Secondary containment seismic performance 1

[DBreyer]

Hi All,

We have to revisit the seismic design of some older flat bottom tanks.
Initial assessment is that the tanks themselves won’t be able to resist the seismic loads. They are built on crushed stone/sand pads, are not anchored, and have a diameter-to-height-ratio of less then 2.
Our legislation, however, requires only primary or secondary containment to be able to resist seismic loads and stay leak tight.

Secondary containment contains of a subgrade clay liner and earthen dikes with clay cores.
I’m wondering if you can demonstrate the seismic performance of this kind of construction?
Before we start putting resources to the task I wanted to ask if you know if this has been successfully done before and maybe have some references/contacts for this?

Kind regards

 

[Geoff13]

Presumably the thinking behind the legislation is that if the primary tank fails during the seismic event, that's OK as long as the secondary containment can handle the failure of the primary tank as well as the seismic event.

However, an earthen dike secondary containment is only designed to hold the volume of contents in the tank. This is OK as long as the tank has a "slow" leak. It might be that the seismic failure of the tank results in a "slow" leak, such as the piping ripping a nozzle out of the shell and slowly draining the tank. My concern would be if the tank has a catastrophic failure, and releases its contents as a large wave. This will certainly wash over the secondary dike, as has happened during other catastrophic tank failures.

Tanks have amazing ability to deform without failure during earthquakes. Many tanks have had significant deformation (elephant foot bulging, permanent uplift at shell, etc) without catastrophically failing. Searching the internet can find examples and reports. However I don't know how to establish how far past the unanchored tank criteria in API 650 Annex E you can go with only permanent deformation, and when a catastrophic failure finally kicks in.

I might consider the primary tank as meeting the legislation if it only exceeds the API unanchored criteria by a "small" amount. If it exceeds by a "large enough" amount that catastrophic failure might result, then I wouldn't accept an earthen dike secondary containment as meeting the legislation.

 

[HTURKAK]

@ DBreyer (Materials)
Your internet country code (DE) .If this is true , Eurocodes are the applicable codes in your zone .

The geometric design and lining ( size , ht, free board etc ) could be designed as per relevant environmental codes / owners demands .
However , the stability of the slope shall be checked as per requirements of Eurocode 8: Design of structures for earthquake
resistance — Part 5: Foundations, retaining structures and geotechnical aspects.

You did not mention about the content, size of the tank,seismicity etc.

Can reducing the max. liquid operation ht be an option ? .So the tanks themselves be able to resist the seismic load ?






According to the grace of God which is given
unto me, as a wise masterbuilder, I have laid the foundation, and another buildeth thereon. . . .
I Corinthians 3:10

 

[DBreyer]

Thanks for the replies so far.

@Geoff13:
Even without an earthquake the described wave is a credible scenario (e.g. due to gutter type corrosion). Only tank in tank systems wouldn't face something like that.
Regarding advanced analysis. I was thinking about performing a pushover analysis. But I am not sure if this will really show significant improvements.

@HTURKAK :
I'm located in Europe, so you are correct, Eurocodes would be the right code.
What I wanted to know is, if someone was already able to demonstrate necessary seismic performance for such secondary containments?
. Adn ground characteristics C-S
Tanks I am talking about mainly handle petrol components. Tanks are around 15m high with diameter ranging from 10 to 25m. Seismic ground acceleration is 0.56m/s2 and ground characteristics C-S

Regarding limiting max. Liquid level. That would be kind of a last resort, as we would limit logistics with this.