An Introduction to Power Dividers, Combiners, and Couplers

A power divider divides an incoming signal into two (or more) output signals. In the ideal case, a power divider can be considered loss-less, but in practice there is always some power dissipation. Because it is a reciprocal network, a power combiner can also be used as a power combiner, where two (or more) ports are used to combine input signals into a single output. Theoretically, a power divider and a power combiner can be the exact same component, but in practice there may be different requirements for combiners and dividers, such as power handling, phase matching, port match and isolation.

Power dividers and combiners are often referred to as splitters. While this is technically correct, engineers typically reserve the word “splitter” to mean an inexpensive resistive structure that splits power over very wide bandwidth, but has considerable loss and limited power handling.

The term “divider” is most often used when the incoming signal will be split evenly across all outputs. For example, if there are two output ports, each would get slightly less than half of the input signal, ideally -3 dB compared to the input signal. If there are four output ports, each port would get about one-quarter of the signal, or -6 dB compared to the input signal.

When 2:1 dividers or combiners are cascaded together, the original signal can be split over many output ports. For example, in some high power solid state amplifiers that replace vacuum tubes, as many as 12 amplifiers might be combined.  If you fed one port on a 12-way divider you would see over 20 dB loss to the common port. But if you fed all 12 ports with equal amplitude signals as a combiner, the real loss might just be a few dB. There are also components (like the Telewave ANTPD3xx and ANTPD4xx power dividers) that are 1:3 or 1:4 dividers. That is, they divide power from a single source over 3 or 4 outputs using just one component. In the case of a three-way divider, there are direct ways of achieving multiple ports without cascading 2:1 structures.


When choosing which type of divider or combiner to use, it’s important to consider isolation. High isolation means that incident signals (in a combiner) don’t interfere with each other, and any energy that isn’t sent to the output is dissipated rather than sent to an output port.  Different types of dividers handle this in different ways.  For example, in a Wilkinson divider, the resistor has 2Z0 value and is strapped across the outputs.  In a quadrature coupler, a fourth port has a termination.  The termination dissipates no energy unless something bad happens, like one amp fails or the amplifiers have different phases.

Types of Dividers

There are many types and subtypes of power dividers or combiners. A few of the more common ones include:

Wilkinson power divider

A Wilkinson divider splits an input signal into two equal phase output signals, or combines two equal-phase signals into one in the opposite direction. A Wilkinson divider relies on quarter-wave transformers to match the split port. A resistor is placed across the outputs, where it does no harm to the input signal at Port 1. This greatly improves isolation and allows all ports to be impedance matched. This type of divider is often used in multi-channel radio frequency systems because it can provide a high degree of isolation between the output ports. By cascading more quarter wave sections, Wilkinson’s can easily handle 9:1 bandwidths of electronic warfare systems.

Reactive power divider

If you deleted the isolation resistor in a Wilkinson power divider, you would have a reactive divider. One outcome of deleting the isolation resistor is that only the common port is impedance matched. Reactive combiners are not recommended for power combining, as they have limited isolation. Reactive combiners are often used in antenna feed networks.


The term “coupler” usually implies that power division is uneven, such as 5 percent of the signal being split off (or coupled) to one output port and the remaining 95 percent of the signal going through to the other port. A “10 dB coupler” couples off 10% of the incident power.

The simplest coupler is a four-port network where one port is the common port, one part is the coupled port, one is the “through” port and one port is the isolated port. The isolated port is typically terminated with a matched load and this is what allows the through power and coupled port to be isolated from each other.

A hybrid coupler usually refers to a network that gets equal power division or combination from a network that is asymmetric. 3dB hybrids are two-way combiners that have equal division (3.01 dB is 10*log(0.5), the 0.5 is half-power). A Lange coupler is one form of 3dB hybrid.

A quadrature coupler has a natural tendency to force 90 degree of phase difference between the divided ports, which is useful for reducing reflection coefficient of combined amplifiers. One excellent use of quadrature couplers is to impedance match pairs of devices. The devices are arranged so that reflections from them are terminated in a load that is isolated from the quadrature coupler’s input. When transistors are combined using quadrature couplers, this is called a balanced amplifier. Quadrature couplers are also used to make reflective attenuator devices (such as shunt PIN diodes) become absorptive. Most coupled-line structures provide quadrature response.

Coupled line couplers versus wired couplers

Coupled line couplers use interaction from transmission lines that are near each other but are not connected. In this case of a coupled-line coupler, at DC (0 Hz), none of the power is transferred to the coupled arm. A Lange coupler is a planar microstrip design that uses multiple strips to achieve 3dB (hybrid) coupling; it is an example of a coupled-line coupler. Coupled line couplers easily provide octave bandwidth.

Wired couplers have direct connections between all ports. There are several types of wired couplers.

Branchline coupler

A branchline coupler is a simple type of quadrature coupler arranged in a ring of circumference of one wavelength, with four ports. A signal entering port 1 is split into two quadrature signals (ports 2 and 3), with the remaining port 4 fully isolated from the input port at the center frequency. The lower output port (port 3) has the most negative transmission phase since it has the farthest path to travel. The branchline coupler is narrow-band compared to coupled-line structures; typically, only 20% bandwidth is possible.

Rat-race coupler

A rat-race coupler gets its name from its circular shape, shown below. The circumference is 1.5 wavelengths. For an equal-split rat-race coupler, the impedance of the entire ring is fixed at 1.41xZ0, or 70.7 ohms for a 50 ohm system. For an input signal Vin, the outputs at ports 2 and 4 are equal in magnitude, but 180 degrees out of phase. Typically 30% bandwidth is achieved.

Gysel coupler

A Gysel coupler (often referred to as a the Gysel divider) provides in-phase outputs, and is configured most commonly as a five-port (for a two-way split with two terminations). The Gysel divider is often used in kilowatt-level power combining. For example, if you want some redundancy in a 50,000 watt television transmitter you could use a five-way Gysel combiner with 15,000 watt tubes and be able to remove one of the tubes for service or replacement without taking the transmitter off-line. The terminations will need to be oil-cooled to dump all of the wasted power, but that is doable. Gysel dividers are finding their way into the millimeter-wave spectrum, when gallium nitride amplifiers are combined in solid-state power amps (SSPAs) that will compete with vacuum tubes such as TWTs. Gysels typically provide 30% bandwidth. The Gysel is very similar to the rat-race (the ring is 1.5 wavelengths with SQRT(2)*Z0 impedance) but with the input on a different port and a second isolation load to make it symmetric.

**All graphics are reprinted with permission from