Directional wattmeter with plug-in couplers

Introduction

A directional wattmeter is used to measure forward and reflected power in a transmission line (PF & PR, respectively). The power that finally reaches the load (e.g. aerial), PL, is simply the difference of the forward and reverse power:

 PL = PF - PR

 A property of the directional wattmeter that is commonly overlooked is that the PF and PR calibration on the meter face is only valid for a matched transmission line. If the line is mismatched (e.g. high VSWR), the PF and PR readings are meaningless. Interestingly, PL is still the simple difference, (PF - PR) even in a mismatched line. So, the amount of power that reaches the load in a mismatched system can still be measured.

 For those who are more conversant with SWR, it can be determined either by calculation or by referring to a chart that can be found in various ARRL handbooks under the title of ‘SWR as a function of forward & reflected power’. [1]

Construction of the main unit

Front of main unit with meter (left), and switches for selecting 1 of 3 couplers (top right) and Fwd/Rev power (bottom right)

Main unit's rear view showing 3 headphone jacks for connecting to couplers

Circuit of main unit

A very useful feature of the detachable coupler is that it can be inserted along the transmission line at close proximity to the aerial’s base.  It can then be linked to the main unit via an interconnecting cable. This is useful technique for measuring SWR at VHF/UHF because a long and lossy coax run can give deceptively favourable results on a SWR meter in the shack, even when the aerial is off-tuned. For a more thorough explanation of this common error in SWR measurement, the reader can refer to the ARRL Handbook.[2]

Coupler attached to rear of main unit

Couplers construction

 A coupler for 5W maximum power and 1 ~ 200 MHz operation is described. Experimenters can use the example as a starting point to build couplers for other frequency ranges or power levels.

 The coupler’s RF-tight enclosure were constructed from rectangular PCB scraps. The edges were soldered together to form a box. The components inside are accessible via a lid secured by screws. The coupler connects to Tx. and Ant. via two BNC connectors on opposing sides. A stereo jack at one end served as the interface to the main unit.

1-200MHz 5W plug-in coupler

Inside the coupler, forward & reflected powers are sampled with a directional coupler. A pair of Schottky diodes (BAT-41) is used to rectify the sampled RF power. The RF portion is shielded from the DC section with a tin-plate shim.  Feed-thru capacitors penetrated the shield to permit DC connections between sections. Small 5kΩ trimmers allow different couplers to be adjusted to the same full-scale deflection on the meter. Small holes on the side of the couplers allow the trimmers to be adjusted with the lid closed.

1-200MHz 5W coupler circuit

View of opened coupler

To prevent RF from leaking around the top edge of the shield, spring fingers were soldered to the coupler’s lid. When the lid is shut, the spring fingers make contact with the top edge of the shield.

Coupler's lid with spring fingers

Almost if not all of the directional wattmeter / SWR meter published in amateur journals are based on the half century’s old Bruene W5OLY bridge. Bruene used a conventional transformer, T1, for the directional coupler. Details of the Bruene bridge can be found in a review of directional wattmeters by White G3SEK.[3]

Unfortunately, the parasitic of the conventional transformer restrict its frequency range to slightly more than a decade. DeMaw W1FB reported a range of 1.5 ~ 30 MHz[4]. In addition, capacitive trimmers are required to null out the bridge.

Bruene W5OLY bridge - the basis of many ham directional wattmeter / SWR meter

Implementing the directional coupler with a Guanella transmission line transformer obviates the requirement for nulling capacitors and hopefully allows a wider frequency range. The Guanella transformer consists of a ‘main’ line (Tx. & Ant.) and a ‘branch’ line (Fwd. & Rev.). The coupling factor, k, between the ‘main’ and ‘branch’ lines is determined by the turn ratio, n:

The ratio of n is varied in accordance to the maximum Tx. power. This allows the same range of power to be coupled to the diode detector via the ‘branch’ line irrespective of actual power on the ‘main’ line. Lower n results in tighter coupling (lower k) and is useful for QRP. Conversely, higher n ratio produces looser coupling for QRO. 

Guanella transmission line transformer winding details

A twin-hole balun core salvaged from an old VHF TV aerial formed the basis of the transformer. I have no knowledge about the RF properties of the twin-hole core and used it on the basis of blind faith. Twin-hole balun cores can be found in TV aerial with folded dipoles for the driven element.

 Each hole carried a single turn and a seven turn windings. Enameled wires of different diameters were used for the windings. SWG 20 was used for the single turn windings and SWG 32 for the seven turn windings.

Windings on a twin hole binocular core

A directional coupler has two critical parameters – the insertion loss between the Tx and Ant ports, ILTx-Ant, and directivity, D. For continuous in-line monitoring use, low values of ILTx-Ant will prevent the reduction of S/N ratio during reception of weak signals.

Afterthoughts

 The non-linear calibration of the meter face was time consuming and resulted in an ugly compressed scale at the higher end. Non-linear scaling could have been avoided by using a DC amplifier ahead of the 50-uA meter. By increasing the meter’s sensitivity, full-scale deflection can be achieved over the diode detector’s square law range (-40 ~ -20 dBm). To reduce the power input to the diode detector to fit into this square law range, the coupling factor, k, will have to be increased beyond 17 dB (looser coupling).

 The plastic housing was too light. Heavy coax tend to tilt the housing. A heavier metal case would have been a better choice.

 The large stereo headphone plugs made the coupler protrude too far out. Unfortunately, small stereo headphone plugs make erratic connections. A twin-axial BNC would have been a nice substitute if it were less pricey.

 Acknowlegement

 I will like to thank Phoa 9M2KT for his construction tips and words of encouragement and Sebastian 9W2?? for his mathematical analysis of the coupler’s power vs. detected voltage relation.

[1] ARRL Antenna Book, 16th ed. Graph of SWR as a function of forward & reflected power is given in fig.13 of chap. 24-8.

 [2] ARRL Handbook, 1992 edition. SWR measurement errors explained in fig. 25 of chap. 16-15.

 [3] White I., ‘Inside a directional wattmeter’, Radcom, Sept. 2002. White referred to Bruene’s original paper, ‘An Inside Picture of Directional Wattmeters’, QST, April 1959.

 [4] DeMaw M.F. (W1FB), Practical RF communication data, Howard W. Sam, 1978.

(c) This work is protected by copyright and may not be reproduced in any form without the expressed consent of the owner.