What do I do when NFPA68 venting is not practicable and roof cannot not be designed as frangible?

[rikywere]

Hello,
I have a 6500 mm ID and 7500 mm height atmospheric tank. The tank stores water but there is a choice of having propylene into it (if something is broken somewere else). So that it has to be designed to withstand propylene deflagration. The problem comes when I go through NFPA 68 and the required venting device is too big (area) to be provided. Then I thought of frangible roof in accordance with API 650. But this small tank do not fulfill the requirements of API 650 for frangible roofs. ¿Any idea of how to solve this problem?

Many thanks,

 

[stookeyfpe]

What specification is the tank constructed to?

 

[rikywere]

It is a API 650 storage tank

 

[stookeyfpe]

Have you reviewed API RP 2000? That will provide you the greatest guidance.

 

[rikywere]

Hello,
I have reviewed ANSI/API standard 2000 "Venting Atmospheric and Low-pressure Storage Tanks". I followed the rules for emergency (fire outside of the tank but not into the tank) and operating venting (basically due to the liquid flow), but this standard is not applicable for deflagration venting (correct me if I am wrong). So, this is why I can not say that the tank is designed for propylene deflagration despite the fact that the tank already have venting device (a nozzle) according to API-2000.

 

[DavidCR]

I know nothing about the propylene but if at atmosferic conditions, I supose it is a gas. API 650 is only for liquids. So the frangible rules should not apply neither. It should be a special process tank maybe not in regular codes. Do you have contact with a plant process designer to see how to categorize this tank. For example we have some power plants that uses pentane with a unique design and some distances and design rules from NFPA or local codes cannot be followed easily.

 

[rikywere]

DavidCR:
No, the tank stores water, it is not designed to store propylene or any other gas. The tank is designed according to API650 to store water. The issue with the propylene will come only in case of failure at one heat exchanger far from the tank. But in case of this failure the propylene can reach the tank (disolved in the water) and this is why the client want us to design the tank to withstand this scenario.

 

[IFRs]

Anchor the tank and make the roof frangible by adding a weak seam that will blow under a calculated pressure?

 

[rikywere]

IFRs:
That was my second thought, but the weak seam do not depends only on the weld, it depends on the weld (no more than 6mm leg) the participating part of the shell, the participating part of the roof and the top angle. The problem is that for small tanks (as a general rule) the addition of all the participating areas are too much to consider the joint as frangible.
Thanks,

 

[IFRs]

I don't think the tank shell will ever buckle inwards but thought the OP could detail the fixed roof in such a way to have a substantial length of welded seam designed to fail at the required pressure. I would also revisit the venting calculations - small tanks such as these are made by the thousands - what makes this one special?

 

[rikywere]

IFRs.
I am sure nothing makes this one special, this is why i ask here, to get some guides from your (all of you) expertise. Any suggestion about how to calculate the lenght of weld?
Thanks,