Homemade Inline RF Sampler (Dec 2006)

A RF sampler provides a low-power image of the RF signal propagating in a coaxial line. It is a 3 port device inserted between a power source (transmitter or amplifier) and the terminating load while its sample output (third port) is connected to a measuring equipment ( power meter, spectrum analyzer, oscilloscope, etc.).

Some characteristics of a typical sampler are summarized below :
- It should have a known attenuation between its main ports and the sample port.
- Its attenuation plot should be as flat as possible over the usable frequency range.
- It should handle the power of the tested transmitter.
- Its Standing Wave Ratio (SWR) should be low over its frequency range.

There are several methods to design such a sampler but one of the simplest is a 50 ohms line associated with a resistive divider. Some years ago, W. Hayward, W7ZOI described such a device in QST (1) : he used a brass stripline inside a Hammond 1590A die cast box.

The reason for my design is that I wanted something easier to duplicate, so I used a 50 ohms microstrip line etched on one side of double-side Epoxy PCB instead of the Hayward's stripline.

1. The sampler.

The schematic appears on the left  below and an inside view on the right.


The sampler is built on ordinary double-side Epoxy PCB so the 50 ohms microstrip is 2.8 mm wide : click here to download the PCB artwork (12 k PDF file)

R1, R2 and R3 values are computed in accordance with the required coupling coefficient  : I used  R1=R2=R3= 820 ohms to get a -40 dB coupling.

This resistor being rated  at 0.25 W each, I can use my RF sampler with transmitters up to 40 W. If you want  a higher power limit (100 W) build this circuit with 3 x  820 ohms 0.5 W each.

The sampler is housed in a "tinplate Schubert box" (37 x 55 x 30 mm) available in many electronics shops in Europe, but other brands and models could probably be used.

2. Measurements.

I have built several RF samplers of this type and measured some of their typical characteristics.

The attenuation plot is shown below :

And the SWR of the main port is :

As suggested by W. Hayward,  I put a gimmick capacitor across R1 : this capacitor, made of a small piece of insulated wire extending along R1 and R2 bodies, was adjusted while watching the SWR on a Network Analyzer. It is the red wire on the rightmost photograph.

These plots show that this RF sampler can be used up to 500 MHz with an error less than 0.5 dB and a SWR <1.05.


Although simple and inexpensive, this little RF sampler is a very useful accessory to my measuring equipment.

1  Wes Hayward, W7ZOI, "Simple RF-Power Measurement," QST June 2001, pp. 38-43.