ATMOSPHERIC DISTILLATION BOTTOM TANK

[roker]

Hello,

We intend to do some tests on atmospheric crude distillation bottom's, for this a tank of 5 to 10 m3 will be used as the start point of the tests.
The tests will be done in the refinery.
Can this tank be vented to the atmosphere (with a flame arrester) or with a pressure control with fuel gas inlet at low pressure and outlet to the flare at high pressure, if so the tank should be designed as a pressure vessel which will dictate also a pressure relief valve.
Is blanketing required ?

regards,
roker

 

[TD2K]

I'm not clear what your system looks like. Atmospheric tower bottoms will be bubble point liquid so unless you are going to cool them (which is likely but not stated). Please provide some more details what this system will look like and what specifically you are using it for (it's too large just to take samples I would think).

 

[roker]

Dear TD2K,

We intend to do some experiments using the atmospheric distillation bottom's, the liquid will be cooled to about 200 C and stored in the tank from there it will be pumped to the downstream experiment equipment.

 

[TD2K]

I'm not usually crazy about routing tanks to a flare unless you have a specific flare/vent header just for this. The problem is that tanks typically take very little pressure although some tanks like API 620 can go up to the transition to pressure vessels.

Inert gas pad should be based on the expected vapor pressure of the crude tower bottoms at the storage conditions and flammability range. 200C seems high enough that I would be concerned it would still have a significant vapor pressure driving you outside the range of an atmospheric tank.

Venting to the atmosphere would depend on your operating air permit (if that is a factor), other equipment in the area and its proximity to this vessel. Other things I would want to look at would be the potential for a vapor blow-by to this tank from the crude unit and the potential to overpressure the vessel.

If you have a PFD of the proposed system, I would suggest doing some form of a hazard study on it. This could be a PHA format, check-list or what-if. My preference is for a Hazop because it gets into the possible interactions and upset issues.