Carsten Groen's (OZ9AAR) personal place

144 MHz - 2300 MHz Directional coupler

Embarking on my venture to create a 23cm power amplifier project, I found myself in need of a suitable directional coupler. While there are directional couplers available on the market, the challenge lies in finding products that can handle high power while maintaining good directivity without breaking the bank. Scouring platforms like eBay for surplus devices can sometimes yield fortunate discoveries, but I sought a more reliable solution that did not rely on chance encounters on online marketplaces.

Thus, I made the decision to take matters into my own hands and create a directional coupler from (almost) scratch.

But, remember:

Making a directional coupler is easy.

Making a good directional coupler is NOT easy.

Remember to also check my Dual RF Head, Dual RF Head - USB and Dual RF Head - M5 pages, they work excellently together with the directional coupler (as seen in my 23cm PA project).

I am able to deliver (limited numbers of) assembled and tested directional couplers for €245,- each plus shipping. Each comes with measurements data for 144,432,1296 and 2300 MHz, both as individual sweeps and a summary sheet (see below).

Send me an email if interested, contact info on the CV/Contact page.

Fortunately, there are several meticulously documented projects available on the internet that provide valuable insights into this domain. One such exemplary resource is the comprehensive article authored by Paul Wade, W1GHZ (link to Pauls webpage). Paul has invested significant effort into experimenting with and measuring directional couplers, achieving commendable results using off-the-shelf aluminum enclosures and other components.

Upon studying Paul's work and drawing inspiration from his methodology, I successfully constructed multiple prototypes.

I looked at Pauls article and using ideas from his work, I managed to build a few prototypes. Through several iterations, I finally ended up with a directional coupler that met the requirements for my projects.

Given the necessity for a coupler capable of handling high power, I opted for DIN 7/16 connectors over N connectors. Additionally, I prioritized achieving reasonable return loss (VSWR) and ensuring good directivity. While there are pre-fabricated couplers available at competitive prices, their directivity may not always be optimal, significantly impacting the accuracy of measurements taken. The coupler had to span the 70 cm and 23 cm bands at a minimum (this one also cover the 2 meter and 13 cm bands (13cm with a little less return loss and directivity). 

The coupler serves as an ideal complement to both my Dual RF Head and my Dual RF Head - USB devices. Together with these, it offers a reasonable cost-effective solution for measuring both forward and reflected (high) power, either independently or in conjunction with my REPAM module for remote monitoring.

Impedance calculator for striplines (one of many), link to online calculator.

Below an example of the directional coupler used together with my Dual RF Head - M5 device as a stand-alone SWR and power meter. Remember also to check out my Dual RF Head (analog) and the Dual RF Head - USB projects!

Forward to approximately 22 minutes into the video to see the Directional Coupler being used.

Example of summary sheet for delivered couplers. Together with this summary sheet, individual sweeps for all ports, coupling and isolation are also included for each coupler.

The design

As mentioned, I did a few iterations of the design before I ended up with something that worked as expected. I took some of the ideas and advises from Paul W1GHZ work, did some simulations for the stripline and ended up doing three iterations of prototypes before I got something I was satisfied with. Below are some screenshots from the CAD program of the final version.

Image below shows the directional coupler together with my Dual RF Head - USB device. Usually (depending on the frequency as well as the maximum power you are going to push thru the coupler) two external attenuators are needed for the FWD and REF ports when connecting these to the Dual RF Head (USB). The Dual RF Head needs a maximum of +10 dBm input power (I try normally to keep it at max 0 dBm), so you need to do some calculations based on the band you will be using the coupler, the coupling factor of the directional coupler (see below) and the limitation of the Dual RF Head.

A few internal screenshots of the coupler.

Updated front of coupler:

Mechanics

Below is a few pictures of the last prototypes.

