Vortex Breaker with Cover Plate

[AndrewTX]

I found this old post on vortex breaker designs and it relates directly to an issue I have now.

thread378-36635

I have a liquid full FWKO where the oil/water interface is only 12" off the bottom of the vessel. The outlet nozzle is 6" and the current vortex breaker is a crossed plate with a height of 6". Hence, the interface is only 6" above the vortex breaker.

This vessel is on a floating offshore platform and the client notices spikes in the oil in water content during rough weather.

We've suggested to our client to put a cover plate on top of the vortex breaker to fource the nozzle to draw water from the bottom of the vessel and eliminate the possibility of down flow from the interface level.

Our client likes the idea but wants to get a quantitative assessment of the improvement in water quality by adding this plate. We've lined up a CFD analysis of the fluid flow inside the vessel. Now the client is interested in a qualitative assessment - meaning is there any empirical or historic data showing the benefits of vortext breakers in general and vortex breakers with cover plates in particular.

Does anyone know of any technical papers, or texts, or any other source that might have this kind of parametric data on the benefits of vortex breakers?

We are still suggesting the CFD analysis as the only way to be sure of the benefits for this specific application, but my client still would like some historic information on vortx breakers.

 

[rcooper]

I was under the impression CFD would not show up vortex formation. These take time to develop and CFD only shows instantaneous flow patterns.

In this situation I would consider a physical model to identify the cause of your problems and help produce solutions which will work. A full scale model test is expensive (20,000 GBP the last one I had done) but if it will cost that much to make alterations to your first solution if it doesn't work, then they are worth it.

 

[rzrbk]

A cover plate larger than the outlet diameter should help by decreasing the fluid velocity around its perimeter (spreading flow toward the nozzle over a larger circle) and by creating a longer path from the interface surface to the outlet. These same effects have been used to reduce vortexing in shallow pump suctions.

Also, this may be a stupid question, but during rough weather, is there enough movement in the platform to create some slow sloshing in the vessel? It probably would not take much to create some blurring of the interface level.

 

[unclesyd]

On a number of separators we use a cover plate over the vortex breakers. This cover plate leaves only about 8" to 10" from the sides of the separator that is 5' in dia. These separators are separating a gas from a solid/liquid stream. We have about 10 vessels with the vortex breaker covering.

This plate had always been refereed to as a "China man's hat".

 

[AndrewTX]

Thanks for the replys, guys.

Rcooper, the physical model will actually be more expensive than the CFD. We aren't trying to model the vortex per se, but to get a feel for the general fluid flow paths into the water outlet nozzle.

Putting the plate over the vortex breaker should force a flow path that draws water primarily in a horizontal direction from the bottom of the vessel; hence, giving cleaner water. That is the flow phenomenon we are looking to analyze.

Rzrbk, yes the motion in the vessel does cause slosh and that is leading to the spikes in oil content in the water outlet. This vessel has no motion baffles. Putting the cover plate on the vortex breaker is intended to mitigate the oil slugs going out with the water. We are also suggesting installing motion baffles.

Unclesyd, if it isn't company proprietary information, how did you come to the conclusion of such a large plate? Was there some experimental evidence or operating history used to justify this sizing?

 

[unclesyd]

As I once said don't worry about proprietary information, "The best thing that could happen is that our competitor pick up on our proprietary information"

All that I can recall is that the diameter of the cover plate was increased as we went up in rates. There has been about a 5 fold increase in throughput of this separator since the original installation.

I forgot to add in the previous post that the cover is conical, probably a 6" rise at the center. This form may be strength consideration, as we did have some fatigue problems.

I just checked and out of three process engineers polled no one knew anything about the design. Unfortunately the design engineer is no longer with us.

Anecdotal:
The irony of this is that this engineer (Chem E) was a scrupulous note/record keeper and kept a shelf of 3 ring binders with every calculation, in text book form, that he had ever done. The collection was tossed as the inheriting party concluded that everything was readily available from other sources.

 

[TLKemp]

I have tried to determine the computational basis for vortex breaker design since 1990 and have found the following:

There is very little in the non-proprietary literature on the subject. Sulzer has published some of their work in their pump handbook.

The important factors are the velocity in the nozzle and the height of liquid above the nozzle (the submergence).

If the nozzle is large and the flow low, the nozzle will be self-venting and no vortex will form. In other words any gas bubble drawn into, or vapor formed in the nozzle, can escape by rising up through the liquid. In my experience, designing for this gives large nozzles for low flows and is not economic.

Vortex breakers are generally specified (not designed) using company standards. None of the engineers working for my clients (U.S. and international oil and petrochemical companies)knows or can find the original basis for their standard. I strongly suspect that most of the designs are based on someone's personal preference.

Most VB designs are either a cross, with or without a hat, or of the subway grating type. Both types work (at least none of my clients have blamed the vortex breaker for pump trouble), but I am very hesitant to introduce any restriction into a pump suction line and prefer to design for adequate submergence. (CE Lummis had a chart based on I-don't-know-what that seemed to work).

The subway grating is preferred by some over the cross type and I think grating is less likely to plug and restrict the flow.

"Design" Rules:
Make the cross at least twice the nozzle diameter wide and as high as the diameter. Put a hat on it if you want, but know that several licensors of process technology do not consider the hat necessary. I extend the cross into the nozzle at least 80mm.

A standard grating design would be three layers of flat stock 25mm by 5mm arranged with 25mm by 75mm gaps. Make the grating square and at least 4D on a side. Weld all this together so that the 75mm spaces are at right angles to each other in alternating layers. Put a gap between the layers and hold it all together with "C" channel. There should be a gap between the bottom layer and the vessel of 0.5D or so. Most installations are st. steel or other corrosion resistant material, but I know some that are plain carbon steel and are replaced at regular intervals.

Don't worry too much about the exact dimensions.

My last thought is that I think the results of any CFD modeling will be strongly dependent on hte initial conditions chosen for the model. Whether these accurately reflect your physical (real-world) situation should be the biggest worry.

I have lost my copy of the Sulzer book. I would appreciate a copy of the section on submergence and vortices if possible. (Look for articles on "Irrotational Flow")

 

[bosse1950]

Excuse me for my simplified approach but sometimes in cases like this a redesign of the existing vortex breaker could suffice - eliminating the existing vessel vortex breaker & installing an loose mount internal vortex breaker mounted at the existing 6" outlet flange as a "bolt in" item. This approach would add the existing 6" internal projection to the vessel separation and allow for more efficient draw off.

 

[Mech85]

AndrewTx,

The following technical articles/standards have some information on vortex breakers/inhibitors

1)"Do-It-Yourself Vortex Breakers" by S. Waliullah P108-109 Chemical Engineering, 9 May 1988.

2) Section 5.3.2 Vortex Inhibitors from Australian Standard AS 2419.1 - 2005 'Fire hydrant installation - System design, installation and commissioning' has some design criteria for sizing plate type vortex inhibitors for fire water tanks.

3) The article "Suction Side Problems - Gas Entrainment" by James H. Ingram from "Pump Handbook - the Series on CD-ROM", Pumps & Systems 1999 Edition has a section on Vortex Breaker Design with criteria for 'hat' type and 'cross' type breakers.