Open Vent (Gooseneck) Sizing 1

[takiyasamsama]

Hi,

API 2000 does not mentioned anything about sizing selection for vent i.e. gooseneck. Thus, how do we decide on determine the size of the vent based on the calculated values (total inbreathing, total outbreathing, and emergency vent).

Say there is a problem where the inbreathing flowrate is the highest the rest and there is no need for emergency venting since the AST is design for frangible attachment. What are the deciding factors for calculating the size? I've also tried using Bernoulli equation to determine the velocity so I can use Q=Av formula to determine the vent pipe size however I don't it's right.

I also stumbled upon Morrison sizing selection and from that article / brochure there is something said about wetted area and vent size selection based on NFPA 30. I'm not sure that can be used or not so I seek experience from all user who have verse experience in determining the open vent size.

Thank you.

 

[IFRs]

If the in-breathing is only from product withdrawal you would be safe matching your open area ( net of any screens ) with the suction pipe size. If you are buying a vent, the manufacturer may have flow vs pressure data. Does API 2000 vents require vents to be calibrated by the manufacturer? Not much wrong with Bernoulli but you will have to determine your own entrance/exit/duct loss factor. If you are in a cold climate, account for freezing on the screen and drifting snow blocking the vent.

 

[takiyasamsama]

IFRs,

Thank you as always for your answer. The tank vent is not of vendor item, our third party design the tank with 2 qty of 12 inch open vents (gooseneck type), the tank outlet (discharge line) is 16" and to have 16" vent for the roof is prohibited as per API 650 for roof opening, thus from your answer I get the point is that our third party provided two 12" vents because they want to cater for the 16" outlet line, or am I wrong?

You also mentioned regarding the mesh open area, is it wise for me to calculate the flow leaving the tank through the vent by using area of the mesh?

Thank you very much

 

[IFRs]

16" is not exactly prohibited in API 650, it's just that the API 650 is silent on the size of the repad required for a 16" nozzle ( table 5.14 stops at 12" ). In any case, two 12" pipes are larger than one 16" so that seems to be the logic for using two 12" vents, and this normally is enough to protect the tank ( see NFPA 30 section 21.4.3.3 ). The open area net of the screens should be used, in my opinion. Also note that when the roof nozzle is used for venting, the neck should be trimmed flush with the roof-line ( see the note in figure 5.19 ).

See section 5.8.5.5 for the need for screens.

Although not mentioned in 5.8.5.5, see sections H.5.2.2.1 and H.5.2.2.2 and H.5.3.3 and G.9 for examples where API 650 requires the area net of the screens to be used, hence my suggestion that it be used for all vents.

A reasonable back-of-the-napkin check might be to use V = Sqrt( 2 * g * H / ρ ) and Q = C * V * A where
V is the velocity through the vent ( calculated )
g is the gravitational constant
H is the pressure across the vent ( usually set to be equal to or less than the weight of the roof plate, maximum of 5 kPa (0,725 psi) per API 2000 6.3.1 )
ρ is the density of air ( perhaps dry air at standard temperature and pressure )
C is a constant chosen to represent entrance, duct and exit losses ( perhaps using 0.60 , from API 2000 6.3.3 of 0.50 as a guide but considering that the free vent may not be quite as bad as a manway )
A is the net open area of the vent ( calculated )
Q is the flow through the vent ( calculated and should be greater than the maximum API 2000 in- or out- flow rate )

Calculation methods for theoretical flow rates are given in API 2000 section 6.3 and section D

See API 2000 Annex C for an image of a gooseneck vent without a screen or flame arrestor

Also look at if a flame arrestor is needed on the vents ( API 2000 4.5.4 )

NFPA 30, section 21.4.3.3 describes normal vents sized using API 2000, and goes on to say that "Alternatively, the
normal vent shall be at least as large as the largest filling or withdrawal connection..." so this may be where the contractor is getting the size.

 

[bimr]

Here is a discussion:

http://www.eng-tips.com/viewthread.cfm?qid=399769

 

[hfgnick]

When sizing vent piping for atmospheric tanks I use the Sptizglas equation for low pressure gas flow and determine the equivalent length of piping of the vent using standard values. For instance a goose neck vent would consist of a pipe inlet equivalent length, a small straight pipe length, two elbows equivalent length and a pipe discharge equivalent length. Using the value of equivalent length, API 2000 calculated flowrates and properties of air plugged into the Spitzglas equation will give you a positive pressure for outflow and a vacuum pressure for inflow resulting in the tank. You then compare this to the design vacuum and positive pressure of the tank. The tank design vacuum pressure is typically much less than the positive pressure but both or on the order of inches water gage for a fixed cone roof tank. I forgot the exact typical values at this time but possibly it may be stated on the tank fabricator drawings. I think that design vacuum is typically about 0.5 inches water gauge and design positive pressure is typically about 1 inches w.g to a few more inches but don't quote me on that as I do not have my references with me at this time.