Dimension of the coupler:

Width: 64.5 mm

Length: 110.0 mm

Height: 38.4 mm

Spacing between the four M3 screws in the DIN 7/16 connectors is 24.75 mm

Weight: 450 gram

Installation of two Dual RF Head - Analog modules, REPAM device, Overcurrent module, directional coupler, High power 90 degree hybrid, low power 90 degree hybrid, 150W 23cm driver PA and 1500W attenuator in my dual 23cm PA for my 4.8 meter dish project:

Norbert, OE3NFC, installation in his water cooled dual module 144MHz power amplifier using a Dual RF Head - Analog module, REPAM device, Overcurrent module and a directional coupler:

Details of measurements

Below is the measurement results from the final revision of the directional coupler. First there is a few details about the measurements and the setup, which port is which etc. I used my R&S ZNLE3 VNA and ZN-Z150 calibrator (both in calibration) for the testing.

The directional coupler is symmetric (within physical variations). The measurements and tests, assumes that the ports are used in the following way (you can swap P1/P2, this will also swap the meaning of P3/P4):

• P1 is the "input port", this is typically where the power comes from the power amplifier.
• P2 is the "output port", this is typically where the antenna(s) are connected.
• P3 will then couple a small amount of the forward power (power from the PA).
• P4 will couple a small amount of the reflected power (power reflected from the antenna).

A few things to be aware of:

Return loss, is how well the directional coupler are matched to 50 Ohms. This will to some extent depend on frequency. This can be expressed in dB and also as a VSWR value. The return loss needs to checked on all four ports.

Coupling factor, the amount, or level, of power that are coupled to P3 and P4 depends on the "coupling factor" of the directional coupler. The coupling factor is dependent on frequency, lower frequency yields lower coupling, high frequency yields higher coupling (the coupling line seems to be "longer" with high(er) frequencies, hence more power is coupled to the coupling port at high(er) frequency).

The directivity is the amount of power that "leaks" into the other coupling port. For example, you inject power into P1, and a part of the power (based on the coupling factor) comes out on P3. But also a small amount of power will be present on P4. If the directivity was "infinite high", no power at all would be present on P4. In the real world, directivity is never "infinite"! We want the (leaked) amount of power coming out on P4 to be as small as possible. Same goes for power reflected on P2, this will be coupled to P4 (based on coupling factor) but a small part are also coupled on P3.

Insertion loss, is the amount of power "lost" (in the form of heat) from P1 to P2. We want this number to be as close to zero as possible. Using good connectors, wide and thick stripline between P1 and P2, helps keeping the insertion losses down.

Measurement results

Below are preliminary data for the coupler (measured with R&S ZNLE3 VNA, calibrated with ZN-Z150). Actual values for different devices will vary a bit.

Port numbers used are the same as shown above on photos and principal schematic.

The coupler is also very useable on 144 MHz, the return loss is excellent, the directivity is around 23 dB on both ports and the coupling factor is around -60 dB (perfect for my Dual RF Head devices).

Return loss (P1/P2), S11/S22:

2m: -46.2 dB / 44.7 dB

70cm: -46.5 dB / 50.6 dB

23cm: -31.4 dB / 32.3 dB

13cm: -22.6 dB / 22.7 dB

Return loss (P3/P4), S33/S44:

2m: -34.4 dB / -33.8 dB

70cm: -30.0 dB / -30.0 dB

23cm: -25.0 / -25.0 dB

13cm: -33.7 / -39.0 dB

Insertion loss (very hard to measure but approximately 15% more than a Spinner 7/16 female/female coupler):

2m: < -0.02 dB

70cm: < -0.02 dB

23cm: < -0.02 dB

13cm: < -0.03 dB

FWD coupling factor, S31:

2m: -61.0 dB

70cm: -52.6 dB

23cm: -43.2 dB

13cm: -38.5 dB

REF coupling factor, S42:

2m: -60.9 dB

70cm: -52.6 dB

23 cm: -43.2 dB

13cm: -38.6 dB

FWD directivity:

2m: -21.2 dB

70cm: 25.7 dB

23cm: 29.1 dB

13cm: 23.2 dB

REF directivity:

2m: -21.3 dB

70cm: 26.0 dB

23cm: 30.7 dB

13cm: 22.1 dB

FWD coupling (from P1 to P3):

REF coupling (from P2 to P4):

Measurement results

Peter, M0LNB, was so kind to share some ADS plots of the directional coupler he got from me. You can find more interesting stuff at Peter's website: rfconsult.uk

S21:

Coupling:

Isolation: