API 650 12th Edition

The proof pages have been issued for review to the committee members for API 650 12th Edition.  This edition will be released later this spring and covers items approved by the committee.  There are some 27 agenda items that are being included in this edition.  It does not include items that were approved in the last couple of committee meetings.  We will be working on the changes to our ITSdesign software to include the changes that are coming.

Annular Ring Thickness Calculations

The following questions was posed to us:


1 - Current edition of section 5.5.3 of API-650 requires us to select the larger of
a) product design thickness plus corrosion allowance or
b) hydrotest thickness

2 - Note b to Table 5.1a requires us to use the larger of
a) product design stress or
b) hydrotest stress
in selecting the column to look at.

For a tank with
corrosion allowance = 1 mm  
bottom shell course thickness = 45 mm,
product design stress for bottom shell course = 220 MPa
hydrotest stress for bottom shell course = 240 MPa  :

item 2 above requires us to use the column for 240 MPa
bottom shell course thickness of 45 mm requires us to use the fifth line of Table 5.1a.
We end up with the 19 mm from among the tabulated values.

item 1 above requires us to use the larger of
a) 19 + 1 = 20 mm
b) 19 mm

If we could use the column for 220 MPa for the design conditions, then we would be comparing the value for design thickness as  a) 16 + 1 with the same b) 19 mm for the hydrotest condition.

With the way it currently reads we always end up having to add the corrosion allowance to the hydrotest thickness calculated.


Our response:

In reviewing the way 5.5.3 reads you are correct.  Since it specifies that you use the higher of the two stresses in determining which column you use and then must use the thickness shown plus corrosion allowance for the design condition, you will always be using the higher required thickness and then adding the corrosion allowance to that.  It would seem that we should be using the column that corresponds to the stress in the shell for whichever condition we are evaluating and use the whichever thickness is greater.  Since this is not how the standard currently reads, I have submitted an inquiry to the API committee for review.  It will take time to get a response and even longer if they decide to address the issue and make a change in the standard.

What questions do you have?

I have decided to do a question/answer for my blog. I would like you to email me a question you have about the API-650 standard that you would like answered on this blog.  Please ask one question per email.  Send your question directly to me at jschroeder@innovativetanksolution.com.

In addition I will try to keep you up to date as to what is happening with the standard.  I am currently getting ready for the next API standards meeting and will post a brief update as I get information.

API 2000 Venting

There is a new version of API 2000 for venting calculations. ITSdesign now includes this new edition for venting. There are two aspects to the new version of API 2000. The first is that to use calculations for venting that were in the previous edition, you must be using Annex A. Annex A does have restrictions that says the liquid stored must be similar to gasoline, the maximum operating temperature of the vapor space is approximately 120 degrees F (48.9 degrees C), the tank is uninsulated, and the capacity is less than 180,000 BBL (30,000 cubic meters). If it meets these criteria and you use Annex A, the results are the same as the previous standard. If, however, you use the base standard you will find that you need a lot more information about what you are storing. You need to know if the product stored is similar to Hexane, what is the Latent Heat of Vaporization of the product stored, what is the Relative Molecular Mass of the vapor, and what is the boiling point of the liquid at the relieving pressure. If you have insulation on the tank and you wish to reduce the emergency venting requirements because of the insulation, you need to know what the thermal conductivity is of the insulation. You also need to know what the approximate latitude is for the location of the tank installation. The answers to these questions could greatly affect the venting requirements especially for normal venting. It appears from some of our preliminary calculations that the normal venting per the base standard can be significantly higher than what is required by Annex A. We have just posted a new version of our ITSdesign software that includes these new calculations.

Insulation thickness added to ITSdesign

We have just released a new version of ITSdesign that includes entries for insulation thickness and weights. You have the option of having insulation on just the shell, just the roof, or both. The insulation thickness and weight can be different for the shell and the roof. You can also specify as to whether the insulation weight is to be used in the resistance to overturning for wind and seismic calculations. The addition of insulation thickness affects the area of the tank exposed to wind. In the printouts for wind you will see a different value for the tank diameter and height when detemining the area exposed when the tank is insulated. This increases the overturning moment and the sliding forces. For seismic you will see an increase in the weights. Insulation thicknesses and weights are also shown in the printouts including in the loadings page.

