case of vessel handling liquids, the design pressure to be considered is the pressure in the vapor space (in your case 0.01 barg) plus the static head considering the vessel full of water (in case the actual liquid is denser than water then use correction for specific gravity of the liquid under consideration).

This is the way liquid filled vessels are specified for design pressure. It is interesting to note that the bottom most plate of the vessel is the one which will see the highest static pressure i.e combination of vapor pressure and liquid static head whereas the topmost plate will not be seeing the static pressure due to liquid head. However, for design different plate thicknesses are normally not used. The calculated plate thickness for the highest static pressure the vessel can see is generally used for the design and construction of the entire vessel.

The above discussion relates only to static pressure and precludes any pressure built-up due to any kind of chemical reactions occuring inside the vessel. If any reactions are expected inside the vessel which could lead to increase in pressure inside the vessel this needs to be accounted for separately and added to the static pressure to arrive at the design pressure of the vessel.

Hope this helps.

Regards,
Ankur

[Art Montemayor]

lafondejs,

Allow me to address your queries:

What you describe is NOT a pressure vessel. ASME describes a pressure vessel as one that has a pressure greater than 15 psig, so you can't apply the ASME code. Nevertheless, you are concerning yourself with the wrong parameter. You should not be concerned about equations, standards, or codes related to "pressure vessels". At the pressure level you are debating, you should concern yourself with the PRACTICAL, COMMON SENSE problem of welding the shell of such a vessel.

As an ex-boilermaker and welder, I have the advantage of being trained to always consider the feasibility of welding a vessel. I know, from experience, that it is next to impossible to weld a vessel made of plate that is 3/16" thick - without it developing a warped surface from the concentrated weld heat. My recommendation is to consider making your steel tank or vessel out of 1/4" to 3/8" thick, minimum. That will allow you to weld it.

If your vessel is a horizontal, 8 meter-long vessel then I would not consider any plate below 3/8" thick for the shell. It is that simple and direct.

[kkala]

"There will be a minimum wall thickness required to ensure that any vessel is sufficiently rigid to withstand its own weight, and any incidental loads. As a general guide the wall thickness of any vessel should not be less than the values given below; the values include a corrosion allowance of 2 mm:"
Table of minimum thickness versus vessel diameter is placed in the attached "mwthick. xls".
Source: Coulson and Richardson's "Chemical Engineering" Vol 6, Chapter 13 - Mechanical design of Process Equipment.
Besides the minimum thickness required for welding, above has to be also considered . A drum of 3.5 m diameter would require a minimum thickness of 12 mm. Length does not seem to have an influence, yet mentioned thicknesses are approximate. Mechanical design could give more precise values of minimum wall thickness, if necessary.
By the way, design pressure refers to upper internal space of the vessel; fabricator takes any additional hydrostatic pressure into account, to calculate max possible pressure at any point in the vessel. This is also applicale to atmospheric storage tanks, http://www.cheresources.com/invision/topic/15273-storage-tank-design-pressure-vs-hydraulic-test-pressure. Thus hydrostatic pressure does not participate in the design pressure written in process data sheets, at least according to what we apply here (and elsewhere).
I would be indebted to know other relevant established practices (if any) internationally.

Attached Files

•  mwthick.xls   17KB   616 downloads

Edited by kkala, 20 August 2012 - 04:38 PM.

[Art Montemayor]

With all due respect to Messrs. Coulson and Richardson, but I seriously doubt that either of them ever “burned a rod” (welded a steel vessel). I respect their ability to publish the data shown on the submitted table as “Minimum practical wall thickness of vessels” and the fact that this data has indeed been reported by them. But …….

However, the data is, I believe, shown to be IMPRACTICAL rather than practical – which is what we engineers claim to be (as opposed to scientists and academicians). For example, I personally have built storage tanks in excess of 3 meters (10 feet) in diameter and with only 3/8” wall plate instead of the given ½” minimum thickness. I don’t know what is the background information or basis for the table submitted, but it seems to be a “general” set of data and not specific at all. In fact, is the table made for pressure vessels? Is it for storage tanks (according to API standards)? I don’t think we know.

