Floating Roof Tank Overflow Protection 4

[HFCkh]

I have been assigned to size an overflow nozzle for both internal and external floating roof tanks. My question is that if I know the maximum fill rate of the tank, the 95% fill height of the tank, the sum of the fill nozzle sizes to the tank, what else is needed to calculate the size of the overflow nozzle(s) needed for the tank. I also have been told that typically there should be two nozzles (or holes) spaced 90 degrees from one another. My initial thought was that the overflow nozzle (or hole) should have a crosssectional area at least one size larger than the sum of the fill nozzles to the floating roof tank. I have as yet been able to find any calculation or standard to support this. Also, would I need to know the maximum roof height as well so as not to damage the floating roof seals?

 

[pleckner]

Suggest you obtain a copy of API RP2350, "Overfill Protection for Storage Tanks in Petroleum Facilities"

 

[HFCkh]

Appreciate your prompt reply, I have perused the API RP2350 and it seems to relate mostly to overfill protection in the form of detectors and switches, which we have on the tanks currently.

As far as I can see it does not give a recommendation or method of calculating the size of an overflow nozzle in the event that the detector/switch would malfunction or give a false reading.

Any further help is appreciated

 

[pleckner]

There is no magical equation. You need to treat this as any fluids calculation. You have a maximum fill rate and want to make sure that the exit is large enough to handle it. With the lack of your company standard to fall back on, just do a hydraulic calculation with a chosen nozzle size. You can use CRANE TP410 for assistance. The usual quicker way is what you've been told, add up all the crosss sectional areas coming in (those that are expected to be simultaneous inputs to the vessel) and increase the overflow nozzle by one size. This is where this basis came from, a fluids hydraulic balance.

 

[HFCkh]

Thanks again for the prompt reply. I truly appreciate your input and will see if the CRANE TP410 can aid further.

 

[agf94510]

It concerns me that someone is considering an overflow system for a situation that should be avoided in the first place.

From the practical point, any nozzle on a side of a tank would interfere with external floating roof seals. As the liquid reaches the nozzle elevation, the seal would be breached and some vapor would be released to the atmosphere resulting in an Air Permit Violation. For an internal floating roof (not Vapor Recovery), tank overfill has identical results. Vapor Recovery tanks (with or without an internal floating pan) would not be affected by this type of an incident.

Second unknown is the ability to recover the overflow volume. The inflow rate to a tank would need to be recovered in a second tank with adequate capacity. Considering that the overflowing liquid will trap some air, the second tank must be able to decant and purge that air.

Tank overfill prevention is achieved by a combination of properly designed tank gauging hardware and customized operating procedures. Consider redundant tank gauging, high level switches, level interlocks, inventory management software, and operator training.

 

[pleckner]

Well done @agf94510. I didn't pick this up because I am not that experienced in floating roof tanks. This is probably why you don't see any recommendations for overflow in API!

 

[JStephen]

In fact, it is not uncommon to have overflow slots or nozzles on petroleum tanks. Overflowing a tank through an overflow fitting is obviously a bad thing, but overflowing one without such a fitting can be worse.

If the nozzle or slot is completely under the fluid surface, you can apply Bernouli's equation between the fluid surface and the opening of the overflow. Assume zero velocity at the fluid surface and zero pressure at the opening and you can solve for velocity and then flow rate as a function of head.

If the nozzle or slot is only partially covered by the fluid, you get into a weir-flow situation. Refer to various fluid references for information on weir flow.

Use of the overflow may require that seals rise high enough to clear the opening.

 

[proinwv2]

Hey guys, don't ignore the friction effect of piping beyond that nozzle. All of the losses result in a pressure in the tank that drives the flow. Simply stated you will have inlet loss, piping friction loss, elevation loss or gain, and exit losses. It is a system unto itself and if undersized could result in tank damage or worse.

And yes it will probably start as weir flow and then progress to full orifice flow if the level rises enough.

I don't like rules of thumb for such critical issues. I assume that this is a rather large and expensive tank with a liquid one does not wish to escape to the environment. It is worth the effort to do it correctly.

Sorry if this appears to be a lecture, that is what happens when an ME reads a ChE forum.

A Happy and Safe New Year to all.

Paul

http://www.ostand.com

 

[IFRs]

Here are some thoughts. 1) The floating roof dips into the liquid at least 4". This dip skirt is likely to be about 6" to 8" from the tank shell. This presents a resistance to flow that may be significant and at the very least is in conflict with weir flow assumptions. 2) Floating roofs often have shoe seals that would block the overflow opening completely. These seals often extend below the floating roof by several inches or more. 3) Inflow into the tank is by high power pumps while your outflow is gravity with nearly zero head - simply doubling the inlet pipe size is likely to be inadequate.

 

[agf94510]

In response to JStephen’s comment “In fact, it is not uncommon to have overflow slots or nozzles on petroleum tanks”, I believe that the observation may be actually refer to vents that are present in an IFR tank just below the roofline. These vents prevent pressure or vacuum load on the external roof during transfers. These vents are well above the floating roof seals. An alternate solution to the vents would have been to install multiple P/V valves on top of the roof. The vents are a cost effective alternative.

The overflow nozzle (piped to about 12” above ground) is a common solution to prevent an “smaller” atmospheric tank from being overfilled. This design uses a small vent at top of the tank to equalize the pressure and a separate overflow line. The overflow line may have a liquid seal to retain tank vapors. In this case, there is no floating roof seal to interfere with the nozzle operation. As an example, this type of tank could be used to store NaOH or Acid. Please note that tank overfill causes a hazzard and should have secondary containment.

Expected roof damage in an overfill situation is directly related to roof design. External Floating Roof can have severe damage if the gauging platform limits travel. If not mechanically limited, the roof will float high and spill the contents over the side. Internal Floating Roof will have severe damage when it comes in contact with the fixed roof. Atmospheric or Vapor Recovery Tanks (no floating pan) can be damaged when the rising liquid pushes against the fixed roof. This normally causes the fixed roof to separate at its connection to the tank wall.

All tank overfills should be avoided.

 

[reena1957]

IFRs, you are 100% right. It is not at all astraight forward overflow nozzle as others have indicated. BEWARE of simple solutions. More often than not, they may lead to major problems!!!!

 

[IFRs]

FYI - the "vents" you see in the tank shell or fixed roof are for ventilation, per API 650 Appendix H. They serve to reduce fixed roof underside corrosion and keep the air/vapor mixture above the IFR below the LEL. The tank will be protected from pressure or vacuum as a by-product of the huge holes required for adequate ventilation (2 square feet per 10 feet of tank diameter). Many of the tanks that I design have no shell vents or overflow ports. You really won't hurt the floating roof that much by forcing it into the fixed roof rafters while a overfill (read "spill") costs so much more that any IFR damage is worth the extra protection of additional shell height. High, high-high and emergency alarms and shutoffs seen to do the trick. I have been designing, manufacturing, installing and repairing aluminum and steel IFRs since 1977 and designing tanks for nearly that long also.