overflow nozzle location

[McCormick93]

A design proposal for an atmospheric storage tank shows the high liquid level all the way up to the top angle. This corresponds to the requested nominal capacity of the tank.

Does API 650 allow the overflow nozzle to be located on the roof? Or is there a minimum freeboard above the stated capacity - which means this supplier has really offered a smaller tank than we asked for?

Thanks in advance.

 

[JoeTank]

The design liquid level may be specified by the purchaser regardless of overflow location. Overflows on petroleum tanks would be very unusual. API 650 does not address location of an overflow pipe as they are seldom used.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[panduru]

API Standards are often used in design of tanks for other liquids as well, many non-volatile.

Are there any other standards/ references commonly used in conjunction with API, for issues not addressed in API, then?

 

[sethoflagos]

The design capacity by API 650 is limited to the lower of the top angle or bottom of the overflow.

The normal arrangement for the overflow is via an internal 90 deg spool from the HLL suface down and out through a side nozzle below HLL.

 

[panduru]

Agreed, as per API 650, the maximum for design level = HLL = top angle.

But in case of other than petroleum products?

 

[McCormick93]

Thank you sethoflagos, that makes perfect sense. As panduru noted, this is not a petroleum tank, but the owner wants API compliance and a controlled overflow path.

 

[JStephen]

The API-650 standard is set up for petroleum products. While the standard might be used for molten sulfur, acid, water, or other miscellaneous products, it doesn't attempt to cover all the details of these other applications.

Typically, water tanks are built to AWWA or NFPA standards. These standards require an overflow, with weir box if required, at some point below the top of the shell. The capacity is supposed to be that below the overflow, although many people still use the API terminology and use a nominal capacity which is figured to the top of the shell.

Depending on the size and configuration of the tank, it might be possible to put an overflow nozzle in the roof, but this would be uncommon. If doing so, it would be helpful to come up with a lower configuration than a normal nozzle + ell arrangement.

In the design of API tanks, assuming the design liquid level is at the top of the shell is a conservative assumption that need not relate to the actual overflow depth.

 

[panduru]

JStephen: In the design of API tanks, assuming the design liquid level is at the top of the shell is a conservative assumption that need not relate to the actual overflow depth. - absolutely right.

As for sethoflagos: The design capacity by API 650 is limited to the lower of the top angle or bottom of the overflow. I was wondering if the argument should have really been the other way round - do we locate the overflow nozzle at a point such that it establishes the maximum liquid level required? Isn't there at least some rule of thumb for the minimum freeboard? Assuming non-petroleum products, of course.

 

[IFRs]

There is no need for freeboard when designing the tank. Operationally there may be need for freeboard ( floating roof, overfill prevention, etc). Tanks are usually designed to safely contain the liquid to the maximum level possible which is usually the top of the shell or top angle, or the elevation at which liquid can not go above (overflow etc). If the overflow is not an open slot (ie pipe) then it may be advisable to design for liquid to be to the top of the top angle in case the pipe were to be blocked. Overflow slots are often seen on API tanks. API 650 references such shell openings. API has a document with overfill protection recomendations.

 

[sethoflagos]

Panduru: A basic rule of thumb that is often used is to calculate the process design working capacity requirement (based on filling frequency, surge capacity, off-loading volumes etc.) and divide by 0.85 for mechanical design. The 15% extra allows for ullage losses at the bottom and a bit of freeboard at the top for process upset, installation of alarm or input trip instrumentation etc.

For fixed roof tanks, if you set the overflow level below the top angle, the tank is going to require more (and possibly thicker) plate for the same working capacity so optimum economic design is to have both at the same level.

 

[panduru]

Thank you, sethoflagos. Now, there's a very useful design guideline.

 

[sethoflagos]

You're welcome.

I would stress that this is only a general guide - there are many dozens of tank specifications worldwide, and even more special process cases. So there are some exceptions to the rule.

 

[JoeTank]

If product will freeze, keep level during overflow to a point that is below rafter flange.

How high above overflow will you let product depth go during overflow event? AWWA D100 limits it to 6 inches.

The limiting factor on most overflow systems is the ability to overcomne entrance losses. Therefore, overbox with generous weir length is often needed.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com

 

[sethoflagos]

Absolutely right, Steve. The volume rating for the overflow must be maximum inflow with no outflow.

Knowing this volume flow you can calculate a weir length for a particular dh above weir lip (2" is an acceptable first estimate for a simple pipe overflow - the volume flow varies with dh^1.5). If, say this weir length computes to 36", the perimeter of a 12" pipe looks like a good candidate for the duty.

If the 'simple' solution starts giving an uncomfortably large overflow pipe, as you say a part or full circumferential weir will become the preferred option.