Seismic calculations against hydrotest weight of vessel 3

[Mech2325]

Hi,
COMPRESS has the option of taking a specific percentage of wind speed against hydrotest weight and then it calculates the base shear and bending moment accordingly. But i dint find this option against seismic load. The COMPRESS software only calculates seismic base shear and bending moment against empty and operating weight of vessel. This makes sense that hydrotest weight is a short term effect and the base shear and bending moments against this weight would be excessive leading to a overdesigned foundation but then PV elite allows you to take a percentage of seismic load against hydrotest weight. Can anybody tell of there is such an option in compress afterall it is a good tool to be used. And any references from asme code where it allows you to take less seismic load against hydrotest weight

 

[TGS4]

Such a load case is not one of the required design load combinations. However, the load combinations in the Code are the minimum required and the user is free to request additional ones.

There is a basis for the reduced wind at the hydrotest case: one could require a maximum wind speed during the hydrotest - something that can be monitored. I'm not sure on what reasonable basis you could develop such a seismic load and how you would control for that.

Think about it.

 

[Mech2325]

TGS4: COMPRESS does not take into account the seismic effect during hydrotest. What if there is an earthquake when the vessel is being hydrotested at field?? The foundation has not been designed for that. This is the question actually

 

[JStephen]

If that's a critical concern, you could do the whole design assuming hydrotest fluid levels.
What if you're hydrotesting during a hurricane, and then have an earthquake, for that matter.

 

[racookpe1978]

What is the normal fluid level and weight and pressure and temperature in the vessel?
What is the difference in weight and fluid level between normal operation and the hydro test?

 

[TomBarsh]

As others have indicated, combining earthquake (seismic) with the test condition is (1) not provided (directly) by COMPRESS, and (2) is not "required" by ASME, or by ASCE 7-10, etc. In our original judgment at Codeware it did not seem to be a realistic requirement, in the sense that due to the short-term duration of the seismic condition it was deemed an acceptable risk. However, we have received inquiries from clients located in high-seismic regions such as Turkey and adjacent areas, where the vessel owner's purchase specification has indicated this as a requirement. Consequently, this item was previously added to our program development list.

You can work around this by changing the calculation mode to "Rating Mode", then change the vessel model to reflect the test conditions: pressure, temperature, wind, seismic, corrosion condition, etc, as required.

I agree with JStephen's comments. Hurricanes give some warning, and one could always shut down the test should the hurricane approach locally. But earthquakes rarely give such notice. In the US and Canada, the building codes do not combine wind loads with seismic loads. It's seen as a statistically unlikely scenario for which the risk due the omission is acceptable to society. Of course, any individual structure owner may have other ideas and specify their design to meet such extreme cases. As I recall, even ASCE 7-10 does not address loads for tornados. We certainly have a lot of loss and destruction due to tornados but so far consideration for this has not made it into the Codes. Perhaps this will change in some manner after that school in "Tornado Alley" was directly hit a year or two ago.

Tom Barsh @ Codeware (but speaking for myself only at this time)

 

[jte]

Reminds me of a time when me and a senior vessel engineer were summoned to the project manager's office. Our client (a refinery) wanted to know what would happen if an earthquake hit on a large field fabricated column during the hydrotest.

Me: "Did Civil provide sufficient drainage for the site?"
Sr. Engineer: "Look at the bright side, the water will help put the fires out."

Project Manager: "I knew you guys would say something like that!"

End of discussion. The vessel is still there, by the way. I see it when I drive past when I'm in that area.

 

[TGS4]

Mech2325 - first off, get your head out of your software. Whether or not any piece of software gives you the ability to perform a specific calculation is neither here nor there. What matters is the reasonable-ness of what you are asking.

I think that TomBarsh has addressed this issue adequately in his response. We, as an engineering community (through our Code and Standards) have determined that it is not reasonable to design for a contingency of simultaneous hydrotest and earthquake. jte's response address the practical aspects of this.

Besides, what are you going to use for design margin / evaluation criteria even if your software allowed you to perform such a calculation?

 

[r6155]

I worked (30 years ago) in nuclear plant Project in site where no earthquake , but the plant is designed for earthquake.

