Lumped Ports and Elements

This page describes how to set up lumped ports and elements in Flexcompute RF (also Flex RF).

The lumped port or element represents a discrete circuit connection between a pair of metallic structures in the simulation domain. The lumped port is used to excite and terminate the structure with a given impedance. The lumped element is used to represent a passive circuit load connected to the structure.

Lumped port or wave port?

The lumped representation is fundamentally an approximation of physical reality. It is typically most accurate when the port/element size is much smaller than the wavelength ($L \ll \lambda/10$).

Wave ports excite the actual transmission line mode supported by the port cross-section, so they remain accurate over the band where that mode is well-defined.

Use lumped ports in the following scenarios:

• The port size is much smaller than the wavelength.
• The port structures are not typical transmission lines.
• A quicker but more approximate simulation is desired.

Use wave ports in the following scenarios:

• The port structures are good transmission lines.
• Perfect impedance match for the outgoing wave is desired.
• Calculating the transmission line mode properties in the 2D port plane.

Note: mixing-and-matching lumped and wave ports is generally discouraged unless the different reference impedances and de-embedding standards are carefully handled.

Lumped Ports

The LumpedPort represents a planar, uniform current excitation with a fixed impedance termination.

# Define a lumped port
my_port_1 = LumpedPort(
    name="My Port 1",
    center=(0,0,0),
    size=(0, 200, 300),
    voltage_axis=2,     # voltage drop direction (2 = z)
    impedance=50,       # port impedance
)

The default impedance is 50. The LumpedPort has the following additional restrictions:

• The LumpedPort must be planar (exactly one zero-size dimension). If a 1D port is desired, instead provide a small but finite width along its lateral axis.
• Only axis-aligned planes are supported at this time.

An example lumped port orientation is shown above. The port is shown in blue, and the connected port structures (metal terminals) in orange. The voltage_axis, in red, should always point in the direction that connects the two terminals.

For coaxial terminals, use the CoaxialLumpedPort.

# Define coaxial lumped port
my_coaxial_port_1 = CoaxialLumpedPort(
    name="My Coaxial Port 1",
    center=(0,0,0),
    inner_diameter=1000,   # inner diameter in um
    outer_diameter=2000,   # outer diameter in um
    normal_axis=0,         # normal axis (0 = x)
    direction="+",         # direction of signal along normal axis
    impedance=50,          # port impedance
)

The CoaxialLumpedPort uses the analytical coaxial mode as the source. This means that the port structures must be circular and must match the physical port dimensions. Any discrepancy will result in parasitic effects in the S-parameters.

Tip

When working with imported coaxial geometries (such as the SMA connector), the sides of the cylinders can be slightly faceted due to face tessellation. This results in slight parasitic effects when using the CoaxialLumpedPort. For improved accuracy, consider using the WavePort or TerminalWavePort instead.

Lumped Elements

For simple resistive elements, use the LumpedResistor or CoaxialLumpedResistor.

# Define a lumped resistor
my_resistor = LumpedResistor(
    name="My resistor",
    center=(0,0,0),
    size=(0, 400, 300),
    voltage_axis=2,       # voltage drop direction (2 = z)
    resistance=100.0,     # element resistance
)

# Define a coaxial lumped resistor
my_coaxial_resistor = CoaxialLumpedResistor(
    name="My Coaxial Port 1",
    center=(0,0,0),
    inner_diameter=1000,   # inner diameter in um
    outer_diameter=2000,   # outer diameter in um
    normal_axis=0,         # normal axis (0 = x)
    resistance=25.0,       # element resistance
)

For general-purpose lumped elements, use the LinearLumpedElement. This object connects the element terminals to a passive circuit with arbitrary topology. Use the CircuitImpedanceModel and LumpedCircuitComponent to specify the circuit topology.

# Build the following circuit using CircuitImpedanceModel
#       (0)         (1)         (2)
#  port+ o---( R )---+---( L )---+---o port-
#                    |           |
#                    +---( C )---+
resistor = LumpedCircuitComponent(
    element_type = 'R',
    node_plus = 0,
    node_minus = 1,
    value = 25.0      # Ohms
)
capacitor = LumpedCircuitComponent(
    element_type = 'C',
    node_plus = 1,
    node_minus = 2,
    value = 1e-12     # 1 pF
)
inductor = LumpedCircuitComponent(
    element_type = 'L',
    node_plus = 1,
    node_minus = 2,
    value = 1e-9     # 1 nH
)
my_circuit = CircuitImpedanceModel(
    components = [resistor, capacitor, inductor],
    port_node_plus = 0,
    port_node_minus = 2,
)

# Apply circuit to lumped element
my_lumped_element = LinearLumpedElement(
    center=(0,0, 250),
    size=(0, 200, 500),
    name='my lumped element',
    voltage_axis=2,
    network = my_circuit     # Connect circuit here
)