This directional coupler is a derivative of the W7EL Bruene coupler variant using three transformers instead of the conventional Bruene Directional Coupler that uses a capacitive voltage divider. A 1999 QST article by Roy Lewallen, W7EL describes the basic design. Some additional data can be found on K6JCA’s web site here.
The coupler consists of 3 1:10 transformers. One samples the RF current while the other two provide forward and reflected voltage samples. This was designed as the sampling element for a larger instrument which will use two AD8362 Log True RMS detectors to provide up to 6 decades of dynamic range of forward and reflected power for any modulation method. The goal was to provide flat coupling and directivity of greater than 20dB over the range of 500kHz to 110MHz which potentially includes the Amateur bands from 600m to 2m.
This directional coupler is being offered as a drop-in component for home brew or kit radios and has jumper selectable built-in diode detector sampling or direct AC sampling for external detectors. The board also provides headers and corner mounting holes to mount on a carrier board with more sophisticated detectors. Standard inputs and outputs are SMA connectors. It is also available in an extruded aluminum enclosure. It has been tested at power levels of up to 40W at 100MHz and is undergoinig continued testing. Test results will be posted as they become available.
The original W7EL article spends most of its time on the vagaries and non-linearity of diode detectors and claims better than +/- 7% accuracy up to 432MHz but doesn’t ever present any data on the directivity of the coupler over that frequency range. The directivity is a measure of the influence of leakage of the forward power sample on the reflected power measurement. With sufficient leakage – say -10dB – then even with a perfect load match the reflected power will still indicate -10dB of reflected power and never indicate less than a VSWR or better than 1.92:1. A directivity of 20dB means you will still see a VSWR of 1.22:1 with a perfect match. The goal should be a directivity of 25dB or better so reflected power is at least accurate down to around 1.1:1.
Several different ferrite materials were tried and the best was Fair-Rite type 61 material. Since the goal was modest power handling capabilities of less than 100W some small cores were chosen that were readily available in inventory. The resulting directional coupler coupling and directivity are shown below. with a 10:1 transformer the coupling is 20dB.
bare board, Populated Board, Enclosed Board