Regard
r6155

 

[Mech2325]

Tom barsh and tgs4: Thank you for your valuable comments.
Just one more question from Mr. Tom Barsh ASME does not restict you to perform seismic calculations against hydrotest but it is considered as an acceptable risk and hence we do not perform such calculations. Am i right in my understanding?

 

[TGS4]

Mech2325 - your understanding is correct - seismic + hydrotest is not one of the minimum required load combinations.

 

[TomBarsh]

I agree with others that ASME, ASCE, et al, do not seem to consider it a defined load combination to consider seismic at test condition. The current Section VIII Division 2 is 'state of the art' for ASME Code and Tables 4.1.1 and 4.1.2 do not indicate this as a required consideration. But on the other hand I'm not even certain at the moment that they indicate anything with regard to the test condition (eg: not defining test + wind). In any case, it seems to be a given that it's acceptable in North America not to consider %seismic + test. We consider the risk to be acceptable that if an earthquake hits while a vessel is being tested we may have to sink in more money to fix it (and everything else).

But this may not be the case world-wide, nor should it follow so automatically, in my opinion. I know for a fact that a number of companies that operate in regions of high seismicity, such as Turkey, Pakistan, etc, do have such requirements to consider %seismic + test. This is "their call", they are the ones paying for it (or paying for a crashed vessel at test). Hopefully, if this design presents an "undue burden" to the fabricator there may be some room for discussion and negotiation on the issue.

I think the concept of risk (and consequences) applies even more greatly if the vessel is in its final operating position in an operating plant. This might be for a future corroded test condition (after repairs, etc) or for field fabricated vessel. If an earthquake hits while the vessel is being tested then the consequences of failure are no longer limited to just the vessel being tested; it may affect a number of other vessels, piping, components, etc.

So really, this concept seems less and less foreign to me. As I mentioned, this is on our (Codeware) development list. But the condition can be investigated by the designer through proper set-up of the model in current builds of COMPRESS.

 

[Mech2325]

Thank you Mr. Tom Barsh. Your comments have been very helpful

 

[TomBarsh]

One more comment to add to this discussion.

Quoting the original post "This makes sense that hydrotest weight is a short term effect and the base shear and bending moments against this weight would be excessive leading to a overdesigned foundation..."

It's the reference to "foundation" that catches my eye.

Another consideration is that many (most?) vessels will be shop tested. Vessels on saddles will certainly be tested in the horizontal position, and many vertical vessels (supported in their final installation on skirts, legs, etc) will also be tested in the horizontal position with temporary saddle supports. Surely the temporary "foundations" for the vessel's shop test are unlikely to be fully designed/investigated for the effect of seismic loads (or even wind loads) at test.

It's likely reasonable to design the vessel's final foundation for the condition of field test condition with a seismic load possibility...assuming that this is a region of high seismicity. But to require this for a shop test may not be so reasonable especially in light of the very temporary nature of this condition.

 

[Mech2325]

TOM Barsh
So, you are saying that foundation should be designed taking into consideration hydrostatic condition too??

 

[TGS4]

If the vessel will be (or is at least anticipated that there is the possibility of) a field hydrostatic test, then absolutely the foundation should be design for that situation.

 

[NRP99]

All have explained about why you should not take occasional loads combined together in rather best way, consider following perspective.

Let us design your equipment considering highest magnitude tornedo/torpedo/hurricane combined with the highest magnitude earthquake both happened in the past. We will consider/design vessel for hydrotest which will be performed in the above conditions. Now all is set and you have taken due care of all these emergencies and you know your design can withstand any of the emergency combined.

Now imagine the vessel cost, weight, foundation design, structural attachment design, support design and most importantly the efforts you/your team/project team has put to transform this design on paper and on the actual site.

Now imagine a tornedo/hurricane/earthquake or all together hit the field of hydrotest or plant during operation which is of very high magnitude not seen earlier or it will be going to set new record. Will your efforts pay you respect? I guess the answer is unfortunately NO.

I will only say, we take calculated risk in designing a vessel or any design which are efficient cost/effort wise. But we cannot avoid risk totally. That is impossible.