Vibrational analysis on Rectangular tank

[kos1]

Hello everyone,

I am seeking your assistance once again for a unique challenge I currently face.

I have two rectangular tanks:
Tank 1: Dimensions - 60in H x 90in L x 88in W
Tank 2: Dimensions - 63in H x 90in L x 58in W

My goal is to obtain certification for both tanks according to the UN 31/A code (

https://publications.gc.ca/collections/collection_2022/ongc-cgsb/P29-043-146-2022-eng.pdf).

However, one particular requirement of the code poses a challenge: conducting vibration tests on the largest tank, which unfortunately no facility in North America is equipped to handle due to its size.

Consequently, the regulatory bodies, namely Transport Canada and DOT, have requested that I present them with a theoretical analysis demonstrating the adequacy of Tank 1's design based on the results obtained from Tank 2.

If any of you have any tips or guidance regarding how I can approach this situation, your input would be greatly appreciated.

Thank you.

 

[centondollar]

Surely, there are several facilities in the US that can accommodate such a small tank? You speak of inches; those dimensions in metric units are less than 2 meters by 2 meters by 2 meters. That is not an object with a large size.

Eigenvalue calculation with e.g., FEA will give you the mode shapes and natural frequencies of the tank. If they do not coincide with vibration expected to be exerted upon the tanks, there should be no vibration issues. You may also perform a linear dynamic analysis with external forcing at expected frequencies with or without damping to investigate the vibration amplitude as a function of time or frequency.

 

[kos1]

Hello @centodollar

Thank you for the replay and suggestion.

If you know of any facility that can accommodate such a large tank, (Note that it will need to be filled to 95% for the test). Please let me know.

 

[271828]

I'm confused why you're referring to these as large tanks. They're very small as tanks in "civil structures" go. Are they on the order of 2 m by 2 m by 2 m like centondollar typed?

There are many MEs out there who are set up to test small stuff. There are structural engineers out there set up to test bigger stuff. Surely someone can test this. What are the desired outputs from the tests? Natural frequencies, mode shape, stresses due to specific excitations, something else?

 

[kos1]

Hello @271828,

I suspect the reason is that the tanks need to be fully loaded with water as part of the test, and no facility has a test bed strong enough to take that load (Gross Mass for Tank 1 = 6150kg). If you know any facility that can do the test for this size/load, please forward it to me and I will be happy to follow up. I have contacted everyone single one found on the internet and they have all turned me down.

In terms of the desired output from the test. There should be no physical deformation or failure of the tank after the testing. And for the testing, the tank needs to be subjected to 1h of vibration with a double amplitude of 25mm frequency as per ASTM D999-96 Method A1.

 

[ThomasH]

I Googled "Seismic testing facility USA", the top result was this:

https://nts.com/services/testing/dynamics/ca-seismic/

Don't they meet your requirements?

 

[dhengr]

Fazlur99:
5'-0" H x 7'-6" L x 7'-4" W is not a particularly large container as far as structural testing goes. But, many labs may not like the idea of that much water all over their facility during the testing process. It would seem that most any lab with an EQ testing ‘shaker table’ could do this testing if they could fit you in their schedule and they didn’t mind the water. What are these containers made of, show us some construction drawings and details. How do you contain the liquid in transit? The transportation authorities want to know that the containers aren’t going to fall apart due to the fairly typical rigors in transit. How are you shipping these containers, by truck, on a railcar, or both, and how are they fixed to the trailer or railcar for transit, that’s an important detail too. By rail, the longitudinal and lateral horiz. design loads can be every bit as severe as the vert. loadings. Your test (Method A1) is a vertical vibration test, with 1" amplitude. What does “double amplitude” mean?

I suspect the AAR Testing Facility out in Pueblo, Colorado could do this for you, but it would be pretty costly.

 

[kos1]

Hi @ThomasH, @dhengr
As I have mentioned multiple times, they all rejected citing because the tank is too big. NTS was one of the first ones I reached out to because of its 50+ years of experience.

Im thinking of loading the model into Autodesk Nastran and using its Vibration analysis tool and compare the two results. And if the results are compatible, then I can make an argument to regulator that if the smaller one passes the vibration test, the larger one will too.

 

[ThomasH]

When I read about one of NTS sites I noticed that they mentioned paylods up to 75000 pounds. I suspect that a potential water leak may be of greater concern .

I also recently read about a seismic test for a building:

https://www.geekwire.com/2023/wood-...y-timber-building-stands-up-to-seattle-shake/

If you plan on going for an analysis I would first check that Autodesk Nastran suitable. Can it do this type of FSI. I have worked with the software but it was a while ago. So it might have improved. But this type of analysis also requires some expertise.

 

[Tmoose]

Do I understand correctly this requirement is related to shipping the tanks to the customer?
If so, I'm surprised they'd be filled with water during shipping.

Or, is the final use of the tanks to be used to transport water/liquids?
In that case I'd understand why durability when "shipping" is a concern.

If really used to transport water, is freeze protection included?

 

[kos1]

These are UN 31A/Y IBC tanks. They are designed to transport fuel from site to site and hence the need to be tested with water. They are designed as per CGSB 43-146 (Canada) & DOT 49 CFR 178 in the USA.
The vibration testing is mandatory as per the code.