17 replies to this topic

[farid.k]

Anyone experience of doing bundwall sizing using Torricelli’s law (refer attached picture)?

some client request to consider that in bundwall sizing. i never experience that in oil and gas industry. Typically design practice i.e. as per IP 19 Fire precautions at petroleum refineries and bulk storage installations will consider the LOPC containment for bundwall sizing.

 

Attached Files

•  Untitled.png   84.89KB   5 downloads

[breizh]

https://cdn.shopify.....compressed.pdf

 

Hi ,

I've attached a link with a reference to your drawing . page 86 .

Note : There is a typo in the comments , it's I which is Square rooted not I^2.

 

First time I saw this type of calculation . No question about the derivation .

 

Better to talk to the safety bureau or fire brigade to get advice . If h is too high , difficulty for the fire brigade to intervain  and it may generate issue with the tank ( stability) .

Make sure that the capacity of retention is big enough to recover the total quantity of material ( chemical) + some water 

My view

Breizh

 

edit: additional  document from  EPA 

https://www.epa.gov/...le_vertical.pdf

Edited by breizh, 25 June 2020 - 12:44 AM.

[farid.k]

https://cdn.shopify.....compressed.pdf

Hi ,
I've attached a link with a reference to your drawing . page 86 .
Note : There is a typo in the comments , it's I which is Square rooted not I^2.

First time I saw this type of calculation . No question about the derivation .

Better to talk to the safety bureau or fire brigade to get advice . If h is too high , difficulty for the fire brigade to intervain and it may generate issue with the tank ( stability) .
Make sure that the capacity of retention is big enough to recover the total quantity of material ( chemical) + some water
My view
Breizh

Thanks sir for feedback..i am not sure to follow that requirement or not. For me, as long as it can hold the containment, not required to spend more money to make it even higher..i never found that requirement in oil and gas related standard like API, NFPA, IP or any equivalent..

[thorium90]

Actually, in my country it is common to size the bundwall height for jetting effects as well. But not to increase the height infinitely, thats not practical. But to adjust the bundwall height and distance from tank to bundwall so that any jet will not exit the bunded area. The typical result is a bund wall that is about 1 to 2m high and a tank to bundwall distance that is half the height of the tank max liquid level.

Ultimately, a bund wall is useless if the liquid can possibly throw itself outside of the containment bund.

Alternatively, another option that I have designed before is to use splash curtains. These will allow the bundwall to be near to the tank due to space constraints and the jet will splash onto the curtain or cover and drip down into the bunded area. The height of the splash barrier required is calculated. The volume of the bund wall will still meet the containment volume required without needing an excessively high wall.

Edited by thorium90, 24 June 2020 - 11:34 AM.

[farid.k]

Actually, in my country it is common to size the bundwall height for jetting effects as well. But not to increase the height infinitely, thats not practical. But to adjust the bundwall height and distance from tank to bundwall so that any jet will not exit the bunded area. The typical result is a bund wall that is about 1 to 2m high and a tank to bundwall distance that is half the height of the tank max liquid level.
Ultimately, a bund wall is useless if the liquid can possibly throw itself outside of the containment bund.

Any specific international standards (oil and gas related) referred? I don't mind to follow as long as have solid reference

[latexman]

I know it as a company standard, not a jurisdiction standard. 

[farid.k]

I know it as a company standard, not a jurisdiction standard.

Thanks sir for valuable feedback. May i know which company?

Edited by farid.k, 24 June 2020 - 01:08 PM.

[Zauberberg]

Tank leakages (loss of containment) typically occur as one of the following:

- Bottom area leaks

- Vapor/liquid interface area leak (mainly due to corrosion)

- Catastrophic tank failure (collapse)

 

If you use any particular standard available out there, you will follow a prescriptive approach - meaning it will be over-conservative for 99% cases but it may not be sufficient for 1% highly unlikely cases with extreme consequences.

 

The real question here is - are you looking to cover all possible scenarios, or to manage the risk effectively? If you want to cover all scenarios, then the answer is obvious: design for it. If you want to manage risks effectively, then look closely at the release (loss of containment) scenarios, their likelihood and consequences, and see where are you on the risk curve. My 2 cents.

[latexman]

I first saw it at Dow Chemical.

[breizh]

Hi,

For those interested a good website : EPA

https://www.epa.gov/...tion-worksheets

 

Breizh

[farid.k]

Hi,
For those interested a good website : EPA
https://www.epa.gov/...tion-worksheets

Breizh

Thanks for such good example. But i didn't see any torricelli factor considered in the calculation isn't?

Edited by farid.k, 25 June 2020 - 10:10 AM.

[farid.k]

I first saw it at Dow Chemical.

Noted sir..maybe DOW practising more conservatives way in any design calculation.

[breizh]

Hi, 

Yes you are correct . EPA is not refering to Torricelli ,  1.5 m < h < 1.8 m in US & UK  with a minimum distance between the tank and the wall.

 

Breizh .

Edited by breizh, 25 June 2020 - 10:54 PM.

[latexman]

It's more like . . . Dow is headquartered in state of Michigan and 

 

"Michigan Division Environmental Standard E-600 requires that a tank shall be located within a containment structure such that the maximum trajectory of a leak from the tank is contained"

 

and Dow is a big believer in standardization.

[astro]

 

Actually, in my country it is common to size the bundwall height for jetting effects as well. But not to increase the height infinitely, thats not practical. But to adjust the bundwall height and distance from tank to bundwall so that any jet will not exit the bunded area. The typical result is a bund wall that is about 1 to 2m high and a tank to bundwall distance that is half the height of the tank max liquid level.
Ultimately, a bund wall is useless if the liquid can possibly throw itself outside of the containment bund.

Any specific international standards (oil and gas related) referred? I don't mind to follow as long as have solid reference

 

Various Australian standards relating to the atmospheric pressure storage of hazardous materials (including hydrocarbons and corrosives ) have referred to the leakage crest locus limits for (I'm guessing) probably for 2 decades or more.

 

Here's a quote from AS 1940-2017 The storage and handling of flammable and combustible liquids, 5.8.3 (Storage in tanks) Design and construction:

 

(h) The location of a bund relative to the closest tank shall be such that the top inside perimeter of the bund is not inside the crest locus limit specified in Figure 5.2, except where it can be demonstrated that a reduced distance would be appropriate due to viscosity or other considerations.

Refer the shot from one the figures referenced in the quote:
 Crest Locus Limit.jpg   19.85KB   3 downloads

Using this design criteria, the bund wall height is a function of step out distance from the wall of the tank.

 

I would have thought that Australia is not the only jurisdiction to apply this sort of approach, which is undoubtedly derived from Toricelli's theorem.

Edited by astro, 29 June 2020 - 07:05 AM.

[breizh]

Hi,

To add to Astro's document , I found similar description in a document issued by New zealand EPA  (2012) . page 59.

 

I've added another document from UK to confirm my previous posts .

Good luck

Breizh 

Edited by breizh, 01 July 2020 - 09:06 AM.

[astro]

Looks like breizh posted a scan of the Kiwi code of practice. If anyone's after a better quality document look for Secondary containment systems (HSNOCOP 47) here:

https://worksafe.gov...substances-cop/

 

In 8. Rereferences, HSNOCOP 47 cites a number of the Australian standards that I was referring to. No surprise given the geographical proximity and the shared approach to standards.

 

I could see reference to the crest locus limit in HSNOCOP 47 but the UK standard HSG176 doesn't appear to offer anything similar. Both provide useful guidance nonetheless. Useful tip.

[farid.k]

Thanks All for such very supportive and valuable feedback..i do really appreciate it!!