Frangible roof design of anchored tanks

At the API committee meeting in November I brought up the subject of the frangible roof design requirements. In particular I asked about the requirement if 5.10.2.6.d which says "For anchored tanks of any diameter, the tank shall meet the requirements of 5.10.2.6.a and the anchorage and counterweight shall be designed for 3 times the failure pressure calculated by F.6 as specified in 5.12." One of the requirements is in 5.10.2.6.a.5 which gives the maximum area of the compression ring that you can have and still be considered frangible. I asked if this requirement can be removed when the tank is anchored since we have to design the anchorage for 3 times the failure pressure which is based on the actual area of the compression area. After some discussion the committee as a group decided that you still must meet the area requirement of 5.10.2.6.a.5. What this means is that for smaller diameter tanks meeting the frangibility requirements is very difficult. For shop built tanks built to Appendix J meeting the frangibility requirements is next to impossible.

What is next in ITSdesign

We are currently working on adding Appendix P.2 calculations to our ITSdesign software. It has turned out to be a bigger challenge than first anticipated and so is taking longer than expected to program. I will keep you informed as to the progress on this. With regard to Appendix P.3 that was deleted from the standard, the committee is still working on the review of the equations in relation to the charts. That work has not been completed and so we do not have a timetable for when this section will be put back into the standard. Since it is not ready for balloting as yet, it will be at least a year.

It is a New Year!

Welcome to the New Year! My first post for 2011 will be about Addendum3 of API 650. Where is it? It is still in the process. I know it should have been released in November but it wasn't. In fact we have not seen the final draft of it to review. The last word is that it should be released in February or March. I am looking forward to its release as there are some changes in it to help with reducing anchorage on some tanks due to wind. As soon as I know it is being released, I will let you know and have the changes ready to release in our ITSdesign software program.

Buckling due to Wind in Corroded Condition

For API 650 tanks the wind calculations (both resistance to overturning and for buckling) are performed in the new condition unless the customer specifies otherwise. As such we have always had a checkbox in the "Wind" portion of our design screen to specify that the wind calculations are to be done in the corroded condition. We have now added a second checkbox to allow you to do the resistance to overturning in the "new" conditiong but do the shell buckling analysis in the "corroded" conditiong. If this checkbox is marked, then the check for wind girders will be evaluated in the corroded condition rather than the new condition.

We continually make additions and modifications to our ITSdesign software to meet the needs of our clients. We have several modifications being evaluated or worked on because of requests received. A trial version of ITSdesign is available for download at www.innovativetanksolution.com.

API Committee Meeting

The next API committee meeting will be coming up in November. As a member of the Subcommittee for Aboveground Storage Tanks design group, I will be attending the meeting and bringing some of my concerns and hopefully revisions to the standards. I have several issues I will be bringing before the committee. If you have a particular concern you think needs to be addressed, you can try to send an inquiry to API through their normal channels or you could email me your concern and I will see if it is something that I think the committee should be addressing. I would need any comments by the end of September. You can email me at its@innovativetanksolution.com.

Corrosion Allowance on Columns

We have released a new version of our software that makes some additions to how corrosion allowances are handled for the columns of column supported roofs. Previously we applied the corrosion allowance to the outside only on pipe columns. For structural columns we applied the corrosion allowance to all surfaces of the beams. We have added a checkbox on the "Rafters & Columns" tab that allows you to specify that the corrosion allowance on pipe columns is to be applied to both the interior and the exterior. For structural columns we added a checkbox that allows you to specify that the corrosion allowance is the "Total" corrosion allowance rather than per surface. We also now note this on the "Design" page of the printout.

Factory Mutual 4020/4021

We have just released a new version of ITSdesign that includes a design for Factory Mutual 4020/4021 (December 2005) tanks. This is currently only in US units as we are working on the conversions for SI units that Factory Mutual did not provide. The Factory Mutual design is still being verified by one of our clients to make sure that we interpreted all of the standard correctly. We will be making changes to this design if any discrepencies are discovered. Factory Mutual has two different options for seismic design using live loads. We chose to use the 25% of live load option. They also require that the sloshing height not exceed the freeboard or you must do an evaluation of the loadings on the roof. We currently have only included the option to require that the freeboard exceed the sloshing height. We did find an error in the equation in 2.15.5 (c) for the design tension load on bolts for seismic. The equation should be multiplied by S rather than divided by S. We have incorporated this in our program.

