How To Determine Governing Terminal Velocity For Vertical 3-Phase Separator Sizing

Identifying the limiting velocity is straightforward in principle: you compute the terminal velocity for each of the three phase-separation tasks and select the slowest one. That slowest velocity dictates the vessel diameter, because it demands the largest cross‑sectional area to prevent carry-over or carry-under.

In a vertical three‑phase separator, three terminal velocities compete for control: (1) the falling speed of liquid droplets in the gas, (2) the rising speed of gas bubbles in the oil, and (3) the falling speed of water droplets in the oil. The smallest of these three velocities becomes the governing velocity, forcing the column diameter to be sized for the most difficult separation.

Understanding the Three Critical Velocities

Every vertical three‑phase separator must simultaneously handle three distinct phase interactions. Each interaction is governed by its own terminal velocity, and the slowest one sets the minimum vessel cross‑sectional area.

Liquid Droplet Settling in the Gas Phase

The first interaction occurs in the top section of the column. Tiny liquid droplets entrained in the gas stream must fall fast enough to reach the liquid‑liquid interface before the gas exits. This terminal velocity is the liquid‑droplet fall velocity in the gas phase (v₁).

Gas Bubble Rise Through the Oil Phase

In the liquid‑collection section, dissolved or entrained gas bubbles must escape upward from the oil. The gas‑bubble rise velocity in the oil phase (v₂) determines how quickly the oil can degas without losing gas into downstream oil lines.

Water Droplet Settling in the Oil Phase

Free water dispersed in the continuous oil phase must settle downward to the water‑oil interface. The water‑droplet fall velocity in the oil phase (v₃) defines how effectively water separates from the oil within the available residence time.

Calculating Each Terminal Velocity

Operators in a university‑pilot‑plant setting usually begin by calculating each terminal velocity using a suitable settling law. The calculation method depends on the flow regime, which varies with droplet or bubble size and fluid properties.

Start with Stokes’ Law for Laminar Conditions

If the droplet Reynolds number is low (< ~0.1), falling or rising objects follow Stokes’ law. The velocity equation is:

[ u_t = \frac{g , d^2 , (\rho_p - \rho_f)}{18 , \mu} ]

where (d) is the particle or bubble diameter, (\rho_p) and (\rho_f) are the densities of the dispersed and continuous phases, and (\mu) is the continuous‑phase viscosity. This gives a first estimate, but it must be validated.

Confirm the Flow Regime with the Friction Group Method

To avoid guessing the regime, many unit‑operations labs use the dimensionless parameter (K):

[ K = d \left[ \frac{\rho_f (\rho_p-\rho_f) g}{\mu^2} \right]^{1/3} ]

(K < 3.3) → laminar (Stokes) regime
(3.3 < K < 43.6) → transition (Allen) regime
(K > 43.6) → turbulent (Newton) regime

Once the regime is known, the corresponding terminal‑velocity equation can be applied.

Typical Design Assumptions for a Pilot‑Scale Column

For the gas‑phase separation, a liquid droplet size of 150 µm is standard in educational pilot plants. In more stringent scenarios (e.g., protecting compressor blades), the design switches to 100 µm droplets and includes a mist eliminator. The drag factor (C) is then obtained from the relation (C_D(\text{Re})^2), and the terminal velocity follows:

[ V_t = \sqrt{\frac{128.8 , D_p , (\rho_l - \rho_g)}{3 , \rho_g , C}} ]

For bubble rise and water‑droplet fall in the oil phase, similar correlations based on the appropriate droplet or bubble size—often 150–500 µm—are used, depending on the emulsion characteristics and oil viscosity.

Identifying the Controlling Velocity

After computing (v_1), (v_2), and (v_3), the operator simply finds the smallest value. That smallest velocity is the governing terminal velocity.