Additionally Length (or height) of a cylindrical “can” DOES MAKE A DIFFERENCE. The static pressure due to liquid head (as Ankur has mentioned) is what will ultimately determine, in part, the thickness of the bottom row of cylindrical welded plates – or “can”. Wind loading and roof weight will also be factors. That is why I prodded the OP to state whether the vessel is vertical or horizontal. It makes a difference and I won’t generalize.

I base my opinion on my welding and fabrication experience and I would urge and recommend anyone else to always refer to a local, experienced vessel fabricator for this type of practical information and not rely on an engineering textbook.

When calculating the required wall thickness of a storage vessel according to the hoop stress equation and arriving at a ridiculously small value (as the OP reports) it is always more practical to rely on experienced empirical estimates rather than academic ones. Some calculated vessel wall thicknesses simply cannot be electric-arc welded because the steel plate bends and warps out of geometric shape due to the amount of weld heat that can’t be removed fast enough together with the steel expansion. I have personally had to apply “strong-backs” on storage tanks because welding on the shell for maintenance has produced what we call “a puckered effect” – localized deformation of the shell resulting in inundations and “bumps”. It not only looks horrible, it destroys the geometric inherent strength of the tank and neutralizes the calculations of the hoop stress.

That’s why I strongly recommend the opinion of a local, experienced vessel fabricator

[lafondejs]

Dear Ankur, Art & kkala,

Thanks for your replies. I'm sorry I forgot to mention that this vessel is a horizontal vessel.

As per Mr. Art's comments, he has strongly recommended the opinion of a experienced vessel fabricator. Of cause I do agree with him. But I'm wondering why don't we have a well known method for these types of vessel designing?. I'm raising this question because, I think these types of storage vessels which have the same geometric shape of normal/ conventional pressure vessels are very commonly used in process and other industries.

Thanks.

Edited by lafondejs, 20 August 2012 - 11:11 PM.

[kkala]

1. Data of minimum wall thickness by Coulson and Richardson (C-R) does not deserve the characterization of impractical in my opinion, even though it is not complete. Structural design has to consider both, welding possibility and vessel ability to withstand its own weight, which become significant as vessel design pressure approaches 0 Barg from positive values.

2. The graph in “comparison” sheet of "mwthickness.xls" (attached) shows all data by C-R’s book (blue lines) versus vessel diameter, as well as minimum wall thickness necessary for welding by Art Montemayor (red lines). The two sets of data comply satisfactorily to each other, yet C-R’s data has the advantage of indicating increased thickness with increase of vessel diameter. Besides, mentoned thickness of 3/8 inche for a 10 ft diameter vessel is close to minimum range by C-R.

3. Usefulness of C-R’s data would be enhanced, if there was distinction between horizontal and vertical vessels, as well as thickness dependence on vessel length. I will sent a note to book Publisher, hoping or more complete data in future editions.

If local hydrostatic pressure requires extra wall thickness, the latter could be calculated (not concerning minimum thickness of this thread). Same for other loads due to wind, earthquake, etc.
4. Since Chapter 13 of the book deals with pressure vessels, presented data by C-R is understood to concern drums, irrespectively of their construction code. Additional clarification in future editions would be useful.
5. Vessel fabricators are usually reluctant to give vessel data, unless a bid is requested or expected (clearly including vessel details). For preliminary design and cost estimate of alternatives, a quick vessel wall thickness is needed, including consideration of minimum wall thickness. The latter could be defined by mechanical engineers (considering statics+strength of materials+codes), but chemical engineers are not familiar with this design, so they have to use generic data like C-R's.

Nevertheless presentation of existing low design vessel data by some member would be quite helpful for comparison.

5a. According to C-R “the walls of pressure vessels are usually relatively thin compared with the other dimensions and can fail by buckling under compressive loads”. Wikepedia indicates buckling is a major design concern for vertical or even horizontal vessels, affecting their wall thickness (not considering stiffening reinforcements, assumed uncommon for vessels). For instance see “buckling of thin cylindrical shells subject to axial loads”, “ – torsional buckling”, etc. But we are not familiar with "buckling" calculations.

5b. "As per section 11.1.3 of API 574: "In low-pressure and low-temperature applications, the required pipe thicknesses determined by the Barlow formula can be so small that the pipe would have insufficient structural strength. For this reason, an absolute minimum thickness to prevent sag, buckling, and collapse at supports should be determined by the user for each size of pipe."
Table 6 (attached ) of the same code provides some data for Carbon and Low-alloy Steel Pipe at less than 205 degree centigrade condition".http://www.cheresour...kness-of-pipe/ .