If you are doing Factory Mutual tanks we would appreciate your evaluation of the design. If you do not currently use ITSdesign software, go to our website at www.innovativetanksolution.com and download a trial copy of our ITSdesign software.

API 650 Approved Agenda Items

We have just posted a new version of ITSdesign in which we did some minor corrections. One of those was to add the test pressure (1.25P) to the liquid height instead of just P for hydrotest when F.7 calculations are required. We also increased the allowable anchor bolt diameter for some grades of material and added A1554-105. We also did some other minor corrections.

The big change is that we have added several approved agenda items from the API committee meetings to our program. You must have client approval to be able to use agenda items that have been approved but not published in an addendum. These are items that will be published in the next addendum. Several of these affect shop built tanks as we have approval to use the materials at DMT listed in Appendix A without the impact testing. We also can build a shop built tank out of material thicker than 1/2". The other big change is changes in the wind calculations in 5.11 and changes in the formulas in Appendix F. These changes will have an affect on whether anchorage of tanks is required for wind and for wind/pressure combinations.

What Seismic Factor to use for Anchor Chairs?

One of our clients brought up a disturbing point in API 650 Appendix E for Seismic designs. I have queried several engineers and reviewed this matter and am still confused as to how to apply this. The problem is as follows. In E.6.2.1.2 the last paragraph says that when you have a mechanically anchored tank, the anchor embedment or attachment to the foundation, the anchor attachment assembly and the attachment to the shell shall be designed for PA. It then defines PA as the lesser of the minimum yield strength of the anchor bolt multiplied by the as-built cross-sectional area of the anchor or THREE times PAB. PAB is given in equation E.6.2.1.2-2 and is the load on the bolt due to seismic. Now the problem. E.7 is Detailing Requirements and the paragraphs in E.7 apply when SDS>0.05g which happens to be any time seismic design is required. E.7.4 says that connections and attachments for anchorage and other lateral force resisting components shall be designed to develop the strength of the anchor or FOUR times the calculated element design load. Based on E.7.4 the value in E.6.2.1.2 should then be 4 and not 3. Until we can get a definitive answer from API on this issue, we have modified our ITSdesign software for Addendum 2 designs to use 4 instead of 3. For many tanks this will have no affect. However, for some tanks it will place an increased design load on your shell and could cause you to need higher chairs or thicker bottom shell plate. All comments on this issue are appreciated.

API 650 Addendum 2 includes changes in roof designs by requiring a design for unbalanced snow loads. For column supported roofs this only affects the center column. If the roof slope is no more than 2 in 12, the unbalanced snow load per square foot is the same as the balanced snow load design. This does not affect rafters or girders. Also, the standard says that you do not need to include it for intermediate columns. The only additional requirement is that you have to check the center column for an eccentric load since the unbalanced snow is only on 135 degrees of the roof. This might cause a slight increase in the column size depending upon the loads.

Self-supporting cone and dome roofs have a new formula for checking unbalanced snow loads. Once again it can have an affect on the roof thickness depending upon the loads.

The biggest affect is on the supported cone roof without a center column that we have in our ITSdesign software. First, if your roof exceeds 2 in 12 pitch, then the snow load portion is increased by 50%. This will affect rafter sizes. The compression ring and tension ring are also affected no matter what the snow load is because of the unbalanced loading on the rings. To determine the stresses in the rings from this loading condition, a check at each rafter and between each rafter must be made to determine the location of the greatest tension load and the greatest compressive load on the rings. If you are using this configuration you may want to try to keep the roof slope to the 2 in 12 pitch. You will probably find the compression and tension rings are going to be larger even at this pitch.

Supported Cone Roof without Column

We have just posted our newest version of ITSdesign. ITSdesign now has a design for a supported cone roof without a center column. This is available for AWWA D100 tanks and API 650 carbon steel field erected tanks (base code and Appendix A designs). The design includes the axial load on the rafter in addition to the bending loads. The center compression ring can be a standard channel, a channel shape fabricated from plate, or a bar shape. Loadings on both the center compression ring and the outer tension ring are evaluated. The rafters and center compression ring may be on the inside of the tank or on the outside of the roof. If they are on the exterior of the roof, they will have to be welded to the roof to transfer the loading on the roof plate to the rafters. This type of support system allows you to place the rafters on the outside of the tank when you have a interior lining on the tank. Also, if you have a tall tank this can eliminate the long center column. If you are not currently a user of our ITSdesign program, be sure to download our free trial version from www.innovativetanksolution.com.

