Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations KootK on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Transmix/Interface in pipelines 1

Status
Not open for further replies.

Augusto5

Mechanical
Nov 2, 2005
2
Hi everybody;
I'm in the operations sector of product pipeline. One of the worst problems we have is the transmix generated between the Naphta for petrochemical (NVI) process and the Butane (BTN). Typically, the interface volume begins around 50 m3. To the terminal station arrives around 200 m3.
Resuming, I need some help with the transmix problem. Maybe, somebody could tell me how much each factor acts. As operator, I can set some conditions, such flow rate, pressure, pressure drop (bypassing prover). But the question is, how can I minimize the volume generated of interface?
I can send you any other specific information or installation's diagrams if you need.

I'll thank you a lot your help.

P.S.:I'd like to get in contact with people in the operations sector of product pipeline and oil pipeline.
 
Replies continue below

Recommended for you


Although I'm not involved in pipeline intermingling of successive products I have an old note that says that laminar flow would result in complete mixing of the products. It also has been established that there is a critical value for the Reynolds Number (Re) at which the rate of intermixing shows a sharp decrease. These depend on the pipe diameter as follows:

dia, inches Critical Re

4 22 000
6 29 000
8 34 000
10 40 000
12 46 000
14 72 000

The source for this info apparently is: Mixing of miscible but dissimilar liquids in serial flow in a pipeline Austin and Palfrey Proc. Instn. Mech. Engrs., 178 (I), 377-396.
 
25362,

I would have thought that laminar flow would inhibit mixing, as each flow regime stays in its own path or layer. I would think turbulent flow would promote mixing - inline mixers, for example.

Augusto5,

Sorry, I can't help you much as I am not on operations side - I'm on the pipeline design side.

The best I know, flow rate and pressure are not variables that my client plays around with. The pressure is pretty much set based on the total pressure loss to the terminal at the given flow rate. And the flow rates are set by the schedulers per the contracted deliveries - usually as much as the system can handle.

 
Pipeflo, a multi phase steady state pipeline hydraulics simulator from Neotec, includes a section that will calculate the size of the interface between different products- it was used on a multi-product pipeline I helped to design & build about 10 years ago. That might help with your problem.
 

To Ashereng, it has been established that when speaking of transferring white or "clean" oils in a multi-product pipeline in a sequence such as naphtha, gasoline, kerosine, gas oil, kerosine, gasoline, naphtha, laminar flows result in complete mixing of the products (probably because of inter-diffusion and longer residence times).

The Reynolds number I quoted are critical in the sense that at these values the rate of intermingling shows a sharp decrease.

Upon arrival at destination switching to appropriate storage is done generally on basis of specific gravity of the products flowing past the detector of the interface. Other properties such as dielectric constant or viscosity have also been used.

Sometimes water-inflated rubber spheres are introduced into the pipeline by special launchers as the switch of one product to another is made at the origin of the pipeline.

At the receiving point special storage tanks are supposed to hold the interfacial mixtures until they are disposed of, usually by blending into a larger volume of compatible products in a ratio that will not put that product off specification.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor