Tests on the directional coupler with HF-output clearly showed differences in reflection damping between different types of low power 50/51 ohms terminating resistors and dummy loads. Good terminators showed over 20dB return loss.
Such a directional coupler with HF-output can be used to measure the match of filters. Conveniently using only microwatt-power, from a
simple signal generator.
It also showed to be ideal in use for sensitivity measurements on our Rotterdam repeaters, which were working in full duplex mode on a dummy load or a antenna. The attenuation (coupling loss) from repeater-transmitter to the signal generator was -20dB to -30dB, so the signal generator could not be damaged, and could not cause IMD. Real sensitivity measurement of a fully operational repeater, without the use of an expensive measuring set, then is easy.
Directional couplers with HF-output.
BNC receptacles are of screw-type with PTFE dielectricum. Casings used are from from TEKO. N receptacles are of flat flange-type with PTFE dielectricum. The flanges of the recepticles are directly soldered onto the bottom copper (back plane) of the PCB, OR soldered to the side connections of the PCB.
Directional couplers with DC-output.
Read also the instructions above about mass connection.
BNC receptacles are of screw-type. Casings from TEKO.
Reverse the diodes if negative DC output is wanted.
Several years ago, i needed an SWR-meter/power-meter for use on 145MHz. As money was short, a cheap one ("Made in Italy") for 27MHz was purchased, as specifications stated is was "designed" for up to 150MHz.
Well, it was rubbish of cause. In the back, soldered on two SO239 receptacles, was a SINGLE sided PCB with some sharply zigzagging tracks and loosely connected wires, using the back plane of the housing at 5mm distance as a "ground plane". This could never result in a proper impedance of the measuring lines, even on 27MHz. The SWR (reflection damping) of this PCB itself was bad on 2mtrs, and power indication was a laugh.
In an old ARRL handbook was a picture of a PCB for use in a SWR-meter. Antenna books described how to make measuring lines from coax or using Toro ides. I tried some of these, but none of them was working properly on VHF, and descriptions did not say what the influence was of the coupling-wires, or the PCB-material on the properties of the whole thing. But i did found details about impedance of PCB-tracks, and decided to try to design my own directional coupler PCB.
Important for construction:
The width of the tracks and the dielectric constant of the PCB material determine the characteristic impedance of the tracks. Therefore, only use PCB with the specified dielectric constant. I was surprised to see that good quality hard-paper PCB (Conrad Electronics) had also the correct dielectric constant. But better use glass-fiber PCB as this is stronger.
The standard RF source- and load impedance of VHF/UHF communication equipment is 50 ohms. The power-line between the input and output connectors of a directional coupler therefore MUST have also that impedance of (exactly) 50 ohms.
The impedance of the measuring lines could have any practical value, as long as one side of each measuring line is terminated with the characteristic impedance of that measuring line.
The back plane of the PCB serves as mass plane and screen. Very short mass connections between receptacles and back plane are a MUST.
If the coupler is to be used as a directional coupler with HF-output from the measuring lines into 50 ohms loads, the measuring lines MUST have an characteristic impedance of 50 ohms as well.
It then is very easy to terminate those measuring lines correctly with resistive 50 ohms loads, made from 3x 150 ohms SMD resistors in parallel. If 90 degr. corners in the tracks are changed to 2x 45 degr; the unit can be used on 450MHz and even higher frequencies, as those corners do not reflect RF-waves back to the source anymore. This makes it easy to construct good performing instruments.
The output of a measuring line (sensitivity) depends on the wavelength of the signal, the length of the coupling line, and the spacing between the coupling line and the power line. 3x higher frequency means 3x higher output from the same configuration. If one PCB should be used for two different frequency bands, the sensitivity should be calibrated for each frequency band separately by trim potentiometers, if one scale is to be used on the meters. For higher frequencies (23cms) the coupling lines should be made shorter. For a low power range, a separate scale should be made to compensate for detector diode non-linearity.
The diodes should have low knee voltage, and suitable for high frequencies. Resistors and condensers are SMD types. Resistors should be soldered on top or beside each other, making good thermal contact to the PCB-tracks. Maximal power handling is determined by the maximum allowable power dissipation of the terminating SMD-resistors.
The coupling between the power line and measuring lines can be adjusted by varying the distance between them. This distance is critical, and care should be taken when photographing and etching the PCB. A coupling attenuation of less than 20dB will influence the SWR of the power line.
Pictures of prototypes:
|Directional coupler with HF-output, -30,5dB @ 145MHz, -20dB @
This unit is very useful for weak coupling of HF-sources, and return loss measurements with very low, to high power.
The housing is a small TEKO box.
|Directional coupler with detectors and DC-output.
This unit can be used as power meter and SWR meter for 28MHz to 435MHz.
|Directional coupler PCB with detectors and DC-output.
Replacing the very poor detector circuit in a cheap citizens band "power/SWR meter".
It now can be used to measure on 28MHz, 145MHz and 435MHz.
As shown here, a simple but accurate PCB works, with receptacles soldered on the mass side (back), screwed to the housing.
To prevent HF radiation, a better solution is to put the PCB in a dedicated, RF-tight box.