Why the Smallest Velocity Controls the Diameter

Separator diameter is derived from the continuous‑phase flow rate and the terminal velocity of the dispersed phase. For a given volumetric flow (Q), the required cross‑sectional area (A) is (A = Q / u_t).

Because area is inversely proportional to velocity, a lower terminal velocity demands a larger area. Therefore, the phase pair with the slowest separation dictates the minimum column diameter. If a smaller‑diameter column were used, that phase would not separate within the available residence time, leading to entrainment and poor product quality.

A Practical Example

Suppose the calculated velocities are:

Liquid in gas: 0.2 ft/s
Gas in oil: 0.03 ft/s (slowest)
Water in oil: 0.05 ft/s

The 0.03 ft/s value controls. All other separations would be faster within the same vessel, but the column must be wide enough to handle the slowest gas‑bubble rise. This reasoning is fundamental to the pilot‑scale design approach taught in unit‑operations laboratories.

Understanding the Trade-offs

Selecting a velocity solely from theoretical settling equations carries some limitations. Pilot‑plant operators must recognize these trade‑offs to avoid scale‑up surprises.

Droplet Size Distribution is Never Uniform

Real dispersions contain a range of droplet and bubble sizes. The chosen design diameter is often a mean or a conservative small size to prevent entrainment. For gas‑liquid separation, 150 µm is a classroom standard, but it may not reflect the actual droplet population. A finer mist would settle even more slowly, demanding an even larger diameter.

Residence Time Versus Turndown Flexibility

Sizing for the slowest velocity yields a safe design at full flow, but the column may be oversized at low turndown. An oversized diameter reduces liquid turbulence, which can improve separation, but it also reduces effective liquid height for level control. Typically, pilot‑plant setups maintain a liquid residence time of around 10 minutes to balance controllability and separation performance.

The Role of Internals

Mist eliminators, demister pads, and inlet diverters can enhance separation and effectively raise the allowable terminal velocity for design. When a mist eliminator is present, the gas‑phase velocity can be increased within the vessel, sometimes shifting the controlling velocity to the liquid‑liquid separation stages.

Making the Right Choice for Your Goal

The method for finding the governing terminal velocity depends on what you are optimizing.

If your primary focus is educational clarity: Calculate all three velocities with simplified Stokes‑law assumptions and identify the smallest. This illustrates the core principle without complex rheology corrections.
If your primary focus is precise pilot‑plant design: Use the friction group method to confirm the flow regime for each phase pair, then compute velocities with the appropriate drag correlation. Select the smallest velocity and size the column diameter accordingly.
If your primary focus is flexible operation: Consider the terminal velocity of the smallest expected dispersed‑phase size (e.g., 100 µm droplets) as the controlling criterion. This provides a conservative design that works across a wider range of feeds and flow rates.

The governing terminal velocity is always the slowest of the three phase‑pair velocities; your role is simply to calculate them honestly and let that smallest number guide the geometry of your separator.

Summary Table:

Velocity Type	Phase Interaction	Key Sizing Impact	Typical Pilot-Scale Assumption
Liquid Droplet Fall (v1)	Liquid droplets in gas phase	Dictates gas-phase separation efficiency	100–150 µm droplet size (with mist eliminator)
Gas Bubble Rise (v2)	Gas bubbles in oil phase	Prevents gas carry-under in downstream lines	Highly dependent on continuous phase viscosity
Water Droplet Fall (v3)	Water droplets in oil phase	Dictates water-oil separation effectiveness	150–500 µm droplet size (emulsion dependent)

Bring hands-on chemical engineering principles to life in your laboratory. LABPARK designs and manufactures high-performance Educational and Vocational Unit Operations Pilot Plants spanning chemical engineering, bioprocess & biotech, and environmental & water treatment. Tailored for universities, research institutes, and enterprises, our systems offer students and researchers real-world experience with fluid dynamics, separation processes, and pilot-scale operations. Ready to enhance your lab's educational impact? Contact us today to explore our customizable pilot plant solutions!