Probably there is something similar in vessel codes, we do not yet know where.

5c. "DesPress" sheet of attached "mwthickness.xls" indicates that minimum wall thickness by C-R corresponds to a drum design pressure of more or less 7 Barg in the diameter range of C-R's data. I had vaguely heard of ~5 barg in the past. Of course a vessel of e.g. 2 Barg design pressure cannot be pressurised to e.g. 4 Barg, even if its thickness would be suitable for the latter.

 

Attached Files

•  mwthicness.xls   43.5KB   283 downloads

[kkala]

Following additional notes may be useful as supplement to above post No 7 (by kkala).

 

1. In post No 6 lafondejs asks for a well known method for actually atmospheric drums. Perry’s Chemical Engineers’ Handbook /Transport and Storage of Fluids /Storage and Process Vessels / Storage of liquids (p. 6-85 and on in 5^th edition) indicates following.

a. API RP 650 deals with cylindrical (apparently vertical) atmospheric tanks.
b. American Water Works Association has issued the AWWA D100-67 standard for steel tanks-standpipes, reservoirs, elevated tanks. Intended for water, but can be used for storage of other liquids.
c. Underwriters Laboratories Inc has issued the standard UL 142 for horizontal and vertical tanks to store flammable liquids above ground (UL 58 is for underground).
Probably (β) or (γ) can fit the case, to be checked for verification.

 

2. Following post No 7, a drum wall thickness will converge to minimum wall thickness (being discussed here) as its design pressure approaches 0 Barg ( well, 1.03 Barg for ASME, 0.5 Barg for other codes (*), etc) from higher values. And this minimum wall thickness is expected to be almost same for identical configuration vessels of any design pressure lower than P. The upper limit P is anticipated to have a value lower than 7 Barg, since the drum bears combined loading due to internal pressure and its own weight (*).

Actual low design pressure drum data for further “investigation” would be welcomed, and / or advice on the point.

For want of a specific code, wall thickness of an atmospheric drum could be considered (in basic preliminary design) as the minimum wall thickness of a pressure vessel as above. Corrosion allowance and any (hydro) static pressure have to be also considered.

 

3. In post No 7 and here design pressure refers to upper internal space of the vessel, without taking (hydro)static pressure into account . This is widely applied, yet clarifications (any other practices, etc) were requested in post No 4.
Probably drum hydrostatic test has to use a test pressure including (hydro)static.

 

4. All above is assumed for temperatures low enough for carbon steel permissible stresses to be same as for ambient. Consequences are not known for design temperatures far from ambient. Relevant query has been placed concerning piping by sumano, see “Calculation of structural minimum thickness for pipe” (Hydrocarbon, oil and gas, 5 Oct 2012, - thread is referred in post No 7 for other reasons) .

 

(*) See para 13.8 “Design of vessels subject to combined loading”, Chapter 13 of “Chemical Engineering, Volume 6 – Design” by Coulson – Richardson. This Chapter is judged quite suitable for the Chemical Engineer interested in the elementary mechanical design of pressure vessels. Present strength of materials knowledge is more science than empirical art.

 

Editing note 15-12-2012: Perry (7th ed, p. 11-21) reports on coils that "when stainless steel or other high-alloy coils are not subject to corrosion or excessive pressure, they may be of schedule 5 or 10 pipe to keep costs at a minimum, although high quality welding is required for these thin walls to assure trouble free service". Pipe thickness for such schedules is 1.7 or 2.8 mm for diameters 1" - 2".

Welding of 1/8" ~ 3 mm plates is understood to be possible, http://www.advrider....ad.php?t=551794 . However techniques seem too specialized and not feasible for constructing a drum, at least from economical point of view.

As a previous post points out, welding plates 3/16" ~ 5 mm thick for vessels is next to impossible.

 

[Robvdh]

Even though your question is from some time ago, maybe this could be of some help in the future?

Attached Files

•  ASME_pressure_vessel_design-A(1).xls   389KB   401 downloads

[lafondejs]

Even though your question is from some time ago, maybe this could be of some help in the future?

Thanks.