In this final post in the series of Flow Science’s 35th
anniversary simulation contest, I will talk about a case study simulating a
part of a microfluidic circuit – pneumatic latching valve. These devices are a relatively
new industry application that Flow Science is exploring, in the context of a
broader exploration of the use of CFD in microfluidics applications, and
results have been very encouraging.
Figure 1. A microfluidic 4-bit demultiplexer for routing
pressures and vacuum pulses. Inside the red box is a single latching valve that
will be simulated in FLOW-3D
A latching valve, as the name indicates, holds (latches) a
valve in open/closed position without continuous application of external
pressure. Latching valves are used for energy efficiency and are analogous to
electrical solenoid valves. Details of the working of a latching valve system
are shown in Figure 2. Stages 1-7 show how the system changes from a closed
state to a latched open state, and then back to a closed state again. An open
state is one where the fluid can flow through the valve, and in a closed state
fluid cannot flow through the valve.
Figure 2. The 7 stages of latching valve system as it
evolves from closed to latched open to closed again. NC means not connected*
Latching valve setup in FLOW-3D
The latching system primarily comprises of 3 types of
features – valves, inlet channels and control channels. Valves and inlet
channels are made from solid components in FLOW-3D
, while the control channels
are directly represented through meshes (seen in black in the figure below). Each
valve has an inlet channel and a control channel except for valve 3. Valve 3
has an inlet channel and two output channels. The inlet channel brings in fluid and the control
channel allows the pressure to be manually controlled externally by the
user/designer. Setup of the entire latching valve system in FLOW-3D
is shown in
Figure 3. Setup of the latching valve system (currently in
stage-7) in FLOW-3D
Time dependent pressure boundary conditions
Being pneumatic valves, the functioning of the latching
system is totally dependent on the application of pressures at the boundaries
of the system. The inlet boundary condition is a time-varying pressure boundary
condition with vacuum (below atmospheric pressure) and pressure pulses (Figure 4).
The control channel for valve 1 has a pressure pulse twice the atmospheric
pressure (Figure 5). The control channel for valve 2 is maintained at
atmospheric pressure. The outlet channel is at atmospheric pressure. Notice
that eventually all the pressures fall back to atmospheric pressure, which means
that no additional external pressure is required by the latching system to stay
in its state (closed in this case).
Figure 4. Time-varying pressur
e boundary condition for the inlet
channel to Valve 1.
5. Time-varying pressure boundary condition for the control channel of Valve 1.
Stages 3-7 were simulated using FLOW-3D
and the results post-processed in FlowSight
. The latching mechanism has been accurately
simulated, as shown in the reference paper, by starting at the open stage
(stage 3) and ending at the closed (stage 7) stage. Pressure pulses in the inlet
channel are 500 Pa, positive or negative and the pulses span over 50
milliseconds. Water is used as the fluid, and compressibility of water is used
to allow some propagation time for the pressures in the system. Opening and
closing of the individual valves can be seen in top three viewports of the
animation below. The simulation below shows the evolution of the system from
Simulation of a pneumatic latching valve used in
microfluidic demultiplexer. The animation starts at stage 3 – the open
stage, and finally evolves to stage 7 – the closed stage.
An accurate simulation of the working of the pneumatic latching
valve can help designers reduce the cost of trials and errors in the design
phase, ensuring that the best design goes to the fabrication stage. Notice that
in the final stage, the valves are in a closed state and would remain so for a
certain period of time, in spite of the absence of external pressures through control channels.
In the upcoming post I will talk about our new optimization and parametric study capabilities using CAESES
, an optimization software by Friendship Systems.
*Reference: William H. Grover, Robin H. C. Ivester, Eric C.
Jensen, Richard A. Mathies, Development and multiplexed control of latching
pneumatic valves using microfluidic logical structures, 2006