API 2000 Venting

Happy New Year! I will be trying to keep my New Year's resolution to blog more frequently. To start the year, let us look at what happens when API 2000 venting requirements are applied to a tank. My concern here comes about because of a tank I just designed for a client. The specifications said this tank was an atmospheric tank. Design and operating pressure are specified as atmospheric. All seems fine until we get futher into the specifications and come across the venting. In this section it was stated that the tank was to meet API 2000 venting requirements and provide the inflow and outflow rates. The rates are quite low as the tank is not very large. Even though the primary vent is to have a flame arrestor, the required pressure to provide the venting required for pressure and vacuum are both less than 1" of water. This had no affect on the tank. However, when we get to emergency venting the specifications said that the pressure required to provide full emergency venting could not exceed the design pressure by more than a certain percentage. Upon contacting the supplier of the emergency vent we discovered that we would have to design the tank for 4" of water pressure to meet this requirement based on the venting capability of the available vents. Needless to say this did have an affect on the tank design. We now have to anchor the tank for pressure where before the tank would not have required anchoring.

Venting requirements and the pressure and vacuum needed to meet these requirements can change the design of your tanks. A tank may be specifed to operate at atmospheric pressure. However, the venting and particular the emergency venting requirements may require the tank to be designed for pressure. Be aware of what pressure is required by the emergency and normal vents to meet the venting capacity. It may start relieving at a relatively low pressure but may require a higher pressure to meet the capacity required.

API 650 11th Edition Addendum 2

Addendum 2 to API 650 11th Edition has been released. The effective date for this addendum is May 1, 2010. I will give just a brief overview of some of the bigger changes. I will do a more complete review later. First, they have changed Tables 5-6a and 5-6b and revised the minimum distance from the bottom of the tank to the center of a nozzle for regular type fittings. Snow load requirements have changed such that you now will have to consider both balanced and unbalanced snow loads. In Appendix V that have changed the stability factor such that there are different values depending upon certain conditions. The one that surprised me the most is that they slipped in Appendix Y. This was added after the proof copies were sent out. Appendix Y has added back into the standard the API monogram system. It appears that you can now place the API monogram on a tank nameplate if you have been certified to do so. You may find in the future that some companies will require this certification to be able to bid on and build their tanks. It will be something to watch out for.

Seismic Av and TL values

The question comes up as to what is the Av and TL values in our ITSdesign program? These values are listed under the seismic input section on the design page. Av is the vertical acceleration value and is entered as a value of g. This input is different than the Ss and S1 values which are entered as a % of g. In API-650 it is up to the client to specify if a vertical acceleration is to be included in the seismic calculations. If it is, they are to provide this value. In AWWA D100 there is a minimum value that is to be used based on the other seismic values. A client can specify a higher value though.

The second value, TL, is the Region-Dependent Transition Period for Longer-Period Ground Motion. This value depends upon the location of the tank and can be 4, 6, 8, 12, or 16. Areas from Eastern Texas to Eastern Kentucky and from Central Minnesota to the Gulf Coast are listed as a 12. The Northeast states are a 6 while the Southeast states down to Florida are an 8. The Southwest is a 6 and the upper Midwest is a 4. West of the Rocky Mountains it is a 6, 8, 12, or 16 depending upon your location. AWWA D100 printed the map of these values in the seismic section, Section 13 Figure 19. Unfortunately, API 650 did not include this map in their standard. If you do not have the AWWA D100 standard, you would need to look in ASCE7 for this map.

Jack Rafter System

We have just released a new version of ITSdesign that has a Jack Rafter System. This allows you to use a single center column on tanks as large as 90' to 100' diameter. With this system half of the rafters go from the center column to the shell. In between each of these Thru Rafters is a Girder that is made from a small channel and is located at approximately the mid point of the Thru Rafter. From the center of the Girder to the shell is placed a Jack Rafter. Thru Rafters get to be quite heavy but the Jack Rafters are relatively small. It does increase the load on the center column and on the soil below the column. However, the cost savings can be significant as you do not need intermediate columns. It also reduces the diameter of the cap plate on the center column considerable. The erection cost for this type of system is less than for one with intermediate columns.

If you haven't downloaded a free trial copy of ITSdesign, do so today and give this new system a try. Just go to www.innnovativetanksolution.com and click on the ITSdesign download button. Follow the instructions for installing the trial program.